Datasets:
add all 2020 transcripts
Browse files
Jess, Steve, and Bryan reflect on Season 1_transcript.txt
ADDED
|
@@ -0,0 +1,1705 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[0.00 --> 14.10] Welcome to On The Metal.
|
| 2 |
+
[14.50 --> 15.46] I'm Brian Cantrell.
|
| 3 |
+
[15.58 --> 16.96] With me as always is Jess Frizzell.
|
| 4 |
+
[17.08 --> 17.50] Hey, Jess.
|
| 5 |
+
[17.64 --> 18.06] Hey, Brian.
|
| 6 |
+
[18.26 --> 20.94] And joining us is our boss, Steve Duck.
|
| 7 |
+
[20.96 --> 21.36] Hey, Steve.
|
| 8 |
+
[21.52 --> 22.14] Glad to be here.
|
| 9 |
+
[22.52 --> 24.10] And that's it in the garage.
|
| 10 |
+
[24.22 --> 25.58] We actually don't have a guest
|
| 11 |
+
[25.58 --> 28.36] because we are at the end of season one.
|
| 12 |
+
[28.44 --> 29.00] This is it.
|
| 13 |
+
[29.00 --> 30.92] What a season, huh?
|
| 14 |
+
[30.92 --> 31.78] That was amazing.
|
| 15 |
+
[31.86 --> 32.60] It was really good.
|
| 16 |
+
[32.84 --> 34.20] Okay, so I've got to ask both of you.
|
| 17 |
+
[34.66 --> 36.24] When we set out to go do this,
|
| 18 |
+
[36.38 --> 38.72] and I think all three of us wanted to do the podcast,
|
| 19 |
+
[38.82 --> 39.62] we thought it was going to be fun.
|
| 20 |
+
[40.14 --> 42.24] Did you think it was going to be as good as it was?
|
| 21 |
+
[42.28 --> 43.30] It was so good.
|
| 22 |
+
[43.56 --> 43.88] Definitely.
|
| 23 |
+
[44.64 --> 48.14] I definitely knew that a certain subset of people would like it,
|
| 24 |
+
[48.20 --> 50.56] but then I think that it grew more than I thought it would.
|
| 25 |
+
[50.66 --> 52.74] Yeah, I mean, you cannot argue
|
| 26 |
+
[52.74 --> 54.18] that this thing did not beat our expectations.
|
| 27 |
+
[54.18 --> 57.58] No, I think the caliber of guests
|
| 28 |
+
[57.58 --> 60.10] that we were able to get was mind-blowing.
|
| 29 |
+
[60.20 --> 60.82] It's mind-blowing.
|
| 30 |
+
[61.00 --> 63.42] Which was also fun because it was like during raising,
|
| 31 |
+
[63.60 --> 63.98] which is cool.
|
| 32 |
+
[64.08 --> 66.78] And I also thought the guests were so different.
|
| 33 |
+
[67.00 --> 68.82] You know, I mean, I loved that.
|
| 34 |
+
[68.96 --> 70.64] And, you know, you would get a guest like,
|
| 35 |
+
[70.70 --> 71.62] well, this is going to be very different.
|
| 36 |
+
[71.68 --> 73.74] And then it takes a delightful path.
|
| 37 |
+
[73.86 --> 74.70] I just thought it was amazing.
|
| 38 |
+
[75.02 --> 76.54] And yet a bunch of common themes.
|
| 39 |
+
[76.92 --> 77.98] A bunch of common themes.
|
| 40 |
+
[78.18 --> 79.22] A bunch of common themes.
|
| 41 |
+
[79.30 --> 80.84] And especially, you know,
|
| 42 |
+
[81.06 --> 82.54] I could suggest you and I would do the show notes,
|
| 43 |
+
[82.54 --> 84.58] which actually, man, for some of these guests,
|
| 44 |
+
[84.68 --> 86.06] the show notes were a truck pull.
|
| 45 |
+
[86.32 --> 87.14] Yeah, no, there was,
|
| 46 |
+
[87.36 --> 90.42] when I did the first one for Jeff Rothschild,
|
| 47 |
+
[90.52 --> 91.44] like those show notes,
|
| 48 |
+
[91.50 --> 93.24] I was like, it was multiple pages.
|
| 49 |
+
[93.30 --> 95.18] And I was like, whoa, I had no idea.
|
| 50 |
+
[95.26 --> 98.66] But also it lets you like listen to it over again
|
| 51 |
+
[98.66 --> 99.90] with like a fine tooth comb.
|
| 52 |
+
[99.98 --> 101.48] So like I like learned a lot.
|
| 53 |
+
[101.62 --> 102.36] I learned a lot too.
|
| 54 |
+
[102.42 --> 104.24] And I felt also there were a couple of things
|
| 55 |
+
[104.24 --> 105.58] that came up a couple of times.
|
| 56 |
+
[105.68 --> 107.08] Technologies that came up a couple of times.
|
| 57 |
+
[107.08 --> 109.50] And it was just, it was a ton of fun.
|
| 58 |
+
[109.50 --> 111.46] For what it's worth, I did volunteer to try and do show notes.
|
| 59 |
+
[111.66 --> 113.28] And I think both of you were like, you know.
|
| 60 |
+
[113.50 --> 113.94] It's okay.
|
| 61 |
+
[114.36 --> 114.68] It's okay.
|
| 62 |
+
[114.78 --> 116.66] The explanation process is probably going to be longer
|
| 63 |
+
[116.66 --> 117.34] than us doing it.
|
| 64 |
+
[117.72 --> 118.24] It's true.
|
| 65 |
+
[118.50 --> 120.50] But it was, it was, it was a lot of fun.
|
| 66 |
+
[120.56 --> 121.48] The whole thing was a lot of fun.
|
| 67 |
+
[122.00 --> 123.70] And there's going to be another season, right?
|
| 68 |
+
[123.72 --> 125.38] This is not the last season of On the Metal.
|
| 69 |
+
[125.42 --> 126.24] Oh, definitely not.
|
| 70 |
+
[126.38 --> 127.08] So we're excited.
|
| 71 |
+
[127.16 --> 127.82] We're going to keep doing it,
|
| 72 |
+
[127.84 --> 130.70] but we are going to take a break for a little bit.
|
| 73 |
+
[131.06 --> 133.22] And I think the three of us wanted to reflect a little bit
|
| 74 |
+
[133.22 --> 134.88] on some of our favorite moments.
|
| 75 |
+
[135.58 --> 138.08] We loved, I think, every conversation
|
| 76 |
+
[138.08 --> 140.50] and loved so much about so many different conversations.
|
| 77 |
+
[140.66 --> 142.94] But what we, each of the three of us has done,
|
| 78 |
+
[143.26 --> 145.44] is we have written down five moments
|
| 79 |
+
[145.44 --> 148.10] that spoke to us in some special way.
|
| 80 |
+
[148.96 --> 151.50] And we actually don't know what we've said.
|
| 81 |
+
[151.70 --> 153.98] So I know, Jess, you are very concerned
|
| 82 |
+
[153.98 --> 155.34] that we are going to steal your favorite moment.
|
| 83 |
+
[155.40 --> 156.62] Yeah, so I would like to go first.
|
| 84 |
+
[156.68 --> 157.42] You get to go first.
|
| 85 |
+
[157.58 --> 158.76] Okay, because my favorite moment,
|
| 86 |
+
[158.86 --> 159.98] I do not want to be stolen.
|
| 87 |
+
[160.10 --> 162.00] And I feel like it might be on other people's list,
|
| 88 |
+
[162.00 --> 164.08] but it was Amir's Silicon Wafers.
|
| 89 |
+
[164.08 --> 165.38] Oh, that was on my list.
|
| 90 |
+
[165.92 --> 167.52] Oh, man, that was on my list.
|
| 91 |
+
[167.62 --> 167.90] And I felt like-
|
| 92 |
+
[167.90 --> 168.34] It's so good.
|
| 93 |
+
[168.42 --> 169.52] That's why I wanted to go first.
|
| 94 |
+
[169.62 --> 169.98] I was like, no.
|
| 95 |
+
[169.98 --> 170.74] Oh, that was on my list.
|
| 96 |
+
[170.74 --> 171.20] I gotta go.
|
| 97 |
+
[171.58 --> 172.22] Okay, all right.
|
| 98 |
+
[172.52 --> 172.92] That was great.
|
| 99 |
+
[172.92 --> 174.00] You know, that was on my list.
|
| 100 |
+
[174.20 --> 177.26] And that's funny because I felt very confident
|
| 101 |
+
[177.26 --> 179.18] that that would be on no one else's list.
|
| 102 |
+
[179.42 --> 179.98] Excuse me.
|
| 103 |
+
[180.02 --> 182.34] I thought about ordering some old Silicon Wafers
|
| 104 |
+
[182.34 --> 184.50] for the office to hang from the ceiling.
|
| 105 |
+
[184.54 --> 185.36] Oh, my God, you should do that.
|
| 106 |
+
[185.38 --> 185.92] Turn around, Brian.
|
| 107 |
+
[186.00 --> 186.52] Yeah, exactly.
|
| 108 |
+
[187.44 --> 189.14] Okay, so that obviously all three,
|
| 109 |
+
[189.26 --> 190.90] that moment spoke to all three of us.
|
| 110 |
+
[190.90 --> 192.30] Why did that moment speak to you so much, Jess?
|
| 111 |
+
[193.20 --> 194.70] The image of it, honestly.
|
| 112 |
+
[195.34 --> 196.42] And I would love doing that.
|
| 113 |
+
[196.54 --> 198.40] It feels like, and I don't know,
|
| 114 |
+
[198.44 --> 200.10] it was like the way he described it.
|
| 115 |
+
[200.52 --> 201.88] I mean, so for those of you
|
| 116 |
+
[201.88 --> 202.98] who may not have caught this moment,
|
| 117 |
+
[203.06 --> 205.04] this is with Amir Michael describing
|
| 118 |
+
[205.04 --> 208.46] hanging Silicon Wafers in his bedroom
|
| 119 |
+
[208.46 --> 209.92] as a child, like a fifth grader.
|
| 120 |
+
[210.36 --> 212.00] And just like the way he described it,
|
| 121 |
+
[212.06 --> 213.32] I felt like I'm in the bedroom.
|
| 122 |
+
[213.56 --> 213.72] Totally.
|
| 123 |
+
[213.92 --> 214.96] It was so evocative.
|
| 124 |
+
[215.34 --> 216.80] And there was so much.
|
| 125 |
+
[216.84 --> 218.36] It's like I was a kid in Silicon Valley.
|
| 126 |
+
[218.56 --> 219.38] I mean, it was so-
|
| 127 |
+
[219.38 --> 220.16] Why didn't I do that?
|
| 128 |
+
[220.24 --> 220.76] It was so good.
|
| 129 |
+
[221.06 --> 221.26] Right.
|
| 130 |
+
[221.36 --> 221.92] Well, and actually,
|
| 131 |
+
[222.02 --> 223.02] you're the anomaly, Stephen,
|
| 132 |
+
[223.10 --> 223.38] that at least,
|
| 133 |
+
[223.56 --> 224.46] well, you grew up in the Bay Area,
|
| 134 |
+
[224.54 --> 225.34] not really Silicon Valley,
|
| 135 |
+
[225.44 --> 226.20] growing up in the East Bay.
|
| 136 |
+
[226.30 --> 226.82] But we, I mean,
|
| 137 |
+
[226.96 --> 228.48] for most of us who are out here,
|
| 138 |
+
[228.54 --> 229.52] we did not grow up out here.
|
| 139 |
+
[229.74 --> 230.20] Yeah, that's true.
|
| 140 |
+
[230.40 --> 232.16] And Amir is unusual in that regard.
|
| 141 |
+
[232.26 --> 234.64] He's a true child of Silicon Valley.
|
| 142 |
+
[234.72 --> 236.30] I just thought that was a very evocative moment.
|
| 143 |
+
[236.38 --> 237.86] So yeah, Jess, you were right to fear it.
|
| 144 |
+
[237.98 --> 238.16] Yeah.
|
| 145 |
+
[238.40 --> 239.08] No, that's-
|
| 146 |
+
[239.08 --> 240.08] That's why you wanted to go early.
|
| 147 |
+
[240.08 --> 240.86] I thought I was less fearful of, so.
|
| 148 |
+
[240.88 --> 241.48] God, I know.
|
| 149 |
+
[241.58 --> 242.54] I am now terrified.
|
| 150 |
+
[242.54 --> 244.18] I actually, I was convinced
|
| 151 |
+
[244.18 --> 245.40] that none of mine were going to get stolen.
|
| 152 |
+
[245.54 --> 246.74] And now, all right, Steve.
|
| 153 |
+
[246.74 --> 248.34] Probably should have written down more than five.
|
| 154 |
+
[249.24 --> 251.46] One of the, I mean, just right off the jump,
|
| 155 |
+
[251.62 --> 253.10] the interview with Jeff Rothschild
|
| 156 |
+
[253.10 --> 256.96] and hearing him describe his path into Intel
|
| 157 |
+
[256.96 --> 259.52] was one that was just fascinating for me.
|
| 158 |
+
[259.62 --> 260.52] Oh my God, that was amazing.
|
| 159 |
+
[260.52 --> 263.72] And hearing about, you know, one,
|
| 160 |
+
[263.86 --> 267.30] how the memory manufacturing process,
|
| 161 |
+
[267.30 --> 269.68] you've got a lot of defects that you can't use,
|
| 162 |
+
[269.68 --> 272.34] how that got repackaged into a solid state product.
|
| 163 |
+
[272.44 --> 277.48] And then that product being slower than fixed head disk
|
| 164 |
+
[277.48 --> 280.00] and this creating a billion dollar issue for Intel.
|
| 165 |
+
[280.00 --> 282.50] And at the time he happens to be interviewing
|
| 166 |
+
[282.50 --> 285.74] and confidently says, oh yeah, I can fix that.
|
| 167 |
+
[285.88 --> 289.04] And then leaves, goes back to Honeywell, right?
|
| 168 |
+
[289.04 --> 289.78] Oh yeah, right.
|
| 169 |
+
[289.78 --> 291.74] Yeah, and goes back and then sees
|
| 170 |
+
[291.74 --> 293.46] in like a published report
|
| 171 |
+
[293.46 --> 295.60] that the person that he had interviewed with was saying,
|
| 172 |
+
[295.72 --> 298.70] you know, yes, we did have a supply issue.
|
| 173 |
+
[298.94 --> 300.98] We've actually resolved it to the market.
|
| 174 |
+
[301.08 --> 302.30] Basically big market announcement
|
| 175 |
+
[302.30 --> 304.66] and that the resolution was that they had hired Jeff
|
| 176 |
+
[304.66 --> 305.32] and he didn't know it.
|
| 177 |
+
[305.46 --> 306.60] Right, he's like, oh, I guess they don't need me.
|
| 178 |
+
[306.64 --> 307.56] He's like, oh, I guess I don't need to,
|
| 179 |
+
[307.58 --> 308.20] I don't need to go back.
|
| 180 |
+
[308.52 --> 310.44] It's like, oh no, it's the exact opposite of that actually.
|
| 181 |
+
[310.50 --> 310.86] You're the fix.
|
| 182 |
+
[311.00 --> 311.76] You're the fix, pal.
|
| 183 |
+
[312.10 --> 314.08] Yeah, that story was amazing.
|
| 184 |
+
[314.34 --> 316.24] And that's one of those where you want to have it
|
| 185 |
+
[316.24 --> 318.78] almost videoed, not for the way,
|
| 186 |
+
[318.78 --> 320.24] for the way he was telling it, yes,
|
| 187 |
+
[320.38 --> 323.40] but the three of us were legitimately slack-jawed.
|
| 188 |
+
[323.46 --> 323.56] Yeah.
|
| 189 |
+
[323.92 --> 326.96] I mean, that was just like a mouth agape,
|
| 190 |
+
[327.12 --> 330.18] like, oh my God, it was amazing.
|
| 191 |
+
[330.66 --> 333.46] And that was, I mean, so many of his stories were incredible.
|
| 192 |
+
[333.46 --> 335.68] And again, I mean, back to the common themes,
|
| 193 |
+
[335.84 --> 337.68] it's like, you know, in each one of these,
|
| 194 |
+
[337.74 --> 340.86] there were so many chunks where we were just soaking it in.
|
| 195 |
+
[341.04 --> 344.00] I mean, just hearing these stories that go back decades
|
| 196 |
+
[344.00 --> 345.92] and that were how these companies were built.
|
| 197 |
+
[345.92 --> 348.24] And they were so personal too.
|
| 198 |
+
[348.42 --> 349.46] I mean, I think that's what-
|
| 199 |
+
[349.46 --> 350.22] That's what I love about it.
|
| 200 |
+
[350.22 --> 351.54] That's what I think we all loved about it.
|
| 201 |
+
[351.62 --> 354.02] I mean, it's like, you're getting such a,
|
| 202 |
+
[354.16 --> 356.28] you know, who these folks actually are,
|
| 203 |
+
[356.38 --> 356.94] which is great.
|
| 204 |
+
[357.56 --> 357.80] All right.
|
| 205 |
+
[357.86 --> 360.24] So mine, that actually,
|
| 206 |
+
[360.32 --> 361.00] I was a little bit worried
|
| 207 |
+
[361.00 --> 362.42] this one was going to get stolen in the first round.
|
| 208 |
+
[362.54 --> 364.08] So I'm not sure how many people have this.
|
| 209 |
+
[364.54 --> 366.28] Ron Minnick and Dr. Nim.
|
| 210 |
+
[366.68 --> 368.14] Oh, that's on my list.
|
| 211 |
+
[368.30 --> 369.14] It's not on my list,
|
| 212 |
+
[369.20 --> 370.36] but that's a good one that I forgot.
|
| 213 |
+
[370.38 --> 370.92] That's my number three.
|
| 214 |
+
[370.92 --> 371.18] All right.
|
| 215 |
+
[371.24 --> 371.44] Okay.
|
| 216 |
+
[371.50 --> 376.64] So, and this is using kind of plastic tumblers effectively
|
| 217 |
+
[376.64 --> 378.68] and marbles to emulate a computer.
|
| 218 |
+
[379.10 --> 380.18] And based on this,
|
| 219 |
+
[380.22 --> 382.34] and Steve, during the conversation,
|
| 220 |
+
[382.34 --> 382.68] you're like,
|
| 221 |
+
[382.72 --> 383.78] I gotta go find one of these on eBay.
|
| 222 |
+
[383.96 --> 384.78] And you can find one on eBay
|
| 223 |
+
[384.78 --> 386.32] for a couple hundred bucks or whatever, right?
|
| 224 |
+
[386.50 --> 386.66] Yeah.
|
| 225 |
+
[386.68 --> 388.06] I've got three kids under six.
|
| 226 |
+
[388.24 --> 390.68] One who I thought maybe getting towards six
|
| 227 |
+
[390.68 --> 392.06] might be appropriate for,
|
| 228 |
+
[392.64 --> 394.04] found it on eBay.
|
| 229 |
+
[394.26 --> 394.44] Right.
|
| 230 |
+
[394.56 --> 395.96] Well, so you should,
|
| 231 |
+
[396.42 --> 397.46] instead of getting that,
|
| 232 |
+
[397.70 --> 399.62] you should buy this thing called Turing Tumblr.
|
| 233 |
+
[399.62 --> 401.84] Well, I'm not buying that because it was $176.
|
| 234 |
+
[402.28 --> 404.12] Well, so Turing Tumblr is a lot cheaper.
|
| 235 |
+
[404.22 --> 404.96] It's like 70 bucks.
|
| 236 |
+
[405.10 --> 407.06] And it is a modern example,
|
| 237 |
+
[407.50 --> 408.54] but it's much more,
|
| 238 |
+
[408.88 --> 409.56] it's elaborate.
|
| 239 |
+
[409.70 --> 410.30] It's well thought out.
|
| 240 |
+
[410.82 --> 413.58] And so I bought this for the kids for Christmas.
|
| 241 |
+
[413.74 --> 414.36] Of course,
|
| 242 |
+
[414.86 --> 416.46] I think I've probably been playing with it
|
| 243 |
+
[416.46 --> 417.28] more than anyone else.
|
| 244 |
+
[417.56 --> 419.52] But it's been really interesting to watch.
|
| 245 |
+
[419.58 --> 421.26] I've got a 15-year-old,
|
| 246 |
+
[421.36 --> 421.92] a 12-year-old,
|
| 247 |
+
[421.96 --> 422.58] and a seven-year-old.
|
| 248 |
+
[422.62 --> 424.54] And they all got into it to varying degrees.
|
| 249 |
+
[424.94 --> 427.30] And it gets hard.
|
| 250 |
+
[427.86 --> 428.78] It gets hard.
|
| 251 |
+
[428.78 --> 429.32] And actually,
|
| 252 |
+
[429.38 --> 429.60] it was like,
|
| 253 |
+
[429.74 --> 430.58] I got one of the,
|
| 254 |
+
[430.62 --> 431.92] I think one of the four-star problems.
|
| 255 |
+
[432.04 --> 432.40] And I'm like,
|
| 256 |
+
[432.64 --> 432.86] I'm like,
|
| 257 |
+
[432.90 --> 433.14] okay,
|
| 258 |
+
[433.40 --> 434.84] four-star must be like the top.
|
| 259 |
+
[434.96 --> 435.44] You go to the end,
|
| 260 |
+
[435.48 --> 435.60] I'm like,
|
| 261 |
+
[435.64 --> 435.90] oh my God,
|
| 262 |
+
[435.92 --> 437.54] there are 10-star problems in this thing.
|
| 263 |
+
[438.16 --> 438.84] And it's good.
|
| 264 |
+
[438.92 --> 439.04] I mean,
|
| 265 |
+
[439.06 --> 439.46] it's like,
|
| 266 |
+
[439.54 --> 440.10] it's not,
|
| 267 |
+
[440.46 --> 441.50] it's not exactly,
|
| 268 |
+
[441.66 --> 443.26] it's computational thinking for sure.
|
| 269 |
+
[443.52 --> 446.98] So Ron's story was very directly inspirational in that regard.
|
| 270 |
+
[447.22 --> 447.54] But also,
|
| 271 |
+
[447.64 --> 450.12] which the way Ron was telling it was so captivating
|
| 272 |
+
[450.12 --> 450.98] about,
|
| 273 |
+
[451.06 --> 451.30] you know,
|
| 274 |
+
[451.30 --> 452.30] learning about,
|
| 275 |
+
[452.30 --> 452.56] you know,
|
| 276 |
+
[452.62 --> 454.74] DeMorgan's laws and so on as a,
|
| 277 |
+
[454.78 --> 455.40] as a kid.
|
| 278 |
+
[455.40 --> 457.44] And so I thought that was great.
|
| 279 |
+
[457.44 --> 458.84] Because his dad didn't steal it.
|
| 280 |
+
[459.02 --> 460.18] He was actually able to use his.
|
| 281 |
+
[460.42 --> 461.84] I hate him not stealing it from like,
|
| 282 |
+
[461.88 --> 462.02] well,
|
| 283 |
+
[462.22 --> 462.48] okay,
|
| 284 |
+
[462.54 --> 464.26] actually this is where my kids would like to speak for themselves.
|
| 285 |
+
[464.36 --> 464.54] And like,
|
| 286 |
+
[464.58 --> 464.72] dad,
|
| 287 |
+
[464.76 --> 466.70] you actually did tell us to go do something else.
|
| 288 |
+
[466.86 --> 467.36] Don't play on it.
|
| 289 |
+
[467.76 --> 468.32] And actually,
|
| 290 |
+
[468.38 --> 469.00] you know what you do?
|
| 291 |
+
[469.04 --> 469.44] It's pathetic.
|
| 292 |
+
[469.58 --> 470.82] I actually want a,
|
| 293 |
+
[470.82 --> 471.94] a phone.
|
| 294 |
+
[472.30 --> 475.40] I want actually a mobile game version of Turing Tumblr.
|
| 295 |
+
[475.48 --> 476.00] Like they,
|
| 296 |
+
[476.00 --> 476.80] they pride themselves.
|
| 297 |
+
[476.80 --> 477.00] Like,
|
| 298 |
+
[477.06 --> 477.14] oh,
|
| 299 |
+
[477.14 --> 478.26] this will get your kids off their phone.
|
| 300 |
+
[478.32 --> 478.42] I'm like,
|
| 301 |
+
[478.46 --> 478.64] actually,
|
| 302 |
+
[478.66 --> 478.74] no,
|
| 303 |
+
[478.74 --> 479.78] I want this on the phone.
|
| 304 |
+
[479.86 --> 481.72] So the kids will do this on their phone because it's,
|
| 305 |
+
[481.80 --> 482.54] it's that good.
|
| 306 |
+
[483.10 --> 483.42] Anyway,
|
| 307 |
+
[483.56 --> 484.32] really good stuff.
|
| 308 |
+
[484.54 --> 484.70] So,
|
| 309 |
+
[484.84 --> 486.96] but I thought that was on my list and Ron was terrific.
|
| 310 |
+
[487.04 --> 487.26] All right.
|
| 311 |
+
[487.28 --> 487.92] Now I'm down to four.
|
| 312 |
+
[488.40 --> 488.76] Exactly.
|
| 313 |
+
[488.92 --> 489.18] Jess.
|
| 314 |
+
[489.52 --> 493.80] So this one's probably not on anyone else's list because it's something that we
|
| 315 |
+
[493.80 --> 496.40] found out after an episode aired.
|
| 316 |
+
[496.40 --> 501.30] So Kenneth was telling me that during Rick's episode where he's talking about how he
|
| 317 |
+
[501.30 --> 504.20] scan the internet for all the vulnerable VMCs.
|
| 318 |
+
[504.38 --> 506.86] He did that from Kenneth's infrastructure,
|
| 319 |
+
[507.36 --> 507.60] which is,
|
| 320 |
+
[507.78 --> 509.98] it's such a great on the metal tie.
|
| 321 |
+
[510.20 --> 510.38] And yeah,
|
| 322 |
+
[510.44 --> 510.72] exactly.
|
| 323 |
+
[510.72 --> 513.20] Cause I love the interconnected world.
|
| 324 |
+
[513.32 --> 515.22] The interconnected worlds are great.
|
| 325 |
+
[515.22 --> 516.26] And I thought,
|
| 326 |
+
[516.34 --> 516.56] I mean,
|
| 327 |
+
[516.56 --> 519.50] there's so much about Kenneth story that really,
|
| 328 |
+
[519.50 --> 520.10] I thought it was,
|
| 329 |
+
[520.20 --> 525.18] what was interesting is the degree that one really resonated with people listening to
|
| 330 |
+
[525.18 --> 525.62] on the metal.
|
| 331 |
+
[525.72 --> 526.08] Yeah.
|
| 332 |
+
[526.16 --> 526.34] Well,
|
| 333 |
+
[526.34 --> 529.40] because it's like the Forrest Gump of like an internet exchange.
|
| 334 |
+
[529.50 --> 529.80] Okay.
|
| 335 |
+
[529.82 --> 530.12] Careful,
|
| 336 |
+
[530.12 --> 530.42] careful.
|
| 337 |
+
[530.42 --> 530.92] Cause this is,
|
| 338 |
+
[531.00 --> 532.62] we're starting to like get into other ones.
|
| 339 |
+
[532.96 --> 533.40] Okay.
|
| 340 |
+
[534.18 --> 535.40] Steve clearly getting finished.
|
| 341 |
+
[535.62 --> 540.44] Go watch myself.
|
| 342 |
+
[540.66 --> 540.86] Yeah.
|
| 343 |
+
[541.04 --> 541.50] But that,
|
| 344 |
+
[541.50 --> 541.98] but that,
|
| 345 |
+
[541.98 --> 543.12] that is amazing that.
|
| 346 |
+
[543.22 --> 545.58] So Rick was obviously aware of Kenneth's,
|
| 347 |
+
[545.58 --> 547.34] of the Fremont internet cabal.
|
| 348 |
+
[547.46 --> 547.62] And,
|
| 349 |
+
[547.62 --> 548.24] um,
|
| 350 |
+
[548.24 --> 548.46] it's,
|
| 351 |
+
[548.50 --> 548.98] it's great.
|
| 352 |
+
[550.14 --> 550.80] All right,
|
| 353 |
+
[550.82 --> 551.22] Steve,
|
| 354 |
+
[551.66 --> 553.88] you want to get that one out there before it gets accidentally stolen?
|
| 355 |
+
[554.26 --> 554.70] Uh,
|
| 356 |
+
[555.02 --> 555.54] well,
|
| 357 |
+
[555.60 --> 556.08] so yes,
|
| 358 |
+
[556.08 --> 561.74] because it was Kenneth's where he was describing basically the DIY ASP,
|
| 359 |
+
[561.74 --> 562.90] like do it yourself,
|
| 360 |
+
[563.14 --> 564.10] create an ASP network.
|
| 361 |
+
[564.10 --> 564.40] And,
|
| 362 |
+
[564.40 --> 568.08] and just the way that he was describing,
|
| 363 |
+
[568.08 --> 576.76] putting this thing together in building blocks and how folks across the aisle from each other in racks were running cables illegally.
|
| 364 |
+
[576.76 --> 578.52] And before you know it,
|
| 365 |
+
[578.52 --> 578.92] you know,
|
| 366 |
+
[578.92 --> 580.30] with a low ASP number,
|
| 367 |
+
[580.76 --> 581.14] he's,
|
| 368 |
+
[581.22 --> 581.56] uh,
|
| 369 |
+
[581.56 --> 581.88] he's,
|
| 370 |
+
[581.88 --> 582.82] he's created something that,
|
| 371 |
+
[582.98 --> 583.16] you know,
|
| 372 |
+
[583.16 --> 583.82] I personally,
|
| 373 |
+
[584.12 --> 584.86] thank you.
|
| 374 |
+
[585.10 --> 585.28] Yeah.
|
| 375 |
+
[585.28 --> 585.60] Uh,
|
| 376 |
+
[585.60 --> 586.90] I personally had thought of as,
|
| 377 |
+
[586.98 --> 587.22] you know,
|
| 378 |
+
[587.26 --> 587.74] being these,
|
| 379 |
+
[587.80 --> 588.42] these very,
|
| 380 |
+
[588.54 --> 593.64] very large companies with tons of infrastructure to be able to connect out in that way.
|
| 381 |
+
[593.74 --> 596.84] Can I tell you how hard I tried in doing the show notes for Kenneth?
|
| 382 |
+
[597.00 --> 600.78] I wanted to find the architectural firm that was sitting on his ASN.
|
| 383 |
+
[600.78 --> 600.80] Yes,
|
| 384 |
+
[600.80 --> 601.24] that had the low ASN.
|
| 385 |
+
[601.32 --> 601.56] God,
|
| 386 |
+
[601.66 --> 601.80] I tried,
|
| 387 |
+
[601.90 --> 603.34] because he just said it's an architectural firm.
|
| 388 |
+
[603.36 --> 603.64] And I'm like,
|
| 389 |
+
[603.72 --> 606.16] I got to know the story of this architectural firm.
|
| 390 |
+
[606.40 --> 609.96] And I tried really hard to find the architectural firm sitting on that ASN.
|
| 391 |
+
[610.00 --> 610.18] And I,
|
| 392 |
+
[610.18 --> 611.02] I couldn't find them.
|
| 393 |
+
[611.98 --> 612.86] And unfortunately,
|
| 394 |
+
[612.96 --> 614.30] like we know that they went out of business,
|
| 395 |
+
[614.30 --> 616.00] but that's how Kenneth got the number.
|
| 396 |
+
[616.00 --> 616.24] Right.
|
| 397 |
+
[616.24 --> 617.40] So it's like,
|
| 398 |
+
[617.52 --> 618.48] like they're already,
|
| 399 |
+
[618.66 --> 619.18] you can't find it.
|
| 400 |
+
[619.22 --> 619.36] Yeah.
|
| 401 |
+
[619.36 --> 620.52] It's actually already not.
|
| 402 |
+
[620.60 --> 622.02] And they probably went out of business because they're like,
|
| 403 |
+
[622.04 --> 623.08] not a very good architectural firm.
|
| 404 |
+
[623.10 --> 624.34] So it's like already like the,
|
| 405 |
+
[624.34 --> 624.54] the,
|
| 406 |
+
[624.98 --> 625.94] it was against me and like,
|
| 407 |
+
[625.98 --> 626.66] I did not find it.
|
| 408 |
+
[626.66 --> 629.10] But learning the lengths people would go to save,
|
| 409 |
+
[629.16 --> 629.32] you know,
|
| 410 |
+
[629.38 --> 636.06] 50 bucks a month and bring a concrete drill into a data center to drill in from the floor above and run cabling.
|
| 411 |
+
[636.30 --> 638.20] That was another one of those that was just very evocative.
|
| 412 |
+
[638.32 --> 641.64] I felt like I was in the data center and I loved kind of like,
|
| 413 |
+
[641.84 --> 644.28] you just got to like walk in with a hard hat and a clipboard.
|
| 414 |
+
[644.30 --> 645.86] And just look like you're official.
|
| 415 |
+
[646.00 --> 646.74] It'll look like you're official.
|
| 416 |
+
[649.04 --> 649.90] That'll get you anywhere.
|
| 417 |
+
[650.10 --> 650.32] Yeah.
|
| 418 |
+
[650.44 --> 650.62] No,
|
| 419 |
+
[650.66 --> 651.22] that was,
|
| 420 |
+
[651.30 --> 651.70] uh,
|
| 421 |
+
[651.70 --> 652.48] that was terrific.
|
| 422 |
+
[653.20 --> 653.52] Uh,
|
| 423 |
+
[653.52 --> 655.16] so another favorite moment of mine.
|
| 424 |
+
[655.24 --> 655.38] It says,
|
| 425 |
+
[655.46 --> 655.64] Steve,
|
| 426 |
+
[655.68 --> 657.12] this is the one that you missed.
|
| 427 |
+
[657.16 --> 658.56] You were out of town when,
|
| 428 |
+
[658.70 --> 658.86] Oh,
|
| 429 |
+
[658.92 --> 659.60] Tom Lyon.
|
| 430 |
+
[659.76 --> 659.98] Yeah.
|
| 431 |
+
[659.98 --> 661.96] And Tom talking about data point.
|
| 432 |
+
[661.96 --> 666.08] And he mentioned this company data point and he recommended a book in particular.
|
| 433 |
+
[666.92 --> 668.62] So book we put in the show notes at the time,
|
| 434 |
+
[668.62 --> 669.38] but data point,
|
| 435 |
+
[669.46 --> 672.38] the lost story of the Texans who invented the personal computer revolution.
|
| 436 |
+
[672.54 --> 675.72] I got the book and read it and really enjoyed it.
|
| 437 |
+
[675.72 --> 676.00] I mean,
|
| 438 |
+
[676.06 --> 680.10] really interesting story about a,
|
| 439 |
+
[680.10 --> 682.40] a company that had been forgotten about.
|
| 440 |
+
[682.46 --> 682.86] And I thought,
|
| 441 |
+
[682.96 --> 683.48] you know,
|
| 442 |
+
[683.56 --> 690.72] one of the things I love about Tom is that Tom is also likes the history of the industry.
|
| 443 |
+
[690.72 --> 691.76] And of course,
|
| 444 |
+
[691.76 --> 694.38] because he's been in it for so much longer than we have,
|
| 445 |
+
[694.94 --> 698.88] he's got this incredible depth of knowledge and interest.
|
| 446 |
+
[699.58 --> 699.98] And,
|
| 447 |
+
[699.98 --> 700.66] and he,
|
| 448 |
+
[700.98 --> 702.52] after the podcast,
|
| 449 |
+
[702.74 --> 702.86] he,
|
| 450 |
+
[703.00 --> 703.16] we,
|
| 451 |
+
[703.26 --> 705.18] he was showing us his manual collection,
|
| 452 |
+
[705.34 --> 705.56] Jess.
|
| 453 |
+
[705.60 --> 706.30] And that was,
|
| 454 |
+
[706.76 --> 708.22] that was just such a great day.
|
| 455 |
+
[708.22 --> 713.10] It was so much fun and just delightful stories.
|
| 456 |
+
[713.10 --> 715.04] And I just loved his interest.
|
| 457 |
+
[715.28 --> 718.28] And I guess he has the belief that I think the three of us have,
|
| 458 |
+
[718.32 --> 719.86] and I think our listeners have,
|
| 459 |
+
[719.86 --> 722.72] which is that history has got a lot to teach us that it,
|
| 460 |
+
[722.94 --> 725.34] we're not just interested in this out of,
|
| 461 |
+
[725.56 --> 727.92] because out of a kind of an antiquarian fascination,
|
| 462 |
+
[727.92 --> 731.66] it's because we see a lot of parallels with the present and future.
|
| 463 |
+
[731.86 --> 734.70] And we think that there's really something to understanding this stuff.
|
| 464 |
+
[734.80 --> 735.50] So I love that.
|
| 465 |
+
[735.58 --> 737.60] I love Tom's description of data point and arc net.
|
| 466 |
+
[737.96 --> 738.28] Yeah.
|
| 467 |
+
[738.28 --> 738.58] I was,
|
| 468 |
+
[738.64 --> 739.58] I was bummed to miss that one,
|
| 469 |
+
[739.62 --> 741.96] but also it was nice to be able to listen to one.
|
| 470 |
+
[742.18 --> 742.54] Oh,
|
| 471 |
+
[742.58 --> 742.90] interesting.
|
| 472 |
+
[743.02 --> 743.20] Yeah.
|
| 473 |
+
[743.46 --> 744.16] And he,
|
| 474 |
+
[744.22 --> 744.98] he was fascinating.
|
| 475 |
+
[745.24 --> 745.60] And,
|
| 476 |
+
[745.66 --> 746.06] and I mean,
|
| 477 |
+
[746.14 --> 746.38] God,
|
| 478 |
+
[746.48 --> 747.10] one of what,
|
| 479 |
+
[747.32 --> 747.88] nine kids.
|
| 480 |
+
[747.88 --> 749.08] That was another moment.
|
| 481 |
+
[749.24 --> 749.84] Like that family,
|
| 482 |
+
[749.86 --> 750.60] the family,
|
| 483 |
+
[750.88 --> 751.22] the family,
|
| 484 |
+
[751.22 --> 753.58] I know family took over.
|
| 485 |
+
[753.82 --> 754.06] They,
|
| 486 |
+
[754.14 --> 754.96] I know you think,
|
| 487 |
+
[755.02 --> 756.94] and like this family should be,
|
| 488 |
+
[757.00 --> 758.40] and he was talking about this family reunion.
|
| 489 |
+
[758.40 --> 758.80] Yeah.
|
| 490 |
+
[759.16 --> 762.04] And it's like the computer history museum should,
|
| 491 |
+
[762.16 --> 765.52] should send a documentary crew or just this family reunion,
|
| 492 |
+
[765.66 --> 766.56] let them hold it there.
|
| 493 |
+
[766.56 --> 766.96] You know,
|
| 494 |
+
[767.02 --> 767.80] I guess.
|
| 495 |
+
[767.98 --> 768.26] Right.
|
| 496 |
+
[768.84 --> 769.14] No,
|
| 497 |
+
[769.18 --> 769.64] it was amazing.
|
| 498 |
+
[769.66 --> 771.34] And then the optical mouse and the,
|
| 499 |
+
[771.34 --> 771.48] the,
|
| 500 |
+
[771.48 --> 771.52] the,
|
| 501 |
+
[771.52 --> 771.54] the,
|
| 502 |
+
[771.54 --> 771.94] it was like,
|
| 503 |
+
[772.16 --> 772.48] and he,
|
| 504 |
+
[772.48 --> 773.04] you know,
|
| 505 |
+
[773.04 --> 774.08] looks behind at the,
|
| 506 |
+
[774.16 --> 774.28] you know,
|
| 507 |
+
[774.28 --> 776.48] the image of the optical mouse behind you in the garage.
|
| 508 |
+
[776.48 --> 776.84] And like,
|
| 509 |
+
[776.86 --> 777.02] it's like,
|
| 510 |
+
[777.02 --> 777.14] yeah,
|
| 511 |
+
[777.18 --> 777.44] you know,
|
| 512 |
+
[777.44 --> 777.88] that's a,
|
| 513 |
+
[778.40 --> 784.06] that's that actually the one my brother invented is actually the superior implementation of an obstacle mouse.
|
| 514 |
+
[784.22 --> 784.34] It's like,
|
| 515 |
+
[784.34 --> 785.16] it's like,
|
| 516 |
+
[785.22 --> 785.58] where are we?
|
| 517 |
+
[785.66 --> 785.78] No,
|
| 518 |
+
[785.80 --> 786.28] it was amazing.
|
| 519 |
+
[786.56 --> 786.80] It was,
|
| 520 |
+
[786.84 --> 787.44] that was such a,
|
| 521 |
+
[788.40 --> 788.74] all right.
|
| 522 |
+
[788.74 --> 789.58] Okay.
|
| 523 |
+
[789.58 --> 793.84] So my next one is also a kind of behind the scenes thing,
|
| 524 |
+
[793.84 --> 800.96] but the three hour long episode with Jonathan blow where like Steve had to leave in the middle for like a call,
|
| 525 |
+
[800.96 --> 801.90] but came back.
|
| 526 |
+
[801.90 --> 806.04] And then I had to like move a task rabbit that was supposed to assemble furniture.
|
| 527 |
+
[806.36 --> 807.08] Cause I was like,
|
| 528 |
+
[807.12 --> 808.24] this is just getting good.
|
| 529 |
+
[808.64 --> 808.84] Oh,
|
| 530 |
+
[809.02 --> 809.56] that was,
|
| 531 |
+
[809.82 --> 812.74] and he was just gathering steam.
|
| 532 |
+
[812.82 --> 815.02] Like we could have gone on for so much longer,
|
| 533 |
+
[815.06 --> 816.06] but it was freezing.
|
| 534 |
+
[816.06 --> 819.12] It is so cold in this garage.
|
| 535 |
+
[819.26 --> 820.56] I also like legitimately,
|
| 536 |
+
[820.56 --> 823.14] like I'm going to miss my flight into New Zealand.
|
| 537 |
+
[823.28 --> 824.60] We were going to New Zealand for,
|
| 538 |
+
[824.74 --> 826.30] for the holidays.
|
| 539 |
+
[826.30 --> 827.66] My wife's family's from New Zealand.
|
| 540 |
+
[827.80 --> 828.00] I'm like,
|
| 541 |
+
[828.00 --> 828.98] you couldn't even copy it.
|
| 542 |
+
[829.02 --> 832.40] We had to hide the SD card because like we were concerned.
|
| 543 |
+
[833.96 --> 834.32] Literally.
|
| 544 |
+
[834.32 --> 835.88] I did not have time to copy the SD card.
|
| 545 |
+
[835.94 --> 836.40] That's true.
|
| 546 |
+
[836.50 --> 837.50] That if I copied the SD card,
|
| 547 |
+
[837.50 --> 841.30] I was in danger of missing Christmas by that.
|
| 548 |
+
[841.36 --> 841.86] That's what,
|
| 549 |
+
[841.86 --> 843.28] what the stakes were,
|
| 550 |
+
[843.34 --> 844.54] but that could have gone on.
|
| 551 |
+
[844.60 --> 845.74] It felt like I couldn't believe it.
|
| 552 |
+
[845.78 --> 848.66] I had to go out for an interview and came back in 45 minutes later.
|
| 553 |
+
[848.74 --> 851.02] And we're still humming gathering strength.
|
| 554 |
+
[851.06 --> 851.28] Yeah.
|
| 555 |
+
[851.48 --> 856.74] It was like a hurricane that had crossed Florida into the Gulf of Mexico.
|
| 556 |
+
[856.74 --> 857.50] And just like,
|
| 557 |
+
[857.58 --> 859.44] and all of a sudden like cat five.
|
| 558 |
+
[860.06 --> 860.72] And no,
|
| 559 |
+
[860.82 --> 861.06] he,
|
| 560 |
+
[861.14 --> 861.84] it was amazing.
|
| 561 |
+
[861.84 --> 862.26] I thought.
|
| 562 |
+
[862.44 --> 863.04] So good.
|
| 563 |
+
[863.16 --> 863.34] And,
|
| 564 |
+
[863.46 --> 866.40] and it seems like everyone seems to be loving the length.
|
| 565 |
+
[866.76 --> 867.08] Oh,
|
| 566 |
+
[867.18 --> 867.42] and so,
|
| 567 |
+
[867.50 --> 868.32] which is,
|
| 568 |
+
[868.40 --> 868.66] so we,
|
| 569 |
+
[868.84 --> 871.20] the first long one was Jeff Rothschild.
|
| 570 |
+
[871.32 --> 872.00] Cause when we got into this,
|
| 571 |
+
[872.02 --> 872.40] we thought like,
|
| 572 |
+
[872.58 --> 872.86] Oh,
|
| 573 |
+
[872.88 --> 876.26] like these will be like 30 minute conversations during your commute,
|
| 574 |
+
[876.44 --> 877.34] but we kind of knew like,
|
| 575 |
+
[877.36 --> 877.50] ah,
|
| 576 |
+
[877.50 --> 878.76] 30 minutes is going to be a bit short.
|
| 577 |
+
[879.24 --> 882.88] And then we had Jeff Rothschild was the first one that we,
|
| 578 |
+
[882.98 --> 886.36] there was no way we were going to cut him off.
|
| 579 |
+
[886.74 --> 887.60] And another one,
|
| 580 |
+
[887.68 --> 889.22] another one that could have gone six hours,
|
| 581 |
+
[889.24 --> 889.96] could have gone six hours.
|
| 582 |
+
[889.96 --> 890.76] And when we ended,
|
| 583 |
+
[890.88 --> 891.30] he was like,
|
| 584 |
+
[891.34 --> 891.48] wow,
|
| 585 |
+
[891.50 --> 891.72] all right,
|
| 586 |
+
[891.72 --> 892.28] are we already done?
|
| 587 |
+
[892.34 --> 892.50] Like,
|
| 588 |
+
[892.52 --> 892.64] God,
|
| 589 |
+
[892.70 --> 893.08] we,
|
| 590 |
+
[893.08 --> 897.40] we didn't want to like continue to like force you to lunchtime.
|
| 591 |
+
[897.94 --> 898.50] But the,
|
| 592 |
+
[898.82 --> 899.86] and then we had a kind of a,
|
| 593 |
+
[899.86 --> 901.60] an extended discussion about like,
|
| 594 |
+
[901.68 --> 905.80] do we do this in like two parts or do we do it as a one parter?
|
| 595 |
+
[905.80 --> 906.48] And we decided like,
|
| 596 |
+
[906.50 --> 906.92] you know what?
|
| 597 |
+
[907.12 --> 909.40] People can hit the pause button if they don't want to.
|
| 598 |
+
[909.42 --> 910.14] Cause it's like Netflix.
|
| 599 |
+
[910.30 --> 911.92] The episodes don't have a set amount of time.
|
| 600 |
+
[911.98 --> 912.46] You just go,
|
| 601 |
+
[912.54 --> 913.00] you just go.
|
| 602 |
+
[913.10 --> 913.66] And it's like,
|
| 603 |
+
[913.66 --> 918.76] we knew that we would not have wanted to be interrupted halfway through that.
|
| 604 |
+
[918.90 --> 919.06] Yeah.
|
| 605 |
+
[919.72 --> 920.38] So it's like,
|
| 606 |
+
[920.38 --> 922.54] don't give me the first half without giving me the second half.
|
| 607 |
+
[922.58 --> 924.12] So we decided to just let them go.
|
| 608 |
+
[924.40 --> 924.92] Long form.
|
| 609 |
+
[925.30 --> 925.90] Long form.
|
| 610 |
+
[926.12 --> 929.12] And Jonathan's was the Naples ultra in that department.
|
| 611 |
+
[929.76 --> 930.44] And it was good.
|
| 612 |
+
[930.64 --> 931.18] So good.
|
| 613 |
+
[931.34 --> 931.76] It was good.
|
| 614 |
+
[931.82 --> 931.90] I,
|
| 615 |
+
[932.02 --> 932.18] yeah,
|
| 616 |
+
[932.26 --> 932.64] I,
|
| 617 |
+
[933.12 --> 933.50] and I,
|
| 618 |
+
[933.72 --> 934.60] and at the end of it,
|
| 619 |
+
[934.64 --> 934.90] I mean,
|
| 620 |
+
[934.90 --> 935.66] it was kind of like,
|
| 621 |
+
[935.68 --> 936.90] I felt like the three of us were,
|
| 622 |
+
[937.02 --> 938.96] he was gathering strength.
|
| 623 |
+
[939.02 --> 939.10] Well,
|
| 624 |
+
[939.10 --> 940.64] the three of us were like trying to hold on.
|
| 625 |
+
[940.70 --> 940.82] Well,
|
| 626 |
+
[940.86 --> 941.06] my,
|
| 627 |
+
[941.06 --> 942.98] my feet were literally frozen after that.
|
| 628 |
+
[942.98 --> 944.20] Like actually frozen.
|
| 629 |
+
[944.60 --> 947.20] It did take a couple hours to get body temperature back up again.
|
| 630 |
+
[947.28 --> 947.46] Right.
|
| 631 |
+
[947.52 --> 948.40] And I did make my flight,
|
| 632 |
+
[948.52 --> 950.20] but like not with a lot of margin.
|
| 633 |
+
[950.44 --> 950.76] I,
|
| 634 |
+
[950.76 --> 950.90] I,
|
| 635 |
+
[950.90 --> 952.06] I don't regret,
|
| 636 |
+
[952.06 --> 952.54] uh,
|
| 637 |
+
[952.54 --> 954.22] having not copied it off the SD card.
|
| 638 |
+
[954.94 --> 955.40] All right,
|
| 639 |
+
[955.42 --> 955.64] Steve,
|
| 640 |
+
[955.70 --> 956.12] what do you got?
|
| 641 |
+
[956.22 --> 956.26] Uh,
|
| 642 |
+
[956.26 --> 956.46] what's,
|
| 643 |
+
[956.46 --> 956.98] what's next?
|
| 644 |
+
[957.92 --> 958.46] I think,
|
| 645 |
+
[958.54 --> 961.64] I think one of the themes that was pretty interesting to,
|
| 646 |
+
[961.82 --> 963.22] to hear was,
|
| 647 |
+
[963.22 --> 964.80] I think everyone held this in common,
|
| 648 |
+
[964.82 --> 967.46] but some people took it more seriously than others is,
|
| 649 |
+
[967.46 --> 968.28] uh,
|
| 650 |
+
[968.34 --> 969.52] preservation of old hardware.
|
| 651 |
+
[969.88 --> 970.28] So,
|
| 652 |
+
[970.44 --> 970.56] yeah.
|
| 653 |
+
[970.60 --> 971.86] And it wasn't just,
|
| 654 |
+
[971.86 --> 972.28] you know,
|
| 655 |
+
[972.30 --> 973.92] I can't throw out old hardware.
|
| 656 |
+
[973.92 --> 974.44] It was,
|
| 657 |
+
[974.56 --> 977.00] I seek out and collect hardware until my,
|
| 658 |
+
[977.26 --> 977.54] you know,
|
| 659 |
+
[977.60 --> 978.04] my,
|
| 660 |
+
[978.62 --> 981.92] where I live is brimming with things and causing family issues.
|
| 661 |
+
[981.92 --> 982.26] Right.
|
| 662 |
+
[982.36 --> 983.72] So in particular,
|
| 663 |
+
[983.72 --> 988.34] I think John masters had been holding onto hardware in his,
|
| 664 |
+
[988.38 --> 988.68] uh,
|
| 665 |
+
[988.68 --> 990.68] in his domicile until that was causing an issue.
|
| 666 |
+
[990.68 --> 991.78] Then had to get a storage locker,
|
| 667 |
+
[992.36 --> 993.26] filled that thing with the brim.
|
| 668 |
+
[993.26 --> 994.10] He said on the podcast,
|
| 669 |
+
[994.10 --> 994.28] like,
|
| 670 |
+
[994.30 --> 996.12] I'm pretty sure my fiance doesn't know about this.
|
| 671 |
+
[996.12 --> 996.46] And I'm like,
|
| 672 |
+
[996.46 --> 996.64] right.
|
| 673 |
+
[996.64 --> 997.86] I'm sure she does now.
|
| 674 |
+
[998.16 --> 999.06] And did you see her tweet?
|
| 675 |
+
[999.70 --> 999.84] No.
|
| 676 |
+
[1000.14 --> 1000.46] Oh yeah.
|
| 677 |
+
[1000.52 --> 1001.22] Her tweet was like,
|
| 678 |
+
[1001.28 --> 1004.10] I guess I learned there's a storage locker filled with hardware.
|
| 679 |
+
[1004.62 --> 1005.24] How many,
|
| 680 |
+
[1005.30 --> 1006.80] how many other storage lockers?
|
| 681 |
+
[1007.30 --> 1007.66] Yeah.
|
| 682 |
+
[1007.76 --> 1008.12] It's like,
|
| 683 |
+
[1008.16 --> 1008.42] Oh,
|
| 684 |
+
[1008.52 --> 1008.82] wow.
|
| 685 |
+
[1008.92 --> 1009.26] Okay.
|
| 686 |
+
[1009.56 --> 1009.94] Sorry.
|
| 687 |
+
[1010.22 --> 1010.60] Don't want to,
|
| 688 |
+
[1010.64 --> 1012.86] we're not trying to create any domestic strife.
|
| 689 |
+
[1013.14 --> 1013.56] That's so good.
|
| 690 |
+
[1013.68 --> 1017.50] But then you also had Rick wanted to buy a cray as furniture.
|
| 691 |
+
[1017.60 --> 1017.84] Right.
|
| 692 |
+
[1018.20 --> 1018.38] Yeah.
|
| 693 |
+
[1018.40 --> 1018.60] Right.
|
| 694 |
+
[1018.70 --> 1019.00] And I,
|
| 695 |
+
[1019.16 --> 1019.38] you know,
|
| 696 |
+
[1019.38 --> 1020.06] looking that up,
|
| 697 |
+
[1020.06 --> 1022.06] it's like that actually could serve as,
|
| 698 |
+
[1022.24 --> 1022.78] as a couch.
|
| 699 |
+
[1022.90 --> 1023.12] Yeah,
|
| 700 |
+
[1023.28 --> 1023.66] exactly.
|
| 701 |
+
[1023.66 --> 1024.14] And,
|
| 702 |
+
[1024.14 --> 1027.58] and then obviously the Amir thing with the wafers on the wall.
|
| 703 |
+
[1027.68 --> 1027.86] Yeah.
|
| 704 |
+
[1027.86 --> 1033.60] There was just a lot of tie in on incorporation of either storage of hardware or incorporation of hardware.
|
| 705 |
+
[1033.68 --> 1034.52] That kind of physicality.
|
| 706 |
+
[1034.64 --> 1034.84] Yeah.
|
| 707 |
+
[1035.28 --> 1035.64] Well,
|
| 708 |
+
[1035.66 --> 1037.60] my next one actually dovetails very well with that.
|
| 709 |
+
[1037.60 --> 1040.20] And that is trammel and the,
|
| 710 |
+
[1040.28 --> 1040.52] the,
|
| 711 |
+
[1040.80 --> 1040.90] Oh,
|
| 712 |
+
[1040.90 --> 1041.54] did I just do one of yours?
|
| 713 |
+
[1041.58 --> 1041.66] No,
|
| 714 |
+
[1041.68 --> 1042.16] I don't know.
|
| 715 |
+
[1043.32 --> 1043.64] Okay.
|
| 716 |
+
[1043.82 --> 1045.40] So I'll keep it.
|
| 717 |
+
[1045.58 --> 1046.96] I'll keep it specific enough.
|
| 718 |
+
[1047.02 --> 1048.10] I'm going to walk a narrow line.
|
| 719 |
+
[1048.16 --> 1049.06] So I don't steal anything,
|
| 720 |
+
[1049.06 --> 1050.50] but the,
|
| 721 |
+
[1050.50 --> 1051.96] the teletype 33.
|
| 722 |
+
[1051.96 --> 1052.46] That's right.
|
| 723 |
+
[1052.46 --> 1052.56] That's all right.
|
| 724 |
+
[1052.62 --> 1053.10] Totally 33.
|
| 725 |
+
[1053.98 --> 1055.82] What about his description of the teletype 33?
|
| 726 |
+
[1056.06 --> 1056.30] And I,
|
| 727 |
+
[1056.62 --> 1056.88] yeah,
|
| 728 |
+
[1056.88 --> 1058.44] I actually do want to expand this now,
|
| 729 |
+
[1058.48 --> 1059.28] but I'm going to steal it.
|
| 730 |
+
[1059.38 --> 1059.94] It's fine.
|
| 731 |
+
[1060.02 --> 1061.12] I'm not even in that genre.
|
| 732 |
+
[1061.36 --> 1061.58] Okay.
|
| 733 |
+
[1061.66 --> 1061.82] Well,
|
| 734 |
+
[1061.82 --> 1062.14] the,
|
| 735 |
+
[1062.20 --> 1063.86] the way he found that,
|
| 736 |
+
[1064.04 --> 1065.44] that whole system on Craigslist,
|
| 737 |
+
[1065.54 --> 1065.70] like,
|
| 738 |
+
[1065.76 --> 1065.92] okay,
|
| 739 |
+
[1065.92 --> 1066.54] but if you take this,
|
| 740 |
+
[1066.60 --> 1067.54] you actually have to take all of it.
|
| 741 |
+
[1067.76 --> 1067.98] Right.
|
| 742 |
+
[1068.02 --> 1068.94] And he gets this,
|
| 743 |
+
[1068.98 --> 1069.34] you know,
|
| 744 |
+
[1069.38 --> 1071.66] system that's in incredible working condition.
|
| 745 |
+
[1071.66 --> 1076.24] And then the way he was describing the teletype 33 and the fact that it was mechanically
|
| 746 |
+
[1076.24 --> 1078.26] locked at this,
|
| 747 |
+
[1078.34 --> 1081.74] what is actually like 110 bod or whatever the rate was very,
|
| 748 |
+
[1081.90 --> 1082.64] very low rate.
|
| 749 |
+
[1083.08 --> 1087.16] And then that dovetailed in with a couple of different conversations,
|
| 750 |
+
[1087.16 --> 1091.46] but one dovetailed back to the Tom lion conversation where just,
|
| 751 |
+
[1091.62 --> 1092.64] we had this moment where you,
|
| 752 |
+
[1092.76 --> 1094.18] Tom and I were talking about white space,
|
| 753 |
+
[1094.24 --> 1094.82] which I loved.
|
| 754 |
+
[1094.82 --> 1097.96] And about like tabs versus spaces.
|
| 755 |
+
[1098.54 --> 1099.42] And you realize,
|
| 756 |
+
[1099.50 --> 1101.38] and I guess I just didn't realize this,
|
| 757 |
+
[1101.96 --> 1102.80] that I never,
|
| 758 |
+
[1103.24 --> 1103.64] I mean,
|
| 759 |
+
[1103.66 --> 1105.26] I've always been a hard tabs person,
|
| 760 |
+
[1105.26 --> 1107.50] but I've never really made the efficiency argument over it.
|
| 761 |
+
[1107.56 --> 1108.24] It's kind of ridiculous,
|
| 762 |
+
[1108.24 --> 1110.80] but actually with a teletype 33,
|
| 763 |
+
[1111.16 --> 1113.08] it's actually an efficiency argument.
|
| 764 |
+
[1113.08 --> 1113.36] It's like,
|
| 765 |
+
[1113.38 --> 1113.56] no,
|
| 766 |
+
[1113.70 --> 1114.56] like you don't understand.
|
| 767 |
+
[1114.68 --> 1117.90] I'm saving a lot of time by like eight spaces.
|
| 768 |
+
[1118.08 --> 1119.68] Like that's like 15 minutes.
|
| 769 |
+
[1119.70 --> 1120.14] Okay.
|
| 770 |
+
[1120.14 --> 1121.12] So like that,
|
| 771 |
+
[1121.18 --> 1123.80] that tab stop is a big deal,
|
| 772 |
+
[1123.86 --> 1124.80] which I thought was really interesting.
|
| 773 |
+
[1124.80 --> 1125.50] And just the,
|
| 774 |
+
[1125.90 --> 1129.50] I feel the way Trammell described the teletype 33,
|
| 775 |
+
[1130.18 --> 1130.48] you,
|
| 776 |
+
[1130.66 --> 1132.28] you can almost hear it.
|
| 777 |
+
[1132.28 --> 1132.86] I had to,
|
| 778 |
+
[1132.86 --> 1133.72] it was very visceral.
|
| 779 |
+
[1134.44 --> 1135.60] And then the teletype 33,
|
| 780 |
+
[1135.60 --> 1138.28] which came up a couple of times in a couple of the different conversations,
|
| 781 |
+
[1138.28 --> 1139.36] the,
|
| 782 |
+
[1139.48 --> 1140.22] in the data point,
|
| 783 |
+
[1140.42 --> 1147.00] the data points machine is the 3,300 because it's a hundred times better than the teletype 33.
|
| 784 |
+
[1147.20 --> 1148.02] They were originally,
|
| 785 |
+
[1148.34 --> 1148.76] yeah,
|
| 786 |
+
[1148.84 --> 1149.88] the terminal corporation.
|
| 787 |
+
[1150.02 --> 1150.16] I'm like,
|
| 788 |
+
[1150.26 --> 1154.24] and I just didn't know anything about the teletype 33 before getting into these.
|
| 789 |
+
[1154.30 --> 1156.22] So I thought that was really delightful and interesting.
|
| 790 |
+
[1156.44 --> 1156.64] All right.
|
| 791 |
+
[1156.64 --> 1157.42] So I did not take your,
|
| 792 |
+
[1157.42 --> 1157.60] your,
|
| 793 |
+
[1157.60 --> 1158.20] your,
|
| 794 |
+
[1158.40 --> 1159.72] send it.
|
| 795 |
+
[1159.74 --> 1160.46] But I do love,
|
| 796 |
+
[1160.56 --> 1161.42] I love the,
|
| 797 |
+
[1161.56 --> 1165.12] this is another one of the like interconnected worlds of the people on the podcast,
|
| 798 |
+
[1165.12 --> 1167.48] because there were things that came up multiple times.
|
| 799 |
+
[1167.48 --> 1173.26] And so Ron and Trammell both overlapped at Sandy labs and worked on firmware and clustering systems.
|
| 800 |
+
[1173.26 --> 1174.00] And I just like,
|
| 801 |
+
[1174.02 --> 1174.62] love the,
|
| 802 |
+
[1174.62 --> 1177.06] the two sides of that story are great.
|
| 803 |
+
[1177.06 --> 1178.66] Like they are great.
|
| 804 |
+
[1178.66 --> 1180.28] And I love having these people that,
|
| 805 |
+
[1180.36 --> 1184.28] that know one another or have worked with one another or have crossed paths in their careers.
|
| 806 |
+
[1184.28 --> 1186.54] I just thought that was so much fun.
|
| 807 |
+
[1186.54 --> 1187.08] And I think,
|
| 808 |
+
[1187.12 --> 1187.76] you know,
|
| 809 |
+
[1187.82 --> 1191.14] so many people that were on the podcast became fans of the podcast,
|
| 810 |
+
[1191.52 --> 1192.24] which was really cool.
|
| 811 |
+
[1192.68 --> 1195.14] Cause I don't think we really told people who else was on it.
|
| 812 |
+
[1195.14 --> 1198.36] We probably could have,
|
| 813 |
+
[1198.42 --> 1198.90] we just like,
|
| 814 |
+
[1198.92 --> 1199.64] it never came up.
|
| 815 |
+
[1200.00 --> 1200.32] Well,
|
| 816 |
+
[1200.36 --> 1204.22] I remember a mirror who we had on a world early.
|
| 817 |
+
[1204.30 --> 1204.82] I think he was our,
|
| 818 |
+
[1204.90 --> 1205.48] maybe our third,
|
| 819 |
+
[1205.48 --> 1207.22] when we were describing the podcast,
|
| 820 |
+
[1207.28 --> 1207.48] he was like,
|
| 821 |
+
[1207.50 --> 1207.62] Oh,
|
| 822 |
+
[1207.62 --> 1208.14] that sounds great.
|
| 823 |
+
[1208.14 --> 1208.30] Like,
|
| 824 |
+
[1208.34 --> 1209.68] I want to listen to all of them right now.
|
| 825 |
+
[1209.68 --> 1209.88] Like,
|
| 826 |
+
[1209.94 --> 1210.16] well,
|
| 827 |
+
[1210.54 --> 1212.06] you are all right now.
|
| 828 |
+
[1212.06 --> 1214.24] So you are the podcast.
|
| 829 |
+
[1214.64 --> 1215.78] We'll have you out there soon.
|
| 830 |
+
[1216.08 --> 1217.04] But I know it was,
|
| 831 |
+
[1217.16 --> 1217.78] that was great.
|
| 832 |
+
[1217.78 --> 1218.34] And,
|
| 833 |
+
[1218.34 --> 1219.74] and listening to that,
|
| 834 |
+
[1219.80 --> 1223.72] cause they basically told Trammell and Ron told the same story,
|
| 835 |
+
[1223.72 --> 1225.74] but from different perspectives.
|
| 836 |
+
[1226.06 --> 1226.24] Yeah.
|
| 837 |
+
[1226.24 --> 1226.82] And they,
|
| 838 |
+
[1227.24 --> 1227.50] right.
|
| 839 |
+
[1227.50 --> 1230.12] They probably had not known that as they were.
|
| 840 |
+
[1230.44 --> 1230.46] No,
|
| 841 |
+
[1230.50 --> 1230.76] it's good.
|
| 842 |
+
[1230.84 --> 1232.18] But I just remember in the show notes,
|
| 843 |
+
[1232.18 --> 1232.54] I was like,
|
| 844 |
+
[1232.58 --> 1232.82] Oh wow,
|
| 845 |
+
[1232.82 --> 1233.62] this is very similar.
|
| 846 |
+
[1233.72 --> 1233.80] Oh,
|
| 847 |
+
[1233.80 --> 1234.28] this is similar.
|
| 848 |
+
[1234.44 --> 1234.66] All right.
|
| 849 |
+
[1234.76 --> 1235.08] Exactly.
|
| 850 |
+
[1235.56 --> 1236.04] Yeah.
|
| 851 |
+
[1238.04 --> 1239.28] Well on a mirror,
|
| 852 |
+
[1239.40 --> 1240.20] that was,
|
| 853 |
+
[1240.32 --> 1241.32] that was another one of mine.
|
| 854 |
+
[1241.44 --> 1241.76] I,
|
| 855 |
+
[1241.76 --> 1242.22] I,
|
| 856 |
+
[1242.22 --> 1242.38] I,
|
| 857 |
+
[1242.38 --> 1242.42] I,
|
| 858 |
+
[1242.42 --> 1242.46] I,
|
| 859 |
+
[1242.46 --> 1243.26] first of all,
|
| 860 |
+
[1243.32 --> 1243.72] just that,
|
| 861 |
+
[1244.20 --> 1245.06] that was fascinating.
|
| 862 |
+
[1245.06 --> 1250.24] Just hearing kind of his story and how he kind of moved through building
|
| 863 |
+
[1250.24 --> 1250.72] hardware,
|
| 864 |
+
[1250.72 --> 1251.50] you know,
|
| 865 |
+
[1251.50 --> 1255.58] with cork boards and bread racks and DRAM swept off the floor that
|
| 866 |
+
[1255.58 --> 1258.68] software would try and correct when it failed and his,
|
| 867 |
+
[1258.84 --> 1263.94] his background and then kind of moving into this desire towards better
|
| 868 |
+
[1263.94 --> 1267.84] efficiency and systems and that kind of guiding a lot of his career and,
|
| 869 |
+
[1267.84 --> 1268.12] you know,
|
| 870 |
+
[1268.12 --> 1269.44] starting OCP and,
|
| 871 |
+
[1269.62 --> 1269.72] you know,
|
| 872 |
+
[1269.74 --> 1271.14] being one of the founding members of OCP.
|
| 873 |
+
[1271.14 --> 1272.54] And what was really,
|
| 874 |
+
[1272.70 --> 1273.72] really interesting was,
|
| 875 |
+
[1273.90 --> 1274.54] you know,
|
| 876 |
+
[1274.54 --> 1275.94] a lot of folks,
|
| 877 |
+
[1275.94 --> 1280.34] at least in the industry and the commercial world are after performance
|
| 878 |
+
[1280.34 --> 1283.52] and efficiency for commercial reasons or deliver a faster system to
|
| 879 |
+
[1283.52 --> 1283.92] customers.
|
| 880 |
+
[1283.92 --> 1285.68] And I want to lower cost.
|
| 881 |
+
[1285.68 --> 1289.08] And I think one of the kind of the core drivers for him that,
|
| 882 |
+
[1289.22 --> 1291.52] that really resonated was how do we,
|
| 883 |
+
[1291.68 --> 1294.72] how do we become more efficiency and more efficient in the planet?
|
| 884 |
+
[1294.82 --> 1296.26] Like how do we actually reduce emissions?
|
| 885 |
+
[1296.26 --> 1300.82] How do we start being less wasteful in terms of all these machines that
|
| 886 |
+
[1300.82 --> 1301.56] are inefficient,
|
| 887 |
+
[1301.74 --> 1302.32] underutilized,
|
| 888 |
+
[1302.44 --> 1302.74] et cetera.
|
| 889 |
+
[1302.74 --> 1308.70] And that kind of driving his startup and his passion and what he really cares
|
| 890 |
+
[1308.70 --> 1309.04] about,
|
| 891 |
+
[1309.54 --> 1311.40] less about the commercial implications of these things.
|
| 892 |
+
[1311.58 --> 1311.72] Yeah.
|
| 893 |
+
[1311.78 --> 1312.26] And I mean,
|
| 894 |
+
[1312.32 --> 1315.12] you could really tell that this is something he felt very earnestly.
|
| 895 |
+
[1315.34 --> 1315.54] Yeah.
|
| 896 |
+
[1315.54 --> 1316.98] This is not something that,
|
| 897 |
+
[1317.06 --> 1320.34] that he was doing for how you might perceive him or what has,
|
| 898 |
+
[1320.44 --> 1320.54] right.
|
| 899 |
+
[1320.54 --> 1322.78] There's something that he really felt,
|
| 900 |
+
[1322.78 --> 1324.02] which I thought was it.
|
| 901 |
+
[1324.06 --> 1324.16] Yeah.
|
| 902 |
+
[1324.16 --> 1324.82] That was interesting.
|
| 903 |
+
[1324.94 --> 1326.02] Cause that was a bit surprising.
|
| 904 |
+
[1326.02 --> 1326.36] I felt.
|
| 905 |
+
[1326.44 --> 1326.58] Yeah.
|
| 906 |
+
[1326.92 --> 1327.08] Well,
|
| 907 |
+
[1327.08 --> 1327.38] you're just,
|
| 908 |
+
[1327.50 --> 1331.24] you're used to the drivers being curiosity about something or a
|
| 909 |
+
[1331.24 --> 1331.98] commercial interest.
|
| 910 |
+
[1332.08 --> 1332.94] And also because I mean,
|
| 911 |
+
[1332.98 --> 1335.60] power efficiency has got such a clear economic driver as well.
|
| 912 |
+
[1335.68 --> 1335.94] That's right.
|
| 913 |
+
[1336.22 --> 1336.58] It's like,
|
| 914 |
+
[1336.58 --> 1339.02] it's very easy to just be doing it for business reasons,
|
| 915 |
+
[1339.02 --> 1339.28] but yeah,
|
| 916 |
+
[1339.28 --> 1339.82] he really wasn't.
|
| 917 |
+
[1339.88 --> 1340.92] He was doing it for,
|
| 918 |
+
[1340.92 --> 1344.96] and I thought similarly in terms of like the driving of OCP being to
|
| 919 |
+
[1344.96 --> 1347.42] share that innovation with more people.
|
| 920 |
+
[1347.42 --> 1348.68] So more people would adopt them.
|
| 921 |
+
[1348.96 --> 1350.50] So we would have more efficient computing.
|
| 922 |
+
[1350.62 --> 1350.76] Yeah.
|
| 923 |
+
[1351.06 --> 1353.52] And not have it locked up within a couple of very large,
|
| 924 |
+
[1353.68 --> 1354.96] very large providers.
|
| 925 |
+
[1355.10 --> 1355.40] Yeah.
|
| 926 |
+
[1355.56 --> 1355.74] That,
|
| 927 |
+
[1355.82 --> 1359.44] that infrastructure privilege that Jess maybe didn't,
|
| 928 |
+
[1359.52 --> 1360.04] maybe coined,
|
| 929 |
+
[1360.14 --> 1360.74] maybe didn't coin,
|
| 930 |
+
[1360.94 --> 1362.28] but I feel just definitely.
|
| 931 |
+
[1362.38 --> 1363.46] I think so.
|
| 932 |
+
[1363.52 --> 1364.98] I think we would only say like,
|
| 933 |
+
[1365.04 --> 1366.76] maybe didn't because I got to tell you,
|
| 934 |
+
[1366.78 --> 1370.06] I was giving a talk last night at Stanford and mentioned infrastructure
|
| 935 |
+
[1370.06 --> 1371.12] privilege and attributed it.
|
| 936 |
+
[1371.50 --> 1374.68] And a Googler who was there got into giggle fits.
|
| 937 |
+
[1374.88 --> 1375.22] I mean,
|
| 938 |
+
[1375.22 --> 1379.56] he was laughing so hard over infrastructure privilege.
|
| 939 |
+
[1379.60 --> 1379.90] And I'm like,
|
| 940 |
+
[1380.24 --> 1380.36] Hey,
|
| 941 |
+
[1380.38 --> 1380.58] look,
|
| 942 |
+
[1380.64 --> 1381.22] like it's,
|
| 943 |
+
[1381.36 --> 1381.62] you know,
|
| 944 |
+
[1381.86 --> 1384.20] the rest of us are actually living in a slum over here.
|
| 945 |
+
[1384.26 --> 1384.44] I mean,
|
| 946 |
+
[1384.44 --> 1384.72] it's like,
|
| 947 |
+
[1384.72 --> 1388.62] like don't hurt yourself laughing with the infrastructure privilege.
|
| 948 |
+
[1388.80 --> 1389.56] Can you go on the hallway?
|
| 949 |
+
[1390.18 --> 1390.58] Exactly.
|
| 950 |
+
[1390.76 --> 1391.60] You may want to like,
|
| 951 |
+
[1391.74 --> 1393.22] but so no,
|
| 952 |
+
[1393.22 --> 1393.46] I think,
|
| 953 |
+
[1393.46 --> 1395.24] I think Jess definitely coined infrastructure privilege.
|
| 954 |
+
[1395.38 --> 1396.36] I'm going with it.
|
| 955 |
+
[1396.54 --> 1396.88] I think so.
|
| 956 |
+
[1397.50 --> 1400.04] So I had another favorite moment just because it took,
|
| 957 |
+
[1400.06 --> 1402.22] me back to a particular time in place.
|
| 958 |
+
[1402.22 --> 1402.68] Um,
|
| 959 |
+
[1402.68 --> 1408.28] when Jonathan blow was describing porting doom to the SGI set top box.
|
| 960 |
+
[1408.86 --> 1409.30] And,
|
| 961 |
+
[1409.42 --> 1409.68] you know,
|
| 962 |
+
[1409.68 --> 1410.62] it's one of those things that like,
|
| 963 |
+
[1411.00 --> 1412.00] it wasn't.
|
| 964 |
+
[1412.00 --> 1413.00] And I think even for Jonathan,
|
| 965 |
+
[1413.00 --> 1414.36] it was not like a,
|
| 966 |
+
[1414.36 --> 1414.66] a,
|
| 967 |
+
[1414.66 --> 1414.92] you know,
|
| 968 |
+
[1414.92 --> 1416.08] a central thing,
|
| 969 |
+
[1416.08 --> 1418.78] but because I had done the same thing,
|
| 970 |
+
[1418.78 --> 1421.16] I'd ported doom to this other operating system at the same time.
|
| 971 |
+
[1421.16 --> 1425.02] It just like immediately transported me back to the,
|
| 972 |
+
[1425.02 --> 1426.00] the mid nineties.
|
| 973 |
+
[1426.00 --> 1427.38] And that project too,
|
| 974 |
+
[1427.46 --> 1432.06] which was a technically interesting and a very ahead of its time project,
|
| 975 |
+
[1432.06 --> 1435.38] but ultimately like didn't succeed as originally envisioned.
|
| 976 |
+
[1435.38 --> 1440.10] But I remember that thinking that the whole idea of video on demand felt
|
| 977 |
+
[1440.10 --> 1443.50] ludicrous and impossible because the last mile problem.
|
| 978 |
+
[1443.50 --> 1448.30] And yet that ultimately we obviously all have video on demand.
|
| 979 |
+
[1448.30 --> 1449.04] And so anyway,
|
| 980 |
+
[1449.30 --> 1451.62] Jonathan's description of that and working on that project and,
|
| 981 |
+
[1451.96 --> 1452.00] you know,
|
| 982 |
+
[1452.00 --> 1453.36] the engineers leaving to go to,
|
| 983 |
+
[1453.48 --> 1455.56] to Nvidia and kind of being in free fall.
|
| 984 |
+
[1455.56 --> 1456.02] It just,
|
| 985 |
+
[1456.10 --> 1458.10] and then going and starting a game problem,
|
| 986 |
+
[1458.18 --> 1460.84] a game company at the worst time to start a game company.
|
| 987 |
+
[1461.42 --> 1461.78] Uh,
|
| 988 |
+
[1461.78 --> 1462.04] I just thought,
|
| 989 |
+
[1462.14 --> 1462.28] again,
|
| 990 |
+
[1462.30 --> 1465.38] very evocative and taking you right back to it,
|
| 991 |
+
[1465.42 --> 1465.86] to a time.
|
| 992 |
+
[1465.90 --> 1466.70] We've all played games.
|
| 993 |
+
[1466.70 --> 1469.10] And I mean that he's been involved in some way,
|
| 994 |
+
[1469.12 --> 1470.78] shape and form and like everything we've,
|
| 995 |
+
[1471.06 --> 1471.26] you know,
|
| 996 |
+
[1471.30 --> 1472.78] he's touched your life at some point.
|
| 997 |
+
[1472.78 --> 1474.00] And well,
|
| 998 |
+
[1474.20 --> 1474.68] and he,
|
| 999 |
+
[1474.84 --> 1479.16] and I'm actually kind of amazed in retrospect that because,
|
| 1000 |
+
[1479.26 --> 1479.38] you know,
|
| 1001 |
+
[1479.38 --> 1480.32] Jess as like,
|
| 1002 |
+
[1480.46 --> 1481.70] just played the,
|
| 1003 |
+
[1481.74 --> 1481.84] you know,
|
| 1004 |
+
[1481.84 --> 1482.32] the witness.
|
| 1005 |
+
[1482.52 --> 1482.64] Yes.
|
| 1006 |
+
[1482.72 --> 1483.80] I played the witness.
|
| 1007 |
+
[1484.02 --> 1485.18] And I have like,
|
| 1008 |
+
[1485.38 --> 1485.64] no.
|
| 1009 |
+
[1485.74 --> 1486.42] So the witness,
|
| 1010 |
+
[1486.56 --> 1487.96] like it is like a,
|
| 1011 |
+
[1488.38 --> 1490.88] it's something where I play video games very fast,
|
| 1012 |
+
[1490.94 --> 1492.12] basically to get to the end,
|
| 1013 |
+
[1492.12 --> 1492.64] because I'm like,
|
| 1014 |
+
[1492.66 --> 1494.14] I just want to like play the game.
|
| 1015 |
+
[1494.38 --> 1497.48] And like one thing that I definitely like wish I would have brought up,
|
| 1016 |
+
[1497.50 --> 1498.24] but I just did it.
|
| 1017 |
+
[1498.28 --> 1499.18] Is that like,
|
| 1018 |
+
[1499.52 --> 1500.62] since I do that,
|
| 1019 |
+
[1500.66 --> 1502.76] like I don't enjoy the game while I'm playing it.
|
| 1020 |
+
[1503.06 --> 1504.00] It's way too competitive.
|
| 1021 |
+
[1504.14 --> 1505.90] So then like his game,
|
| 1022 |
+
[1506.18 --> 1507.48] that's just to get to the last chapter.
|
| 1023 |
+
[1507.48 --> 1507.92] It's like,
|
| 1024 |
+
[1507.98 --> 1508.22] yeah,
|
| 1025 |
+
[1508.28 --> 1509.68] somehow I can't envision that.
|
| 1026 |
+
[1509.76 --> 1510.74] I can't envision that.
|
| 1027 |
+
[1510.98 --> 1512.94] You unlock all these things along the way.
|
| 1028 |
+
[1512.96 --> 1513.28] And I'm like,
|
| 1029 |
+
[1513.30 --> 1513.50] okay,
|
| 1030 |
+
[1513.50 --> 1513.94] it's fine.
|
| 1031 |
+
[1514.02 --> 1516.02] Like I'll go back when I'm done and like,
|
| 1032 |
+
[1516.04 --> 1516.16] go,
|
| 1033 |
+
[1516.50 --> 1518.34] go check out all the worlds and stuff.
|
| 1034 |
+
[1518.60 --> 1521.18] And so then when you're done with the game,
|
| 1035 |
+
[1521.18 --> 1523.38] it pops you back to the beginning.
|
| 1036 |
+
[1523.98 --> 1525.82] And so you weren't able to go back.
|
| 1037 |
+
[1525.92 --> 1527.48] Like I'd have to play the whole game again.
|
| 1038 |
+
[1527.66 --> 1529.54] You would have had to have enjoyed it along the way.
|
| 1039 |
+
[1529.64 --> 1529.76] Well,
|
| 1040 |
+
[1529.78 --> 1530.36] but the other thing.
|
| 1041 |
+
[1530.36 --> 1535.12] I think there's a hidden lesson there.
|
| 1042 |
+
[1536.00 --> 1536.36] Well,
|
| 1043 |
+
[1536.40 --> 1537.38] it's the other thing about it.
|
| 1044 |
+
[1537.38 --> 1539.04] This game is there is,
|
| 1045 |
+
[1539.10 --> 1539.48] it's not really,
|
| 1046 |
+
[1539.62 --> 1541.00] I don't think it's really an Easter egg,
|
| 1047 |
+
[1541.06 --> 1541.14] right?
|
| 1048 |
+
[1541.14 --> 1542.36] The videos that are in it.
|
| 1049 |
+
[1542.44 --> 1543.32] Definitely not an Easter egg.
|
| 1050 |
+
[1543.36 --> 1546.42] Like you're supposed to kind of collect all these like little puzzles,
|
| 1051 |
+
[1546.54 --> 1547.94] but it's definitely not part of winning.
|
| 1052 |
+
[1548.04 --> 1552.84] Like you don't have to collect all these like little puzzles to unlock the YouTube videos to win.
|
| 1053 |
+
[1552.84 --> 1553.44] But like,
|
| 1054 |
+
[1553.74 --> 1554.52] once you feel,
|
| 1055 |
+
[1554.68 --> 1555.34] once you get one,
|
| 1056 |
+
[1555.40 --> 1555.64] you're like,
|
| 1057 |
+
[1555.70 --> 1555.96] oh shit,
|
| 1058 |
+
[1555.98 --> 1557.02] I got to get like all the others.
|
| 1059 |
+
[1557.22 --> 1558.52] But suffice it to say,
|
| 1060 |
+
[1558.60 --> 1561.52] these videos rocked Jess's boat a little bit.
|
| 1061 |
+
[1561.52 --> 1562.52] So like,
|
| 1062 |
+
[1562.64 --> 1565.78] because I play video games so fast,
|
| 1063 |
+
[1565.94 --> 1568.16] like I am very much like in the game.
|
| 1064 |
+
[1568.36 --> 1569.44] Like I am like,
|
| 1065 |
+
[1569.72 --> 1571.48] you have to like tear me away.
|
| 1066 |
+
[1571.84 --> 1574.36] This is also why I didn't bring all this up because it's like,
|
| 1067 |
+
[1574.36 --> 1574.58] okay,
|
| 1068 |
+
[1574.66 --> 1575.74] comes off a little crazy.
|
| 1069 |
+
[1578.00 --> 1579.62] So like the video,
|
| 1070 |
+
[1579.74 --> 1580.70] like the YouTubes,
|
| 1071 |
+
[1580.76 --> 1585.40] like some of them are related and some of them are like physics professors and stuff like that.
|
| 1072 |
+
[1585.40 --> 1585.84] And like,
|
| 1073 |
+
[1586.48 --> 1588.46] like Richard Feynman is in one of the videos.
|
| 1074 |
+
[1588.46 --> 1589.08] And then there's,
|
| 1075 |
+
[1589.32 --> 1589.48] yeah,
|
| 1076 |
+
[1589.64 --> 1591.30] like most of them are physics related,
|
| 1077 |
+
[1591.44 --> 1591.74] honestly.
|
| 1078 |
+
[1591.92 --> 1598.46] And then there's a couple like just curve balls in there that are just so weird that you want to tie it all together.
|
| 1079 |
+
[1599.00 --> 1599.98] And like it,
|
| 1080 |
+
[1600.10 --> 1600.68] it just,
|
| 1081 |
+
[1600.76 --> 1602.84] it creates a weird thing when you do that.
|
| 1082 |
+
[1602.94 --> 1603.44] It's all,
|
| 1083 |
+
[1603.58 --> 1603.84] and,
|
| 1084 |
+
[1604.36 --> 1604.90] I mean,
|
| 1085 |
+
[1604.96 --> 1606.88] so you were having me watch these videos.
|
| 1086 |
+
[1606.98 --> 1608.36] It must've been after the conversation with John.
|
| 1087 |
+
[1608.36 --> 1609.68] It was definitely after because I was like,
|
| 1088 |
+
[1609.72 --> 1611.84] I don't know if I should bring it up or not.
|
| 1089 |
+
[1612.00 --> 1612.12] Right.
|
| 1090 |
+
[1612.16 --> 1612.44] And then,
|
| 1091 |
+
[1612.46 --> 1613.92] so you were showing it to me and I'm like,
|
| 1092 |
+
[1614.02 --> 1614.36] God,
|
| 1093 |
+
[1614.42 --> 1614.80] you have these videos.
|
| 1094 |
+
[1614.88 --> 1615.10] And then,
|
| 1095 |
+
[1615.16 --> 1619.54] but it was clear that these videos were transporting you back to a time and place.
|
| 1096 |
+
[1619.64 --> 1620.76] And then we play this game and you're just like,
|
| 1097 |
+
[1620.84 --> 1621.66] what does the,
|
| 1098 |
+
[1621.78 --> 1622.90] what does the video mean?
|
| 1099 |
+
[1623.18 --> 1624.04] What does it mean?
|
| 1100 |
+
[1624.08 --> 1624.88] It means something,
|
| 1101 |
+
[1624.98 --> 1625.66] but I don't understand.
|
| 1102 |
+
[1625.86 --> 1626.06] It's like,
|
| 1103 |
+
[1626.06 --> 1627.88] even the guy in the YouTube comments was like,
|
| 1104 |
+
[1627.94 --> 1629.62] what do these videos mean?
|
| 1105 |
+
[1629.74 --> 1630.08] And I'm like,
|
| 1106 |
+
[1630.14 --> 1630.70] that is me.
|
| 1107 |
+
[1631.12 --> 1631.48] Right.
|
| 1108 |
+
[1631.66 --> 1632.22] And some of them,
|
| 1109 |
+
[1632.44 --> 1633.30] because the other one was like,
|
| 1110 |
+
[1633.30 --> 1634.50] some of them have a clear meaning.
|
| 1111 |
+
[1634.50 --> 1636.40] So now that you want to get clear meaning out of all of them,
|
| 1112 |
+
[1636.62 --> 1641.68] one of the things that's interesting is one of those videos is by Brian Moriarty,
|
| 1113 |
+
[1642.10 --> 1644.38] who was the Infocom who did Trinity,
|
| 1114 |
+
[1644.50 --> 1645.32] the game that he mentioned,
|
| 1115 |
+
[1645.46 --> 1646.82] the Infocom game that he mentioned.
|
| 1116 |
+
[1647.22 --> 1648.04] So it was all,
|
| 1117 |
+
[1648.14 --> 1650.96] as I was like doing the show notes and I'm like,
|
| 1118 |
+
[1651.04 --> 1651.22] Ooh,
|
| 1119 |
+
[1651.22 --> 1652.58] this is one of the videos like,
|
| 1120 |
+
[1652.70 --> 1652.86] Ooh,
|
| 1121 |
+
[1652.86 --> 1653.88] should I bring this up with Jess?
|
| 1122 |
+
[1653.88 --> 1654.94] Like bit of a landmine.
|
| 1123 |
+
[1654.94 --> 1659.42] Maybe it's just the one with the candle.
|
| 1124 |
+
[1659.50 --> 1660.26] That makes no sense.
|
| 1125 |
+
[1660.72 --> 1661.48] In your defense,
|
| 1126 |
+
[1661.48 --> 1663.12] the one with the candle really does not make sense.
|
| 1127 |
+
[1663.40 --> 1665.52] It's like a Polish art film or something.
|
| 1128 |
+
[1665.64 --> 1665.84] I don't even,
|
| 1129 |
+
[1665.84 --> 1667.98] that one's thrown in there just to mess with people.
|
| 1130 |
+
[1668.08 --> 1668.44] I don't know.
|
| 1131 |
+
[1668.82 --> 1670.24] Especially people playing the game fast.
|
| 1132 |
+
[1670.34 --> 1670.62] Yeah.
|
| 1133 |
+
[1670.96 --> 1671.30] I mean,
|
| 1134 |
+
[1671.30 --> 1672.48] if that was its intent,
|
| 1135 |
+
[1672.58 --> 1673.68] like mission accomplished.
|
| 1136 |
+
[1673.68 --> 1674.38] It really did.
|
| 1137 |
+
[1674.38 --> 1676.22] Like it really did.
|
| 1138 |
+
[1676.62 --> 1676.74] I,
|
| 1139 |
+
[1676.90 --> 1678.22] you almost wonder if you brought it up with Jonathan,
|
| 1140 |
+
[1678.32 --> 1678.48] Jonathan,
|
| 1141 |
+
[1678.56 --> 1680.60] like I kind of overshot the mark on this one.
|
| 1142 |
+
[1681.36 --> 1682.76] I've been waiting for someone to bring that one up.
|
| 1143 |
+
[1682.76 --> 1684.46] Like I was trying to like mess with you,
|
| 1144 |
+
[1684.54 --> 1685.58] not screw you up,
|
| 1145 |
+
[1685.66 --> 1686.96] but I think I may have.
|
| 1146 |
+
[1687.02 --> 1687.12] No,
|
| 1147 |
+
[1687.16 --> 1687.54] but I was,
|
| 1148 |
+
[1687.66 --> 1690.12] I'm amazed in hindsight that you did not actually.
|
| 1149 |
+
[1690.42 --> 1691.56] I was thinking about it a lot though,
|
| 1150 |
+
[1691.56 --> 1694.86] but then I did imagine him like going back to his like office and being like,
|
| 1151 |
+
[1694.92 --> 1698.34] you will not believe what this girl on the podcast said about the videos.
|
| 1152 |
+
[1698.34 --> 1698.78] Like,
|
| 1153 |
+
[1698.96 --> 1699.78] yeah,
|
| 1154 |
+
[1699.84 --> 1700.94] I don't think I,
|
| 1155 |
+
[1701.04 --> 1703.46] and I was just impressed that because underneath the,
|
| 1156 |
+
[1703.60 --> 1703.76] you know,
|
| 1157 |
+
[1703.82 --> 1707.48] that your relative still waters during the conversation must have just been this,
|
| 1158 |
+
[1707.66 --> 1709.22] like the candle,
|
| 1159 |
+
[1709.36 --> 1710.22] the candle,
|
| 1160 |
+
[1710.44 --> 1711.80] what is going on to the candle?
|
| 1161 |
+
[1712.54 --> 1713.62] But it was amazing.
|
| 1162 |
+
[1713.70 --> 1714.88] Now we just,
|
| 1163 |
+
[1714.94 --> 1715.88] have we exhausted your list?
|
| 1164 |
+
[1716.04 --> 1716.14] No,
|
| 1165 |
+
[1716.18 --> 1716.60] I have one more.
|
| 1166 |
+
[1716.68 --> 1716.82] Okay.
|
| 1167 |
+
[1716.84 --> 1717.18] Go for it.
|
| 1168 |
+
[1717.64 --> 1717.90] Okay.
|
| 1169 |
+
[1718.00 --> 1721.80] So my last one is during John masters,
|
| 1170 |
+
[1721.80 --> 1725.98] when we were all reminiscing about where we were during specter and meltdown,
|
| 1171 |
+
[1725.98 --> 1727.32] which like is crazy.
|
| 1172 |
+
[1727.32 --> 1729.00] Cause it was such a huge thing.
|
| 1173 |
+
[1729.00 --> 1730.66] And it was funny to see like everyone else.
|
| 1174 |
+
[1730.66 --> 1733.24] Cause I thought I was like freaking out like where I was,
|
| 1175 |
+
[1733.28 --> 1735.14] but like everyone else was freaking out where they were.
|
| 1176 |
+
[1735.18 --> 1735.54] Yeah.
|
| 1177 |
+
[1736.10 --> 1736.40] That,
|
| 1178 |
+
[1736.50 --> 1737.48] that was funny.
|
| 1179 |
+
[1737.50 --> 1738.72] And I loved your story too,
|
| 1180 |
+
[1738.72 --> 1739.96] about like your,
|
| 1181 |
+
[1740.12 --> 1742.66] your sister is going like shopping for her wedding dress.
|
| 1182 |
+
[1742.66 --> 1743.08] And you're like,
|
| 1183 |
+
[1743.14 --> 1743.76] Oh my God,
|
| 1184 |
+
[1743.82 --> 1745.16] do you not appreciate what's happening?
|
| 1185 |
+
[1745.56 --> 1747.02] The world is ending.
|
| 1186 |
+
[1747.04 --> 1747.92] Look at my phone.
|
| 1187 |
+
[1748.04 --> 1749.22] And I can just admit her to be like,
|
| 1188 |
+
[1749.34 --> 1749.74] can you,
|
| 1189 |
+
[1749.86 --> 1751.26] can I not have my day?
|
| 1190 |
+
[1751.44 --> 1751.80] Like,
|
| 1191 |
+
[1751.88 --> 1754.64] why must it always be about you?
|
| 1192 |
+
[1754.78 --> 1755.14] And you're like,
|
| 1193 |
+
[1755.26 --> 1755.48] no,
|
| 1194 |
+
[1755.62 --> 1756.52] this is like crisis.
|
| 1195 |
+
[1756.52 --> 1756.96] Yeah.
|
| 1196 |
+
[1757.12 --> 1758.92] All of the computers everywhere are broken.
|
| 1197 |
+
[1759.04 --> 1759.96] I was like freaking out.
|
| 1198 |
+
[1761.38 --> 1761.70] Right.
|
| 1199 |
+
[1761.82 --> 1762.04] No,
|
| 1200 |
+
[1762.10 --> 1762.34] I just,
|
| 1201 |
+
[1762.40 --> 1763.54] I felt like I definitely,
|
| 1202 |
+
[1763.64 --> 1765.26] and I remember simply where I was.
|
| 1203 |
+
[1765.38 --> 1765.54] Yeah.
|
| 1204 |
+
[1766.26 --> 1766.84] What a,
|
| 1205 |
+
[1766.92 --> 1767.88] what a mess.
|
| 1206 |
+
[1768.04 --> 1768.62] But that was,
|
| 1207 |
+
[1768.70 --> 1771.10] that was not a good day to be in the cloud provider business.
|
| 1208 |
+
[1771.24 --> 1772.00] It was not a good day.
|
| 1209 |
+
[1772.28 --> 1772.92] It was not a good day.
|
| 1210 |
+
[1773.12 --> 1773.44] Yeah.
|
| 1211 |
+
[1773.54 --> 1774.66] That was not a good week.
|
| 1212 |
+
[1774.86 --> 1774.90] It's like,
|
| 1213 |
+
[1774.92 --> 1775.08] wait,
|
| 1214 |
+
[1775.20 --> 1776.46] do we run a multi-tenant service?
|
| 1215 |
+
[1776.60 --> 1776.80] Right.
|
| 1216 |
+
[1776.86 --> 1777.18] Exactly.
|
| 1217 |
+
[1777.24 --> 1778.02] It was not a good month.
|
| 1218 |
+
[1778.08 --> 1778.66] It was not a good,
|
| 1219 |
+
[1778.88 --> 1779.58] but no,
|
| 1220 |
+
[1779.62 --> 1779.96] that was,
|
| 1221 |
+
[1780.08 --> 1780.66] that was fun.
|
| 1222 |
+
[1780.66 --> 1782.80] And especially that because for,
|
| 1223 |
+
[1782.94 --> 1784.36] because recall for John Masters,
|
| 1224 |
+
[1785.06 --> 1786.24] one of our founding engineers,
|
| 1225 |
+
[1786.34 --> 1787.74] Robert Moustaki was here.
|
| 1226 |
+
[1787.80 --> 1787.98] Yeah.
|
| 1227 |
+
[1788.18 --> 1789.98] Like watching the podcast live.
|
| 1228 |
+
[1789.98 --> 1795.60] And Robert was the one who implemented Colonel Page Table Isolation for us while we were exposed.
|
| 1229 |
+
[1795.60 --> 1802.42] So Robert Moustaki and Alex Wilson were actually having to do brain surgery while we were exposed.
|
| 1230 |
+
[1802.42 --> 1803.14] So that was,
|
| 1231 |
+
[1803.26 --> 1803.52] I mean,
|
| 1232 |
+
[1803.54 --> 1805.18] I know it's evocative for him as well,
|
| 1233 |
+
[1805.22 --> 1805.72] for sure.
|
| 1234 |
+
[1806.26 --> 1807.06] That was just amazing.
|
| 1235 |
+
[1807.14 --> 1808.80] And all the stories of the vulnerabilities were,
|
| 1236 |
+
[1808.90 --> 1809.70] were amazing.
|
| 1237 |
+
[1810.32 --> 1810.76] And,
|
| 1238 |
+
[1810.88 --> 1811.46] and yeah,
|
| 1239 |
+
[1811.50 --> 1811.80] exactly.
|
| 1240 |
+
[1811.86 --> 1812.44] It takes us back.
|
| 1241 |
+
[1812.54 --> 1812.70] It like,
|
| 1242 |
+
[1812.78 --> 1812.92] yeah,
|
| 1243 |
+
[1813.02 --> 1816.94] I was like back there with my sister trying on wedding dresses and her being like,
|
| 1244 |
+
[1817.16 --> 1818.00] what the fuck?
|
| 1245 |
+
[1819.60 --> 1820.04] Right.
|
| 1246 |
+
[1820.32 --> 1821.72] Must it be about you?
|
| 1247 |
+
[1822.18 --> 1822.40] It's like,
|
| 1248 |
+
[1822.44 --> 1823.38] it's not about me.
|
| 1249 |
+
[1823.66 --> 1824.66] It's the world.
|
| 1250 |
+
[1824.66 --> 1825.78] It's humanity.
|
| 1251 |
+
[1827.62 --> 1827.98] Steve,
|
| 1252 |
+
[1828.06 --> 1829.04] any more on your list?
|
| 1253 |
+
[1829.38 --> 1830.38] It was actually,
|
| 1254 |
+
[1830.44 --> 1832.08] that was kind of a partial mine.
|
| 1255 |
+
[1832.18 --> 1832.48] It was,
|
| 1256 |
+
[1832.64 --> 1840.16] it was around the vulnerabilities stories from John Masters and then the naming of them and kind of how the names were derived and the branding and all that.
|
| 1257 |
+
[1840.16 --> 1844.92] It was just hearing all the terrifying stories of vulnerabilities out there.
|
| 1258 |
+
[1845.10 --> 1846.04] Very interesting in that one.
|
| 1259 |
+
[1846.86 --> 1848.18] So my last one,
|
| 1260 |
+
[1848.24 --> 1848.36] I mean,
|
| 1261 |
+
[1848.38 --> 1848.54] I've got,
|
| 1262 |
+
[1848.60 --> 1849.08] he's got a lot,
|
| 1263 |
+
[1849.20 --> 1849.34] we could,
|
| 1264 |
+
[1849.68 --> 1850.40] but Rick,
|
| 1265 |
+
[1850.56 --> 1854.64] I loved Rick when we were talking about Windows NT back in the day.
|
| 1266 |
+
[1855.66 --> 1862.66] And he was able to recall that NT ran on MIPS by bringing up the folder in his brain.
|
| 1267 |
+
[1862.76 --> 1865.72] You could see he was visualizing the folder of the drive.
|
| 1268 |
+
[1865.82 --> 1865.96] He's like,
|
| 1269 |
+
[1866.06 --> 1866.56] and as he was,
|
| 1270 |
+
[1866.64 --> 1866.82] he's like,
|
| 1271 |
+
[1866.84 --> 1867.48] he was on alpha.
|
| 1272 |
+
[1867.64 --> 1868.46] And then it was on,
|
| 1273 |
+
[1868.86 --> 1872.28] because I just feel like that Windows NT ran on MIPS is a fact.
|
| 1274 |
+
[1872.28 --> 1875.32] Not many people know or would retain.
|
| 1275 |
+
[1875.94 --> 1876.88] And again,
|
| 1276 |
+
[1876.94 --> 1880.42] very evocative for me because I had the top 10 reasons to run NT on MIPS,
|
| 1277 |
+
[1880.54 --> 1881.74] but it was just amazing.
|
| 1278 |
+
[1881.90 --> 1883.66] I thought there's a whole Wikipedia page.
|
| 1279 |
+
[1883.66 --> 1884.16] I found,
|
| 1280 |
+
[1884.32 --> 1886.22] I think on Windows NT on MIPS,
|
| 1281 |
+
[1886.32 --> 1889.44] or there is a specific website on it because I linked to it in the show notes.
|
| 1282 |
+
[1889.46 --> 1889.76] And I was like,
|
| 1283 |
+
[1889.82 --> 1890.52] this was a thing.
|
| 1284 |
+
[1890.62 --> 1895.74] And I think one of the things I really appreciated actually in this whole exercise is Wikipedia.
|
| 1285 |
+
[1896.02 --> 1896.34] Cause Jess,
|
| 1286 |
+
[1896.40 --> 1897.78] you and I went to Wikipedia all the time.
|
| 1287 |
+
[1897.86 --> 1899.06] There's good stuff on there.
|
| 1288 |
+
[1899.16 --> 1900.38] There is good stuff on there.
|
| 1289 |
+
[1900.38 --> 1902.78] And so many of these things are,
|
| 1290 |
+
[1903.00 --> 1905.62] these relatively are very obscure artifacts.
|
| 1291 |
+
[1905.88 --> 1908.38] I've got actually really good Wikipedia pages.
|
| 1292 |
+
[1908.62 --> 1913.18] Plus like the other thing that I found through doing the show notes is it's not just the Wikipedia page that's great.
|
| 1293 |
+
[1913.28 --> 1915.26] There's like this references thing at the bottom.
|
| 1294 |
+
[1915.26 --> 1915.76] And like,
|
| 1295 |
+
[1915.78 --> 1917.00] if you click through those,
|
| 1296 |
+
[1917.04 --> 1918.64] then you find out way more information.
|
| 1297 |
+
[1918.92 --> 1918.94] Yeah.
|
| 1298 |
+
[1919.06 --> 1919.82] Way more information.
|
| 1299 |
+
[1919.82 --> 1921.90] I could have done the show notes with Wikipedia.
|
| 1300 |
+
[1922.06 --> 1922.52] I think it's,
|
| 1301 |
+
[1922.66 --> 1924.20] I think I'm taking one on at some point.
|
| 1302 |
+
[1924.80 --> 1925.16] Oh,
|
| 1303 |
+
[1925.96 --> 1926.40] God.
|
| 1304 |
+
[1926.70 --> 1927.18] Come on.
|
| 1305 |
+
[1927.80 --> 1929.54] I thought next season's pop culture,
|
| 1306 |
+
[1929.62 --> 1929.80] right?
|
| 1307 |
+
[1929.80 --> 1930.40] Yeah,
|
| 1308 |
+
[1930.40 --> 1930.84] exactly.
|
| 1309 |
+
[1931.84 --> 1932.24] Yeah.
|
| 1310 |
+
[1932.32 --> 1933.20] So any,
|
| 1311 |
+
[1933.34 --> 1934.38] any lowlights?
|
| 1312 |
+
[1934.98 --> 1935.18] No,
|
| 1313 |
+
[1935.36 --> 1935.56] no,
|
| 1314 |
+
[1935.64 --> 1935.82] no,
|
| 1315 |
+
[1935.90 --> 1935.98] no.
|
| 1316 |
+
[1937.04 --> 1938.82] I can't not talk about it.
|
| 1317 |
+
[1940.38 --> 1944.36] The bumpkin that gave us the three star review.
|
| 1318 |
+
[1944.54 --> 1944.86] Oh,
|
| 1319 |
+
[1944.86 --> 1945.44] come on.
|
| 1320 |
+
[1945.44 --> 1946.70] We're going to talk about this.
|
| 1321 |
+
[1947.62 --> 1949.44] We have to hear all five star reviews.
|
| 1322 |
+
[1949.54 --> 1951.56] There's only one three star review.
|
| 1323 |
+
[1951.56 --> 1951.84] Oh my God.
|
| 1324 |
+
[1951.92 --> 1952.16] Come on.
|
| 1325 |
+
[1952.28 --> 1952.40] No,
|
| 1326 |
+
[1952.40 --> 1952.56] no,
|
| 1327 |
+
[1952.56 --> 1952.90] no,
|
| 1328 |
+
[1952.90 --> 1953.20] no,
|
| 1329 |
+
[1953.20 --> 1953.50] no,
|
| 1330 |
+
[1953.50 --> 1953.66] no,
|
| 1331 |
+
[1953.66 --> 1954.28] get it back on track.
|
| 1332 |
+
[1954.36 --> 1955.16] I thought you were going to say like,
|
| 1333 |
+
[1955.24 --> 1955.42] Oh,
|
| 1334 |
+
[1955.46 --> 1956.06] and Steve said,
|
| 1335 |
+
[1956.22 --> 1956.60] no,
|
| 1336 |
+
[1956.72 --> 1957.06] no,
|
| 1337 |
+
[1957.06 --> 1957.56] instruction,
|
| 1338 |
+
[1957.66 --> 1958.28] instruction set.
|
| 1339 |
+
[1958.36 --> 1958.52] No,
|
| 1340 |
+
[1958.58 --> 1958.80] no,
|
| 1341 |
+
[1958.80 --> 1959.04] no,
|
| 1342 |
+
[1959.04 --> 1959.36] no,
|
| 1343 |
+
[1959.36 --> 1959.60] no.
|
| 1344 |
+
[1959.60 --> 1960.28] I was really,
|
| 1345 |
+
[1960.44 --> 1960.68] I know.
|
| 1346 |
+
[1960.72 --> 1962.06] I was trying to set myself up.
|
| 1347 |
+
[1962.14 --> 1963.24] I was trying to set myself up.
|
| 1348 |
+
[1963.34 --> 1964.44] I just want to get it out of myself.
|
| 1349 |
+
[1964.66 --> 1965.04] All right.
|
| 1350 |
+
[1965.16 --> 1965.52] I just,
|
| 1351 |
+
[1965.62 --> 1965.74] can I,
|
| 1352 |
+
[1965.82 --> 1966.80] let me just get it out of here.
|
| 1353 |
+
[1966.90 --> 1967.82] Let me just get it out of here.
|
| 1354 |
+
[1968.00 --> 1968.76] We'll be done.
|
| 1355 |
+
[1968.88 --> 1969.10] Okay.
|
| 1356 |
+
[1969.50 --> 1970.08] So first of all,
|
| 1357 |
+
[1970.08 --> 1970.90] the podcast is great.
|
| 1358 |
+
[1970.90 --> 1971.40] And this,
|
| 1359 |
+
[1971.52 --> 1974.44] the one of these reviews and the reviews are obviously like,
|
| 1360 |
+
[1974.48 --> 1975.18] I love this podcast.
|
| 1361 |
+
[1975.18 --> 1975.48] Of course,
|
| 1362 |
+
[1975.50 --> 1976.40] like who wouldn't.
|
| 1363 |
+
[1976.46 --> 1977.42] And also like,
|
| 1364 |
+
[1977.46 --> 1979.64] I feel like we don't ask people to leave reviews,
|
| 1365 |
+
[1979.74 --> 1980.40] like leave a review.
|
| 1366 |
+
[1980.40 --> 1981.10] If you want to leave a review,
|
| 1367 |
+
[1981.18 --> 1981.34] I mean,
|
| 1368 |
+
[1981.38 --> 1982.64] people should be free to do what they want,
|
| 1369 |
+
[1982.90 --> 1984.82] except for the person who left,
|
| 1370 |
+
[1985.02 --> 1985.78] who said like,
|
| 1371 |
+
[1985.82 --> 1985.96] Oh,
|
| 1372 |
+
[1985.96 --> 1987.00] I like the podcast,
|
| 1373 |
+
[1987.00 --> 1988.44] but you know,
|
| 1374 |
+
[1988.54 --> 1991.18] Brian should be like asking just less frequently.
|
| 1375 |
+
[1991.18 --> 1993.38] If she's ever heard of something like that's getting a little old.
|
| 1376 |
+
[1993.42 --> 1994.28] That's my recommendation.
|
| 1377 |
+
[1994.92 --> 1995.72] Three stars.
|
| 1378 |
+
[1996.32 --> 1997.48] And first of all,
|
| 1379 |
+
[1997.80 --> 1998.72] I'm like,
|
| 1380 |
+
[1999.00 --> 1999.34] and for,
|
| 1381 |
+
[1999.40 --> 1999.50] yeah,
|
| 1382 |
+
[1999.58 --> 2000.70] just thought it was like hilarious.
|
| 1383 |
+
[2000.74 --> 2000.96] I'm like,
|
| 1384 |
+
[2000.98 --> 2001.96] this is not hilarious.
|
| 1385 |
+
[2002.74 --> 2004.96] This is not hilarious at all for,
|
| 1386 |
+
[2005.12 --> 2005.92] because one,
|
| 1387 |
+
[2006.04 --> 2006.36] I,
|
| 1388 |
+
[2006.48 --> 2007.58] I don't agree.
|
| 1389 |
+
[2007.92 --> 2008.36] I mean,
|
| 1390 |
+
[2008.70 --> 2009.32] you know,
|
| 1391 |
+
[2009.32 --> 2009.50] Jess,
|
| 1392 |
+
[2009.62 --> 2009.96] I am,
|
| 1393 |
+
[2010.04 --> 2010.54] I'm what?
|
| 1394 |
+
[2010.76 --> 2012.14] 16 years older than you.
|
| 1395 |
+
[2012.22 --> 2013.82] It's not an unreasonable question.
|
| 1396 |
+
[2014.26 --> 2015.48] And like on BBS is,
|
| 1397 |
+
[2015.54 --> 2015.66] okay.
|
| 1398 |
+
[2015.66 --> 2017.04] Like BBS,
|
| 1399 |
+
[2017.18 --> 2019.46] the last BBS was shut off when you were in utero.
|
| 1400 |
+
[2019.76 --> 2025.04] I only know BBS is through like internet culture history.
|
| 1401 |
+
[2025.22 --> 2025.34] Right.
|
| 1402 |
+
[2025.38 --> 2025.74] Exactly.
|
| 1403 |
+
[2025.84 --> 2026.04] Like,
|
| 1404 |
+
[2026.06 --> 2027.80] this is not an unreasonable question.
|
| 1405 |
+
[2027.80 --> 2028.90] If you've heard of a BBS,
|
| 1406 |
+
[2029.02 --> 2030.04] like I'm not being,
|
| 1407 |
+
[2030.36 --> 2031.90] I don't feel like I'm being,
|
| 1408 |
+
[2032.12 --> 2034.44] so I feel like I felt obviously very defensive.
|
| 1409 |
+
[2034.52 --> 2035.58] And then I feel like,
|
| 1410 |
+
[2035.58 --> 2036.54] I don't know.
|
| 1411 |
+
[2036.54 --> 2037.56] We're getting two stars.
|
| 1412 |
+
[2039.00 --> 2041.86] We could lose two stars because I'm asking if Jess never heard of a BBS.
|
| 1413 |
+
[2042.10 --> 2042.62] It could have been three.
|
| 1414 |
+
[2042.96 --> 2043.58] It could have been three.
|
| 1415 |
+
[2043.76 --> 2043.98] I know.
|
| 1416 |
+
[2044.04 --> 2044.56] That's what you're saying.
|
| 1417 |
+
[2044.66 --> 2044.94] I know.
|
| 1418 |
+
[2045.02 --> 2045.20] I know.
|
| 1419 |
+
[2045.20 --> 2046.06] I just need to be like,
|
| 1420 |
+
[2046.10 --> 2046.84] only knocked off too.
|
| 1421 |
+
[2046.88 --> 2047.02] Like,
|
| 1422 |
+
[2047.04 --> 2047.98] why am I reading the comments?
|
| 1423 |
+
[2048.12 --> 2048.90] Why are you in the comments?
|
| 1424 |
+
[2049.02 --> 2049.56] Because no,
|
| 1425 |
+
[2049.56 --> 2049.72] no,
|
| 1426 |
+
[2049.72 --> 2049.84] no,
|
| 1427 |
+
[2049.84 --> 2049.92] no,
|
| 1428 |
+
[2049.92 --> 2050.10] no.
|
| 1429 |
+
[2050.10 --> 2050.82] Actually in fairness,
|
| 1430 |
+
[2050.82 --> 2052.10] the comments were great.
|
| 1431 |
+
[2052.38 --> 2054.12] And you didn't read the comments.
|
| 1432 |
+
[2054.22 --> 2055.46] I read them and then I posted it.
|
| 1433 |
+
[2055.58 --> 2056.06] Cause I was laughing.
|
| 1434 |
+
[2056.06 --> 2056.74] The hard one,
|
| 1435 |
+
[2056.80 --> 2058.38] this is all my fault.
|
| 1436 |
+
[2058.38 --> 2059.22] It's all my fault.
|
| 1437 |
+
[2059.38 --> 2061.58] Technically you did actually send me that comment.
|
| 1438 |
+
[2061.86 --> 2062.52] That's true.
|
| 1439 |
+
[2062.70 --> 2063.20] That's true.
|
| 1440 |
+
[2063.40 --> 2063.72] That's true.
|
| 1441 |
+
[2063.76 --> 2065.10] But also because it is,
|
| 1442 |
+
[2065.22 --> 2068.92] the comments have been so positive just because it has been so positive.
|
| 1443 |
+
[2068.92 --> 2069.76] It's been so great.
|
| 1444 |
+
[2069.84 --> 2070.22] You know,
|
| 1445 |
+
[2070.28 --> 2072.12] it's been so much fun.
|
| 1446 |
+
[2072.12 --> 2072.22] Super,
|
| 1447 |
+
[2072.28 --> 2076.50] super grateful at everyone who takes the time to listen to the long form podcast.
|
| 1448 |
+
[2076.74 --> 2076.88] Yeah.
|
| 1449 |
+
[2077.00 --> 2078.32] And given feedback.
|
| 1450 |
+
[2078.44 --> 2079.24] It was great.
|
| 1451 |
+
[2079.24 --> 2082.40] We do have time with the ad.
|
| 1452 |
+
[2082.66 --> 2083.14] We had,
|
| 1453 |
+
[2083.26 --> 2083.62] uh,
|
| 1454 |
+
[2084.04 --> 2090.86] obviously and understandably listening to those same three ads again and again and again would be grating for anybody.
|
| 1455 |
+
[2091.02 --> 2091.32] Hey,
|
| 1456 |
+
[2091.56 --> 2091.82] uh,
|
| 1457 |
+
[2091.84 --> 2094.94] better than the ads we didn't run when we recorded those ads.
|
| 1458 |
+
[2095.06 --> 2095.34] Like,
|
| 1459 |
+
[2095.40 --> 2096.70] let's just say we had some experiments.
|
| 1460 |
+
[2097.62 --> 2099.18] There were some stinkers in there.
|
| 1461 |
+
[2099.18 --> 2100.58] So really bad.
|
| 1462 |
+
[2100.58 --> 2104.18] You don't know how good they would have appreciated us rolling out the throwouts,
|
| 1463 |
+
[2104.18 --> 2105.18] uh,
|
| 1464 |
+
[2105.18 --> 2106.60] just to mix things up.
|
| 1465 |
+
[2106.60 --> 2109.32] It was the,
|
| 1466 |
+
[2109.38 --> 2109.48] yeah,
|
| 1467 |
+
[2109.52 --> 2109.74] it was,
|
| 1468 |
+
[2109.78 --> 2111.82] it was the fact that they're just listening to the same ads all the time.
|
| 1469 |
+
[2111.82 --> 2112.08] Not petition.
|
| 1470 |
+
[2112.22 --> 2114.04] And one of the things like people are not even like,
|
| 1471 |
+
[2114.12 --> 2116.62] don't run the ads or don't run the ads for oxide.
|
| 1472 |
+
[2116.88 --> 2117.34] They're just like,
|
| 1473 |
+
[2117.38 --> 2119.50] can you just run different ads for oxide?
|
| 1474 |
+
[2119.96 --> 2120.68] Advertise differently.
|
| 1475 |
+
[2120.88 --> 2121.06] Right.
|
| 1476 |
+
[2121.16 --> 2121.80] So one,
|
| 1477 |
+
[2121.90 --> 2122.54] one listener,
|
| 1478 |
+
[2122.86 --> 2123.18] uh,
|
| 1479 |
+
[2123.18 --> 2123.46] I mean,
|
| 1480 |
+
[2123.46 --> 2126.12] as we said in the ad just did unsolicited,
|
| 1481 |
+
[2126.34 --> 2127.54] send us an email and said,
|
| 1482 |
+
[2127.64 --> 2128.22] you know,
|
| 1483 |
+
[2128.44 --> 2128.88] look,
|
| 1484 |
+
[2129.16 --> 2129.48] uh,
|
| 1485 |
+
[2129.48 --> 2132.66] I have only been a taker of the podcast this entire time.
|
| 1486 |
+
[2132.72 --> 2133.38] I've given nothing.
|
| 1487 |
+
[2133.58 --> 2135.50] So I felt like I should give you something.
|
| 1488 |
+
[2135.50 --> 2138.06] The repetitive ads are killing me.
|
| 1489 |
+
[2138.16 --> 2141.14] So I've recorded my own ad for you that you're free.
|
| 1490 |
+
[2141.22 --> 2142.56] You're free to use if you'd like.
|
| 1491 |
+
[2142.68 --> 2145.34] And so obviously Steve and I were delighted by this.
|
| 1492 |
+
[2145.34 --> 2147.90] The Jonathan blow episode was the next episode up.
|
| 1493 |
+
[2147.90 --> 2148.90] So this is on,
|
| 1494 |
+
[2148.98 --> 2150.10] this is on Friday.
|
| 1495 |
+
[2150.24 --> 2150.94] This is last Friday.
|
| 1496 |
+
[2150.94 --> 2151.62] Cause this is a,
|
| 1497 |
+
[2151.62 --> 2153.12] Jonathan's episode was going to be,
|
| 1498 |
+
[2153.16 --> 2154.24] it was going to going on Monday.
|
| 1499 |
+
[2154.60 --> 2155.00] Jess,
|
| 1500 |
+
[2155.04 --> 2156.92] you were out of town and we're like,
|
| 1501 |
+
[2157.14 --> 2157.38] Steve,
|
| 1502 |
+
[2157.46 --> 2157.60] I can,
|
| 1503 |
+
[2157.70 --> 2159.80] we obviously have to get ads cut around this.
|
| 1504 |
+
[2159.80 --> 2161.96] We got to go to the garage now.
|
| 1505 |
+
[2162.66 --> 2164.38] And then there was this moment of like,
|
| 1506 |
+
[2164.56 --> 2165.58] we can't get the power on,
|
| 1507 |
+
[2165.72 --> 2166.06] but fortunately,
|
| 1508 |
+
[2166.06 --> 2168.14] this is where me,
|
| 1509 |
+
[2168.22 --> 2170.28] I am seeing absolutely everything.
|
| 1510 |
+
[2170.28 --> 2170.98] It was the wind.
|
| 1511 |
+
[2171.40 --> 2171.60] I know,
|
| 1512 |
+
[2171.68 --> 2175.18] but let's just say that like the IOT came through because the power doesn't work.
|
| 1513 |
+
[2175.62 --> 2176.38] Without it.
|
| 1514 |
+
[2176.62 --> 2176.80] Well,
|
| 1515 |
+
[2176.90 --> 2177.14] I mean,
|
| 1516 |
+
[2177.14 --> 2178.48] Steve's got a point.
|
| 1517 |
+
[2178.62 --> 2179.34] He's got a point.
|
| 1518 |
+
[2179.82 --> 2180.44] He does have the point.
|
| 1519 |
+
[2180.44 --> 2180.82] But yes,
|
| 1520 |
+
[2180.86 --> 2182.68] when the power does not work,
|
| 1521 |
+
[2182.68 --> 2183.18] right.
|
| 1522 |
+
[2183.22 --> 2185.30] It is nice to have someone in another state,
|
| 1523 |
+
[2185.38 --> 2186.10] be able to turn it on.
|
| 1524 |
+
[2186.10 --> 2189.58] I was able to GM you in another state and you were able to turn the power on,
|
| 1525 |
+
[2189.68 --> 2190.36] which was pretty great.
|
| 1526 |
+
[2190.48 --> 2191.28] It was pretty great.
|
| 1527 |
+
[2191.46 --> 2191.82] It was great.
|
| 1528 |
+
[2191.94 --> 2192.10] And,
|
| 1529 |
+
[2192.12 --> 2192.92] and truthfully,
|
| 1530 |
+
[2193.16 --> 2194.68] Steve's caveat didn't even occur to me,
|
| 1531 |
+
[2194.72 --> 2194.94] but right.
|
| 1532 |
+
[2194.98 --> 2195.24] Of course,
|
| 1533 |
+
[2195.28 --> 2195.50] if I could,
|
| 1534 |
+
[2195.54 --> 2196.24] there's just a power switch,
|
| 1535 |
+
[2196.30 --> 2196.94] we would have had to,
|
| 1536 |
+
[2197.18 --> 2197.44] but it,
|
| 1537 |
+
[2197.44 --> 2197.96] but it was fine.
|
| 1538 |
+
[2198.00 --> 2198.46] It was great.
|
| 1539 |
+
[2198.46 --> 2199.90] It worked and we got it recorded,
|
| 1540 |
+
[2200.40 --> 2201.06] but that was a lot of,
|
| 1541 |
+
[2201.12 --> 2201.96] that was a lot of fun.
|
| 1542 |
+
[2202.04 --> 2202.16] I mean,
|
| 1543 |
+
[2202.22 --> 2203.28] how great is it to have,
|
| 1544 |
+
[2203.70 --> 2204.70] and I think that the,
|
| 1545 |
+
[2205.18 --> 2205.36] you know,
|
| 1546 |
+
[2205.38 --> 2206.10] Chris at Humblepot,
|
| 1547 |
+
[2206.20 --> 2206.66] our producer,
|
| 1548 |
+
[2206.82 --> 2207.14] I think,
|
| 1549 |
+
[2207.26 --> 2211.54] I think it's fair to say he had never quite seen that one before.
|
| 1550 |
+
[2211.86 --> 2215.82] He had never seen a listener offer up,
|
| 1551 |
+
[2215.82 --> 2216.68] their own ad.
|
| 1552 |
+
[2216.94 --> 2217.34] Please,
|
| 1553 |
+
[2217.52 --> 2217.98] please.
|
| 1554 |
+
[2218.14 --> 2218.30] Anything.
|
| 1555 |
+
[2218.72 --> 2220.96] So like we'll take ads in the interim.
|
| 1556 |
+
[2221.14 --> 2221.30] Yeah.
|
| 1557 |
+
[2221.90 --> 2224.74] And we totally appreciate you.
|
| 1558 |
+
[2224.84 --> 2224.94] I,
|
| 1559 |
+
[2225.12 --> 2225.20] you,
|
| 1560 |
+
[2225.26 --> 2225.72] the listeners,
|
| 1561 |
+
[2225.72 --> 2228.94] it has been so much fun for us.
|
| 1562 |
+
[2229.20 --> 2229.36] Yeah.
|
| 1563 |
+
[2229.38 --> 2234.34] We have had so many conversations with people who have listened to the podcast for whom it's been meaningful.
|
| 1564 |
+
[2234.46 --> 2238.56] We've had people who've been in the industry a long time for whom they've reminisced.
|
| 1565 |
+
[2238.64 --> 2242.80] What I've loved is some of the younger folks who've listened to the podcast and realized,
|
| 1566 |
+
[2242.96 --> 2243.16] oh wow,
|
| 1567 |
+
[2243.18 --> 2244.48] there's a whole world down there that I didn't,
|
| 1568 |
+
[2244.62 --> 2245.54] I wasn't aware of.
|
| 1569 |
+
[2245.82 --> 2247.00] Which is kind of the point,
|
| 1570 |
+
[2247.30 --> 2247.54] you know,
|
| 1571 |
+
[2247.62 --> 2247.84] I mean,
|
| 1572 |
+
[2247.84 --> 2250.22] at some level it was really not,
|
| 1573 |
+
[2250.26 --> 2259.78] not to reminisce so much as it is to talk about this area of engineering that's still very active and interesting and has lots of hard problems.
|
| 1574 |
+
[2259.78 --> 2261.74] But I think to the credit of the folks that were on,
|
| 1575 |
+
[2261.74 --> 2262.16] I mean,
|
| 1576 |
+
[2262.16 --> 2262.18] I mean,
|
| 1577 |
+
[2262.18 --> 2262.58] the,
|
| 1578 |
+
[2262.58 --> 2262.66] the,
|
| 1579 |
+
[2262.66 --> 2264.94] the descriptions of these,
|
| 1580 |
+
[2264.94 --> 2265.52] of the,
|
| 1581 |
+
[2265.52 --> 2266.72] the lower levels of the stack,
|
| 1582 |
+
[2266.80 --> 2272.72] the descriptions of some of these hard technology problems were told in such a way that were very,
|
| 1583 |
+
[2272.80 --> 2273.58] very interesting,
|
| 1584 |
+
[2274.26 --> 2274.60] you know,
|
| 1585 |
+
[2274.66 --> 2276.10] historical and,
|
| 1586 |
+
[2276.10 --> 2279.44] and then relevant to a lot of the things that we're doing today.
|
| 1587 |
+
[2279.44 --> 2279.76] Yeah.
|
| 1588 |
+
[2279.76 --> 2280.94] And just captivating.
|
| 1589 |
+
[2280.94 --> 2281.14] I mean,
|
| 1590 |
+
[2281.14 --> 2281.38] it was,
|
| 1591 |
+
[2281.38 --> 2281.88] it was,
|
| 1592 |
+
[2282.04 --> 2282.44] it was,
|
| 1593 |
+
[2282.52 --> 2283.82] it was really,
|
| 1594 |
+
[2283.92 --> 2284.44] really interesting.
|
| 1595 |
+
[2285.08 --> 2286.42] Captivating and deeply personal.
|
| 1596 |
+
[2286.42 --> 2286.82] As you said,
|
| 1597 |
+
[2286.86 --> 2287.06] Jess,
|
| 1598 |
+
[2287.10 --> 2287.50] at the beginning,
|
| 1599 |
+
[2287.54 --> 2289.38] like we love stories and that's,
|
| 1600 |
+
[2289.46 --> 2290.14] that's what it's about.
|
| 1601 |
+
[2290.28 --> 2290.50] It's about,
|
| 1602 |
+
[2290.62 --> 2293.38] I think a lot of that just happened because we went into it with a,
|
| 1603 |
+
[2293.70 --> 2295.56] like there was no season before this.
|
| 1604 |
+
[2295.56 --> 2296.18] So it was like,
|
| 1605 |
+
[2296.20 --> 2296.90] everyone was just like,
|
| 1606 |
+
[2296.94 --> 2297.74] so what do I talk about?
|
| 1607 |
+
[2297.76 --> 2298.10] And we're like,
|
| 1608 |
+
[2298.56 --> 2299.66] let's just start with everything.
|
| 1609 |
+
[2299.66 --> 2300.10] Let's just go.
|
| 1610 |
+
[2300.40 --> 2303.22] Which we should also say the first intro,
|
| 1611 |
+
[2303.34 --> 2305.64] the first intro where I was like,
|
| 1612 |
+
[2305.66 --> 2307.62] I handed it over to Jess to make the intro.
|
| 1613 |
+
[2307.62 --> 2308.50] And Jeff's like,
|
| 1614 |
+
[2308.78 --> 2308.92] we,
|
| 1615 |
+
[2309.08 --> 2310.16] and we have,
|
| 1616 |
+
[2310.28 --> 2310.72] I think it was,
|
| 1617 |
+
[2310.80 --> 2311.42] it was at Rick.
|
| 1618 |
+
[2311.50 --> 2312.12] I wasn't Rick.
|
| 1619 |
+
[2312.68 --> 2313.00] And yeah,
|
| 1620 |
+
[2313.12 --> 2313.52] yeah.
|
| 1621 |
+
[2313.70 --> 2315.10] It was either Rick or Jeff.
|
| 1622 |
+
[2315.14 --> 2315.84] We were just like,
|
| 1623 |
+
[2315.90 --> 2316.92] and we have,
|
| 1624 |
+
[2317.20 --> 2317.74] it was,
|
| 1625 |
+
[2317.78 --> 2318.28] it was Jeff.
|
| 1626 |
+
[2318.40 --> 2319.48] And it's like,
|
| 1627 |
+
[2319.56 --> 2320.76] and we have Jeff here.
|
| 1628 |
+
[2320.82 --> 2321.92] Jess is going to intro who we have got.
|
| 1629 |
+
[2322.24 --> 2323.28] So I actually,
|
| 1630 |
+
[2323.38 --> 2324.26] Jeff Rothschild here.
|
| 1631 |
+
[2324.66 --> 2328.60] So I actually put a lot of thought into that though,
|
| 1632 |
+
[2328.66 --> 2329.28] behind the scenes,
|
| 1633 |
+
[2329.36 --> 2329.78] because I was like,
|
| 1634 |
+
[2329.84 --> 2330.10] well,
|
| 1635 |
+
[2330.18 --> 2333.20] maybe he doesn't want to be like intro does the thing that like everyone knows him for.
|
| 1636 |
+
[2333.28 --> 2334.96] Maybe he like wants to be known for something else.
|
| 1637 |
+
[2334.96 --> 2335.38] So that was like,
|
| 1638 |
+
[2335.56 --> 2337.08] maybe I'll just let him intro himself.
|
| 1639 |
+
[2337.08 --> 2337.46] Appreciate that.
|
| 1640 |
+
[2337.64 --> 2338.56] And then like,
|
| 1641 |
+
[2338.64 --> 2338.78] yeah.
|
| 1642 |
+
[2338.78 --> 2339.28] And also,
|
| 1643 |
+
[2339.40 --> 2339.60] I mean,
|
| 1644 |
+
[2339.66 --> 2340.52] it should be said that like,
|
| 1645 |
+
[2340.58 --> 2342.70] we were all three of us are a bit starstruck by Jeff.
|
| 1646 |
+
[2343.22 --> 2344.02] And I mean,
|
| 1647 |
+
[2344.02 --> 2344.40] he's just,
|
| 1648 |
+
[2344.52 --> 2346.64] he's had such an interesting career and been to so many different,
|
| 1649 |
+
[2346.72 --> 2347.26] interesting places.
|
| 1650 |
+
[2347.26 --> 2350.52] And so you don't want to intro him like in a way that he doesn't,
|
| 1651 |
+
[2350.52 --> 2350.78] you know,
|
| 1652 |
+
[2350.82 --> 2351.56] he wouldn't like.
|
| 1653 |
+
[2351.74 --> 2352.32] So then I was like,
|
| 1654 |
+
[2352.36 --> 2352.98] oh crap.
|
| 1655 |
+
[2352.98 --> 2353.34] Smart.
|
| 1656 |
+
[2353.78 --> 2355.30] So it was like,
|
| 1657 |
+
[2355.36 --> 2355.64] okay,
|
| 1658 |
+
[2355.68 --> 2356.40] back to me.
|
| 1659 |
+
[2356.54 --> 2356.70] So,
|
| 1660 |
+
[2356.78 --> 2358.62] and then I think we just decided like,
|
| 1661 |
+
[2358.92 --> 2360.86] that's the way we intro people around here.
|
| 1662 |
+
[2360.96 --> 2361.96] We just give your name.
|
| 1663 |
+
[2362.56 --> 2363.50] We didn't put them in a box.
|
| 1664 |
+
[2363.80 --> 2364.32] We didn't put them in a box.
|
| 1665 |
+
[2364.44 --> 2364.52] No,
|
| 1666 |
+
[2364.58 --> 2365.12] it was great.
|
| 1667 |
+
[2365.18 --> 2365.86] And it was fun.
|
| 1668 |
+
[2366.04 --> 2367.68] And it was so much fun.
|
| 1669 |
+
[2368.18 --> 2368.94] We loved it.
|
| 1670 |
+
[2369.08 --> 2371.58] So season two is coming.
|
| 1671 |
+
[2372.00 --> 2372.20] Yeah.
|
| 1672 |
+
[2372.36 --> 2374.00] And we're excited that you,
|
| 1673 |
+
[2374.14 --> 2375.96] all the listeners are excited.
|
| 1674 |
+
[2376.92 --> 2378.18] So stay tuned,
|
| 1675 |
+
[2378.24 --> 2378.46] right?
|
| 1676 |
+
[2379.38 --> 2379.98] Stay tuned.
|
| 1677 |
+
[2380.10 --> 2380.48] All right.
|
| 1678 |
+
[2380.48 --> 2380.84] Thanks.
|
| 1679 |
+
[2380.94 --> 2381.26] And Hey,
|
| 1680 |
+
[2381.64 --> 2382.20] don't forget,
|
| 1681 |
+
[2382.20 --> 2386.96] head on over to oxide.computer and join our mailing list.
|
| 1682 |
+
[2387.12 --> 2387.42] Thanks.
|
| 1683 |
+
[2387.50 --> 2387.98] Stay tuned to everyone.
|
| 1684 |
+
[2388.10 --> 2388.76] See you in season two.
|
| 1685 |
+
[2389.06 --> 2389.36] See ya.
|
| 1686 |
+
[2389.36 --> 2395.78] You've been listening to on the metal tales from the hardware software interface for show notes,
|
| 1687 |
+
[2395.78 --> 2398.58] to learn more about our guests or to sign up for our mailing list,
|
| 1688 |
+
[2399.00 --> 2401.30] visit us at on the metal.fm.
|
| 1689 |
+
[2401.78 --> 2404.22] On the metal is a production of oxide computer company.
|
| 1690 |
+
[2404.22 --> 2406.88] It is recorded in the oxide garage in Oakland,
|
| 1691 |
+
[2407.04 --> 2407.56] California.
|
| 1692 |
+
[2407.56 --> 2409.94] To learn more about oxide.computer.com.
|
| 1693 |
+
[2409.94 --> 2410.94] Visit us at oxide.computer.com.
|
| 1694 |
+
[2410.94 --> 2412.76] On the metal is hosted by me,
|
| 1695 |
+
[2412.88 --> 2413.50] Brian Cantrell,
|
| 1696 |
+
[2413.62 --> 2414.66] along with Jess for sale.
|
| 1697 |
+
[2414.88 --> 2416.60] And we are frequently joined by our boss,
|
| 1698 |
+
[2416.84 --> 2417.32] Steve Tuck.
|
| 1699 |
+
[2417.62 --> 2421.50] Our original and awesome theme music is by JJ Wiesler at pollen music group.
|
| 1700 |
+
[2421.72 --> 2425.42] You can learn more about JJ and pollen at pollen music group.com.
|
| 1701 |
+
[2425.42 --> 2430.58] We are edited and produced by Chris Hill and his crew at humble pod from Jess,
|
| 1702 |
+
[2430.78 --> 2431.40] from Steve,
|
| 1703 |
+
[2431.52 --> 2434.20] from me and from all of us at oxide computer company.
|
| 1704 |
+
[2434.32 --> 2435.82] Thanks for listening to on the metal.
|
| 1705 |
+
[2455.42 --> 2485.40] We are edited and produced by Chris Hill and his crew at home.
|
Jon Masters_transcript.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Jonathan Blow_transcript.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Kenneth Finnegan_transcript.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Rick Altherr_transcript.txt
ADDED
|
@@ -0,0 +1,1235 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[0.00 --> 14.42] Welcome to On The Metal, tales from the hardware software interface.
|
| 2 |
+
[14.64 --> 18.52] I'm Brian Cantrell. With me, as always, is Jess Frizzell. Hi, Jess.
|
| 3 |
+
[18.82 --> 19.34] Hi, Brian.
|
| 4 |
+
[19.92 --> 23.84] You know, what's so hysterical about that? You know, I can't say my name without you.
|
| 5 |
+
[24.00 --> 26.74] It was mostly because you told me to say, hi, Brian.
|
| 6 |
+
[26.74 --> 29.38] So then I was like thinking of something else to say really fast.
|
| 7 |
+
[29.38 --> 32.28] You kept saying hello as opposed to like, sounds like-
|
| 8 |
+
[32.28 --> 33.08] Yeah, it's like hello. Hey.
|
| 9 |
+
[33.24 --> 33.40] All right.
|
| 10 |
+
[33.46 --> 34.48] I thought it was a dramatic intro.
|
| 11 |
+
[34.70 --> 38.92] All right. So, and speaking of whom, Steve Tuck also with us, our boss. Steve, welcome.
|
| 12 |
+
[39.30 --> 40.12] Glad to be invited.
|
| 13 |
+
[40.34 --> 41.26] Be on your best behavior.
|
| 14 |
+
[41.66 --> 41.86] Always.
|
| 15 |
+
[42.30 --> 44.90] And Jess, you want to introduce our esteemed guest?
|
| 16 |
+
[45.30 --> 50.02] Yeah. So our guest today is Rick Alter. He was one of the first people who I met, honestly,
|
| 17 |
+
[50.18 --> 55.14] in the Firmar space. And I've learned a lot from him over just the past few months.
|
| 18 |
+
[55.14 --> 60.86] And most recently, he found a vulnerability in a lot of super micro BMCs.
|
| 19 |
+
[61.16 --> 62.52] A lot of BMCs.
|
| 20 |
+
[62.74 --> 67.76] Called USB Anywhere. So do you want to maybe tell us a little bit about that?
|
| 21 |
+
[68.22 --> 74.62] Sure. So hello. Yeah. USB Anywhere was a fun one in that I work for a company now that
|
| 22 |
+
[74.62 --> 79.96] explicitly looks for security vulnerabilities in system firmware and produces an enterprise
|
| 23 |
+
[79.96 --> 81.96] product around this. So like, this is my day job now.
|
| 24 |
+
[81.96 --> 85.66] I mean, that's a target-rich environment, as Don Rumsfeld would say.
|
| 25 |
+
[85.96 --> 90.70] It definitely is. And, you know, things have been getting slightly better in the firmware
|
| 26 |
+
[90.70 --> 94.88] scene. I'm sure we'll talk about that in more depth in a bit. But, you know, USB Anywhere
|
| 27 |
+
[94.88 --> 102.32] was really started from just me saying, I know that this feature exists where I can mount a ISO
|
| 28 |
+
[102.32 --> 108.18] image as a CD-ROM drive on a server over the internet. I wonder how that works. And did they do it
|
| 29 |
+
[108.18 --> 114.22] correctly? And it turns out, no, they didn't. And it was done in probably one of the most dangerous
|
| 30 |
+
[114.22 --> 119.70] ways possible. So after doing a bunch of research about a lot of reverse engineering, working with
|
| 31 |
+
[119.70 --> 125.42] Supermicro to get fixes developed, and announced that and give a talk about it at OSFC.
|
| 32 |
+
[125.68 --> 128.10] A great talk. OSFC, the open source firmware conference.
|
| 33 |
+
[128.18 --> 128.42] Right.
|
| 34 |
+
[128.42 --> 129.50] Loved it. Great conference.
|
| 35 |
+
[130.04 --> 135.80] And also, you know, with that revealed that I had actually scanned the internet and found 47,000
|
| 36 |
+
[135.80 --> 139.08] servers just sitting on the internet waiting to be exploited via this.
|
| 37 |
+
[139.42 --> 142.70] I mean, that is 47,000 BMCs on the internet.
|
| 38 |
+
[143.02 --> 143.18] Right.
|
| 39 |
+
[143.32 --> 146.00] I mean, that is like we put the brainstem on the internet.
|
| 40 |
+
[146.76 --> 151.88] Yeah. Yeah, exactly. It's, you know, here's your best practice. Here's what people really do.
|
| 41 |
+
[151.88 --> 156.82] I got in contact with some of the larger network operators that had these on there, and everyone
|
| 42 |
+
[156.82 --> 161.84] was just as mystified as I was. It's, no one clearly knew why these were on the internet,
|
| 43 |
+
[162.10 --> 163.66] and, but everybody knew they shouldn't be.
|
| 44 |
+
[163.98 --> 167.26] Was that number higher than you thought it was going to be? That is such an astonishingly
|
| 45 |
+
[167.26 --> 167.70] high number.
|
| 46 |
+
[168.12 --> 172.50] It's somewhere in the range that I was expecting. There have been people in the past that have
|
| 47 |
+
[172.50 --> 176.72] done similar scans looking for BMCs, and they've found significantly higher numbers in
|
| 48 |
+
[176.72 --> 178.28] the past. So this is actually a reduction.
|
| 49 |
+
[178.68 --> 179.40] It's the good news.
|
| 50 |
+
[179.52 --> 179.88] Yeah.
|
| 51 |
+
[179.88 --> 184.36] Yeah. Were these highly clustered, or was it distributed and kind of everyone had one?
|
| 52 |
+
[184.80 --> 188.46] It kind of came in lumps. Like if you broke it down geographically, the highest density
|
| 53 |
+
[188.46 --> 194.28] was in the United States, but then they were scattered all over Europe, Asia, South America.
|
| 54 |
+
[194.52 --> 198.72] Fairly well distributed, but definitely concentrated in the US. There were a couple of network operators
|
| 55 |
+
[198.72 --> 201.34] that were very heavily affected by it.
|
| 56 |
+
[201.50 --> 201.72] Interesting.
|
| 57 |
+
[201.76 --> 206.66] And I got in touch with them directly to say, look, this is a real problem. You should figure
|
| 58 |
+
[206.66 --> 209.86] out who these machines are owned by and do something about it.
|
| 59 |
+
[209.88 --> 214.44] I mean, you seem like a very law-abiding person, but it must have been tempting to be like,
|
| 60 |
+
[214.68 --> 221.10] wow. I mean, especially to go through the 47,000, like, do I have any of my enemies list
|
| 61 |
+
[221.10 --> 222.20] on here, by any chance?
|
| 62 |
+
[222.20 --> 228.02] I mean, we did look up as to, part of after getting 47,000 IP addresses, right? Because
|
| 63 |
+
[228.02 --> 233.22] that's really what happened. I scanned the entire IPv4 address space. And so I only knew the IP
|
| 64 |
+
[233.22 --> 236.76] addresses initially. And so trying to do like reverse DNS lookups and then look up what
|
| 65 |
+
[236.76 --> 242.18] autonomous system they're attached to and things like that, trying to narrow down who it was,
|
| 66 |
+
[242.36 --> 247.12] kind of taking a peek through it and going, is there anybody that I like know personally or that I,
|
| 67 |
+
[247.12 --> 251.00] you know, have some interaction with that company that I can talk to them about this.
|
| 68 |
+
[251.32 --> 255.70] And thankfully, like it wasn't anybody I knew, but like, as I said, there were some very large
|
| 69 |
+
[255.70 --> 259.30] network operators that were affected by it. And it was just terrifying.
|
| 70 |
+
[259.56 --> 263.64] You know, thankfully they got in touch and asked good questions about how to fix it.
|
| 71 |
+
[263.80 --> 267.42] So how did you, I mean, I've got a question about how you actually found that vulnerability.
|
| 72 |
+
[267.60 --> 270.78] What was the process of reverse engineering for you? I mean, you obviously you're going into this
|
| 73 |
+
[270.78 --> 273.70] piece of functionality thinking, if this hasn't been implemented properly,
|
| 74 |
+
[273.70 --> 277.54] there's likely a vulnerability here. What was your tool set, your mindset? How did you go about
|
| 75 |
+
[277.54 --> 281.42] doing that? So I actually gave a talk about this at Mountain View reverse engineering group recently.
|
| 76 |
+
[281.76 --> 286.86] So unfortunately those usually aren't recorded, but the main thing was I knew what the external
|
| 77 |
+
[286.86 --> 293.50] functionality was and having been in the BMC space for quite a while developing or being part of early
|
| 78 |
+
[293.50 --> 299.86] part of open BMC and developing BMC firmware, I kind of knew what the capabilities of the chip actually
|
| 79 |
+
[299.86 --> 303.66] were. And so it was really just a question of what was all the in-between, like,
|
| 80 |
+
[303.70 --> 308.28] if I knew on the host side that this looks like a USB device and I know that the hardware,
|
| 81 |
+
[308.86 --> 316.10] like the BMC SOC actually has the ability to look like a USB device, practically any USB device,
|
| 82 |
+
[316.24 --> 320.80] then clearly that's the feature that's being used here. What does the full stack look like?
|
| 83 |
+
[321.12 --> 326.60] And that's going to be split between the firmware actually on the BMC and the Java application
|
| 84 |
+
[326.60 --> 329.56] that was running on my laptop, you know, my remote workstation.
|
| 85 |
+
[329.56 --> 335.78] And I'm one to really dig into protocols, like that's sort of my thing is getting into,
|
| 86 |
+
[335.78 --> 340.50] you know, serialized binary formats. So I just grabbed the packet capture of the network and
|
| 87 |
+
[340.50 --> 350.04] looked to see what was going on. And as soon as I saw a TCP 623, which was the port that this service
|
| 88 |
+
[350.04 --> 356.58] was running on, and I saw strings of USB in them, you know, it was just sort of like a telltale,
|
| 89 |
+
[356.68 --> 362.64] like something is very wrong. The actual making a proof of concept was the more difficult part.
|
| 90 |
+
[362.76 --> 369.30] So like the early that there is something wrong was, you know, an hour of work, and it really involved
|
| 91 |
+
[369.30 --> 376.08] Wireshark, and that's about it. Actually understanding the protocol enough to implement my own client
|
| 92 |
+
[376.08 --> 383.50] so that I could then use it in malicious ways. That involved getting the firmware updates,
|
| 93 |
+
[383.74 --> 390.80] unpacking it. BMCs often are a Linux-based system on an ARM processor. And so you can just unpack the
|
| 94 |
+
[390.80 --> 396.38] file system and start opening binaries and things. And then using tools like Ghidra to actually
|
| 95 |
+
[396.38 --> 400.46] decompile the binaries and start looking through how do they actually work.
|
| 96 |
+
[400.46 --> 408.06] This is presumably an A-speed 2400, 2500? Yes. Well, some of them. Some of them were A-speed.
|
| 97 |
+
[408.28 --> 413.60] Some of them were Nuvitan. Oh, really? Wow. Older Nuvitan. Interesting. I have a slightly
|
| 98 |
+
[413.60 --> 420.30] off-topic question. What is your favorite computer that you've ever owned and why? Oh, see, you added
|
| 99 |
+
[420.30 --> 426.32] that I ever owned. Well, okay. That makes it hard. Or, you know, came into contact with.
|
| 100 |
+
[426.92 --> 429.42] Jess is looking for a love story. Is that a fair statement? I am.
|
| 101 |
+
[429.42 --> 432.88] Well, see, so- She wants to know when you first fell in love with a computer.
|
| 102 |
+
[433.10 --> 441.90] So my love for a computer would definitely be the Cray, the large circular Cray computers.
|
| 103 |
+
[441.94 --> 445.32] With the seats? Yeah, like the YMP? Like a Cray YMP1?
|
| 104 |
+
[445.54 --> 448.00] Yeah, yeah. Wow. The seats are cool.
|
| 105 |
+
[448.10 --> 455.14] How? Okay. And it's interesting that I've actually got to sit on one. Like, I mean,
|
| 106 |
+
[455.14 --> 456.76] these are museum pieces. Right. Yeah, exactly.
|
| 107 |
+
[457.18 --> 459.20] Okay, this is not a home computer. No.
|
| 108 |
+
[459.20 --> 466.30] No, no, no, no, no. No. They actually show this in the movie Sneakers, right? Or a related model.
|
| 109 |
+
[466.84 --> 473.00] But yeah, so the Deutsches Museum in Munich actually has one and it's just out there. And they're like,
|
| 110 |
+
[473.06 --> 478.70] yes, you may sit on it and enjoy the machine. So I've actually had a discussion with my wife
|
| 111 |
+
[478.70 --> 481.44] about whether or not we can incorporate this into the decor at home.
|
| 112 |
+
[482.14 --> 482.88] How did that go?
|
| 113 |
+
[482.88 --> 486.90] Yeah. She actually worked with me on it. We found a location that it would work if I could actually
|
| 114 |
+
[486.90 --> 492.46] find one. So I'm open for it. If somebody has one that would like to sell one, please get in touch.
|
| 115 |
+
[492.74 --> 498.30] And is your wife actually interested in this thing? Or does she just, is her love for you that
|
| 116 |
+
[498.30 --> 502.62] unconditional that she's willing to tolerate a Cray supercomputer around the house?
|
| 117 |
+
[502.62 --> 505.82] I think it's more, she can appreciate the aesthetic.
|
| 118 |
+
[506.20 --> 509.88] There you go. It actually, the aesthetic, I mean, if people haven't seen a picture of it,
|
| 119 |
+
[509.94 --> 511.04] they should. It is.
|
| 120 |
+
[511.14 --> 511.26] It's cool.
|
| 121 |
+
[511.44 --> 514.82] It looks like a canonical supercomputer. It looks very bad.
|
| 122 |
+
[514.82 --> 516.42] It looks like 2001 Space Odyssey.
|
| 123 |
+
[517.14 --> 519.12] It does. Steve, you ever seen a photo of these things?
|
| 124 |
+
[519.26 --> 521.70] Well, I asked if it was a home computer, so I think the answer is no.
|
| 125 |
+
[521.70 --> 525.46] No. They're cylindrical. So they-
|
| 126 |
+
[525.46 --> 527.10] Well, they make a letter C from above.
|
| 127 |
+
[527.62 --> 528.38] Did they?
|
| 128 |
+
[528.52 --> 528.84] Interesting.
|
| 129 |
+
[528.84 --> 530.78] God, Seymour. Oh, Seymour.
|
| 130 |
+
[530.90 --> 532.36] Oh, that makes so much sense.
|
| 131 |
+
[532.36 --> 536.10] I did not actually know that. I'm embarrassed to say I did not know that. Yeah. And have
|
| 132 |
+
[536.10 --> 537.26] you ever programmed one of those machines?
|
| 133 |
+
[537.32 --> 539.06] Oh, no. I've never even seen one powered up.
|
| 134 |
+
[539.16 --> 542.60] Okay. So it's just, it is, for you, the Cray is purely the aesthetic.
|
| 135 |
+
[542.86 --> 546.24] Yeah. Oh, yeah. Well, I mean, there's a whole bunch of aspects to it. Like when you get
|
| 136 |
+
[546.24 --> 551.38] into the details of how it actually works and the intricacy of design internally. I mean,
|
| 137 |
+
[551.38 --> 552.14] they're fascinating.
|
| 138 |
+
[552.24 --> 552.86] They are amazing.
|
| 139 |
+
[553.04 --> 557.88] It would be completely pointless to have one that you powered up and tried to use these
|
| 140 |
+
[557.88 --> 562.28] days. I mean, it would be fun from a nostalgia perspective, but it's hard to justify the
|
| 141 |
+
[562.28 --> 563.98] power budget to run one of those.
|
| 142 |
+
[564.12 --> 568.64] You also just have to love the fact that Seymour Cray did not believe in SMP. It's like
|
| 143 |
+
[568.64 --> 572.34] did not believe that God intended multiprocessors. He's like, no, we are going to make this single
|
| 144 |
+
[572.34 --> 575.04] processor bigger and faster. Get out of my way.
|
| 145 |
+
[575.30 --> 576.08] And put seats.
|
| 146 |
+
[576.48 --> 577.02] And seats.
|
| 147 |
+
[577.52 --> 579.22] Just in case you want to like sit and watch it.
|
| 148 |
+
[579.36 --> 581.24] Oh, no. There's, and his, I don't know if you're, if you're,
|
| 149 |
+
[581.24 --> 587.52] his machines from CDC, like the 1604 to the 6600, amazing machines. Now he was a ridiculously
|
| 150 |
+
[587.52 --> 589.18] amazing guy.
|
| 151 |
+
[589.58 --> 594.36] Now in terms of machines I've owned, probably the best machine I ever had, or my favorite
|
| 152 |
+
[594.36 --> 599.60] was I had a deck station or sorry, an alpha station. So it was, it was still branded digital,
|
| 153 |
+
[599.74 --> 600.68] but it had an alpha station.
|
| 154 |
+
[600.80 --> 604.02] Nice. What operating system did you run on the alpha station?
|
| 155 |
+
[604.18 --> 608.44] Well, see, interestingly enough, I got this at a time when I was actually administrating
|
| 156 |
+
[608.44 --> 609.06] True64.
|
| 157 |
+
[609.62 --> 610.56] Ooh, there you go.
|
| 158 |
+
[610.66 --> 612.98] So, so that was my, like my day job.
|
| 159 |
+
[613.20 --> 615.52] True64 without the E, there is no E in True64.
|
| 160 |
+
[615.56 --> 616.68] There is no E in True64.
|
| 161 |
+
[616.92 --> 617.62] T-R-U.
|
| 162 |
+
[617.82 --> 620.22] T-R-U 64. It sounds like a pack of like cigarettes.
|
| 163 |
+
[621.36 --> 624.62] It's like, are those, do you have any True64 lights, please?
|
| 164 |
+
[624.88 --> 626.72] They were very much ahead of their times in branding.
|
| 165 |
+
[626.72 --> 627.58] They were.
|
| 166 |
+
[627.76 --> 628.24] They were.
|
| 167 |
+
[628.38 --> 628.78] Yes.
|
| 168 |
+
[629.20 --> 630.48] True.ly 64.
|
| 169 |
+
[631.46 --> 631.90] Yeah.
|
| 170 |
+
[632.00 --> 636.66] So I had, I had that at home and, you know, interesting thing with the, that generation
|
| 171 |
+
[636.66 --> 640.26] of alphas is you could run True64, you could run Windows NT.
|
| 172 |
+
[640.38 --> 640.78] Yes.
|
| 173 |
+
[641.26 --> 641.64] Yes.
|
| 174 |
+
[641.66 --> 644.10] Which was always a little awkward and interesting.
|
| 175 |
+
[644.72 --> 644.88] Yes.
|
| 176 |
+
[644.88 --> 645.94] You could run Windows NT.
|
| 177 |
+
[646.08 --> 650.32] Well, in the NT, you could run NT on, this was the era that NT was going to be everywhere.
|
| 178 |
+
[650.60 --> 650.86] Right.
|
| 179 |
+
[651.04 --> 652.30] And they ported it to the Alpha.
|
| 180 |
+
[652.72 --> 652.90] Yep.
|
| 181 |
+
[653.08 --> 654.38] Which was an amazing CPU.
|
| 182 |
+
[654.72 --> 656.72] And the, they also had.
|
| 183 |
+
[657.18 --> 657.80] Go on.
|
| 184 |
+
[658.20 --> 658.60] Spark.
|
| 185 |
+
[658.82 --> 659.10] Yes.
|
| 186 |
+
[659.18 --> 659.44] And.
|
| 187 |
+
[659.70 --> 662.04] Oh, and what was the other?
|
| 188 |
+
[662.04 --> 662.62] Oh, they had PowerPC.
|
| 189 |
+
[662.82 --> 663.38] They had PowerPC.
|
| 190 |
+
[663.70 --> 664.56] There's another ISA.
|
| 191 |
+
[664.74 --> 665.22] And MIPS.
|
| 192 |
+
[665.54 --> 666.28] Very good.
|
| 193 |
+
[666.36 --> 666.64] MIPS.
|
| 194 |
+
[666.88 --> 667.04] Nice.
|
| 195 |
+
[667.04 --> 667.30] Okay.
|
| 196 |
+
[667.60 --> 668.70] So I had.
|
| 197 |
+
[668.70 --> 669.10] You know what's really sad?
|
| 198 |
+
[669.18 --> 672.28] I had to think of the actual folder structure inside the installer.
|
| 199 |
+
[672.56 --> 673.64] That's, that's very good.
|
| 200 |
+
[673.82 --> 679.36] I, the only reason that I, that I had to have NT on MIPS on the mind is because back in
|
| 201 |
+
[679.36 --> 685.36] the day, the Palo Alto Goodwill was a glorious place to go because you'd go into the t-shirt
|
| 202 |
+
[685.36 --> 691.10] bin and for a nickel, you could buy corpses effectively because everyone would get these,
|
| 203 |
+
[691.16 --> 694.18] all these t-shirts for various initiatives and then they would give them away.
|
| 204 |
+
[694.50 --> 698.04] So I, for, I think it was actually 25 cents, not a nickel.
|
| 205 |
+
[698.04 --> 706.18] I bought a t-shirt from SGI that had the top 10 reasons to run NT on MIPS on the back.
|
| 206 |
+
[706.78 --> 707.10] What?
|
| 207 |
+
[707.30 --> 707.98] Like listed.
|
| 208 |
+
[708.14 --> 710.30] And the reasons, this is not sophisticated ad copy.
|
| 209 |
+
[710.54 --> 713.26] I think number five was Spark Schmark.
|
| 210 |
+
[713.64 --> 716.14] It was just like, I mean, come on.
|
| 211 |
+
[716.70 --> 718.70] So SGI made these shirts.
|
| 212 |
+
[718.86 --> 723.26] SGI made these shirts because SGI was banking on SGI, Silicon Graphics.
|
| 213 |
+
[723.26 --> 729.48] Um, I mean, I know Rick knows this because he's dropping NT on MIPS, but it's crazy.
|
| 214 |
+
[729.66 --> 733.78] I mean, it's just like how, this is one of these things you wonder how many people ran
|
| 215 |
+
[733.78 --> 734.68] NT on MIPS?
|
| 216 |
+
[735.00 --> 739.54] Oh, I, I mean, I've not encountered many people who ran NT on alpha and I'm pretty sure that
|
| 217 |
+
[739.54 --> 740.48] was much more common.
|
| 218 |
+
[740.72 --> 741.02] Yes.
|
| 219 |
+
[741.44 --> 745.00] When you had NT on all these different platforms, one of the issues was the binary compatibility,
|
| 220 |
+
[745.28 --> 745.46] right?
|
| 221 |
+
[745.50 --> 749.20] It's still, if you got an application, it had to be for that specific architecture.
|
| 222 |
+
[749.20 --> 749.44] Right.
|
| 223 |
+
[749.44 --> 752.32] But on the alpha, they had a program called FX32.
|
| 224 |
+
[752.58 --> 757.18] So you could take an x86 binary, drop it on FX32, and it would convert it to run on,
|
| 225 |
+
[757.18 --> 758.06] on alpha.
|
| 226 |
+
[758.22 --> 762.18] And this is at a time when alpha is so much faster than x86 that it may have been faster
|
| 227 |
+
[762.18 --> 763.28] than x86 to do that.
|
| 228 |
+
[763.60 --> 764.94] It may have been.
|
| 229 |
+
[765.04 --> 768.14] I, you know, I never actually got around to playing with NT on it.
|
| 230 |
+
[768.20 --> 770.98] I just, I know all the details and I looked into how it worked.
|
| 231 |
+
[771.02 --> 776.02] And, uh, yeah, it, it actually took me like six months to acquire the correct RAM to boot
|
| 232 |
+
[776.02 --> 776.58] the machine.
|
| 233 |
+
[776.58 --> 780.96] And true 64, if I'm, that's an ILP 64 kernel, right?
|
| 234 |
+
[781.08 --> 781.30] Yes.
|
| 235 |
+
[781.56 --> 781.72] Yeah.
|
| 236 |
+
[781.80 --> 783.68] So I, so integers were 64 bit.
|
| 237 |
+
[783.86 --> 786.00] It's like, that's going to break some software.
|
| 238 |
+
[786.36 --> 789.16] And that was not a good approach as it turns out.
|
| 239 |
+
[789.22 --> 791.98] I think we, humanity went a different direction on that.
|
| 240 |
+
[792.06 --> 793.00] Went to LP 64.
|
| 241 |
+
[793.68 --> 794.16] That was great.
|
| 242 |
+
[794.20 --> 795.40] So you, you own that box?
|
| 243 |
+
[795.64 --> 796.34] Uh, not anymore.
|
| 244 |
+
[796.50 --> 798.00] I had that for quite a while.
|
| 245 |
+
[798.10 --> 798.58] That's dope.
|
| 246 |
+
[798.66 --> 799.30] Is that as sufficient?
|
| 247 |
+
[799.40 --> 801.04] That's a good answer.
|
| 248 |
+
[801.22 --> 803.20] I mean, I wanted a love story and we got two.
|
| 249 |
+
[803.40 --> 803.92] We got two.
|
| 250 |
+
[804.16 --> 805.78] Are you counting NT on MIPS as a love story?
|
| 251 |
+
[805.78 --> 806.42] I don't think that one counts.
|
| 252 |
+
[806.42 --> 806.58] No, the cray.
|
| 253 |
+
[806.84 --> 807.88] Oh, the cray got away.
|
| 254 |
+
[807.96 --> 809.52] The one that got away and the one that he had.
|
| 255 |
+
[809.62 --> 809.88] Yeah.
|
| 256 |
+
[810.60 --> 813.90] I found a cray, by the way, sold in 2002 for $45,000.
|
| 257 |
+
[814.58 --> 816.66] See, I mean, that's a worthwhile investment.
|
| 258 |
+
[816.80 --> 816.92] Yeah.
|
| 259 |
+
[817.28 --> 821.48] I feel there is a lost art of making computers as furniture.
|
| 260 |
+
[821.78 --> 824.04] It's clearly something that needs to come back.
|
| 261 |
+
[824.08 --> 824.56] It does.
|
| 262 |
+
[825.38 --> 827.52] Maybe there's something that the Amish could pick up.
|
| 263 |
+
[827.66 --> 830.74] You know, they seem to, they've got the furniture or craftsmanship.
|
| 264 |
+
[830.74 --> 833.32] They don't, they distinctly don't like power.
|
| 265 |
+
[833.52 --> 836.58] You know, but they're willing to resell people what they want.
|
| 266 |
+
[836.66 --> 837.78] I think that they should pick this up.
|
| 267 |
+
[837.82 --> 838.88] I think this should be an Amish thing.
|
| 268 |
+
[839.42 --> 844.22] I'm saying they should buy old computer gear and, you know, put some niche.
|
| 269 |
+
[844.46 --> 844.68] Nice.
|
| 270 |
+
[844.92 --> 845.36] A little niche.
|
| 271 |
+
[845.60 --> 845.84] All right.
|
| 272 |
+
[846.26 --> 846.92] I'd buy it.
|
| 273 |
+
[847.38 --> 851.58] Can I ask one question back to the BMC exploits?
|
| 274 |
+
[851.74 --> 851.88] Yeah.
|
| 275 |
+
[851.88 --> 852.86] Or just the vulnerability.
|
| 276 |
+
[853.28 --> 857.54] So with 45,000 hanging out on the internet, seven, right?
|
| 277 |
+
[857.58 --> 858.26] 47,000.
|
| 278 |
+
[859.08 --> 861.60] How vulnerable is someone at that point?
|
| 279 |
+
[861.66 --> 863.22] Like, is this easily exploitable?
|
| 280 |
+
[863.36 --> 863.62] I know.
|
| 281 |
+
[863.68 --> 864.02] Oh my God.
|
| 282 |
+
[864.04 --> 865.52] I almost spit up Diet Coke under my mic.
|
| 283 |
+
[865.86 --> 868.68] How vulnerable is someone with a BMC on the internet?
|
| 284 |
+
[868.96 --> 874.24] I know it's like, you know, hanging wires into water, but I mean, is it easy?
|
| 285 |
+
[874.34 --> 874.86] Go ahead, Rick.
|
| 286 |
+
[874.86 --> 878.64] How vulnerable am I by hanging my BMC out on the internet?
|
| 287 |
+
[879.50 --> 880.62] Oh, so many ways.
|
| 288 |
+
[880.62 --> 886.32] I mean, the thing is that like BMCs fundamentally were designed in the 90s.
|
| 289 |
+
[886.96 --> 893.32] And a lot of the standards that are implemented around BMCs actually never evolved much past the early 2000s.
|
| 290 |
+
[893.80 --> 900.32] And so a lot of the, what you see today as the state of the art is actually things like IPMI version two,
|
| 291 |
+
[900.48 --> 904.90] where according to the standard, you have to support things like no authentication.
|
| 292 |
+
[904.90 --> 906.80] Like it's, it's a required mode.
|
| 293 |
+
[906.94 --> 908.00] Can't turn that off.
|
| 294 |
+
[908.00 --> 913.30] The only mandated ciphers are all things that are, have known weaknesses.
|
| 295 |
+
[913.84 --> 920.14] And ultimately the IPMI implementers forum, I forget what the actual standards organization is.
|
| 296 |
+
[920.18 --> 922.10] They're like, we're just not going to support this anymore.
|
| 297 |
+
[922.10 --> 927.62] So there's this new thing called Redfish that isn't actually done yet, but the spec is there.
|
| 298 |
+
[927.72 --> 929.00] So you should just do that instead.
|
| 299 |
+
[929.80 --> 936.98] But what it means is that pretty much any BMC that gets put on the internet today is, it's terrible.
|
| 300 |
+
[937.62 --> 938.06] There are-
|
| 301 |
+
[938.06 --> 938.66] You're walking right into-
|
| 302 |
+
[938.66 --> 942.88] Well, also you're, you're talking about the vulnerabilities absent the actual vulnerability that you found.
|
| 303 |
+
[942.88 --> 943.46] Yeah, yeah.
|
| 304 |
+
[943.48 --> 944.24] I mean, it's a reminder.
|
| 305 |
+
[944.52 --> 944.68] Yeah.
|
| 306 |
+
[944.78 --> 953.00] Like Rick found a vulnerability in which, and like, if your BMC gets owned by someone who's sophisticated, you are done.
|
| 307 |
+
[953.88 --> 954.28] Yeah.
|
| 308 |
+
[954.62 --> 960.72] I mean, it's, it's also hard to get at yourself a guarantee about, have I actually removed the exploit from this machine?
|
| 309 |
+
[960.72 --> 961.04] Right?
|
| 310 |
+
[961.14 --> 965.32] Like you can maybe reflash the BMC, but what else has been contaminated on the system?
|
| 311 |
+
[966.48 --> 974.80] So when you start working through firmware security and everything, it gets into these situations where at what level do I just write off the hardware and, and feed it through a shredder?
|
| 312 |
+
[974.94 --> 975.04] Yeah.
|
| 313 |
+
[975.06 --> 980.46] I mean, Rick is saying it gently, but he's saying if Rick owns your BMC, you got no way of getting him out.
|
| 314 |
+
[980.92 --> 982.28] I mean, you got to pulp the machine.
|
| 315 |
+
[982.36 --> 983.86] If you are a sophisticated actor.
|
| 316 |
+
[984.30 --> 984.62] Yeah.
|
| 317 |
+
[984.98 --> 985.16] Yeah.
|
| 318 |
+
[985.28 --> 990.32] I mean, a lot of folks wouldn't like, that's, that's hard to justify as a business model, right?
|
| 319 |
+
[990.32 --> 990.92] Oh yeah.
|
| 320 |
+
[991.18 --> 991.74] Like, absolutely.
|
| 321 |
+
[991.90 --> 993.08] It's criminal almost certainly.
|
| 322 |
+
[993.22 --> 1000.74] So people are going to say, well, if this level of attack happens, like I'm, I'm just going to do these things to recondition it and call it good.
|
| 323 |
+
[1000.88 --> 1007.74] But yeah, I mean, this is kind of what my research is, is looking at ways that you can own the machine and at deeper and deeper levels.
|
| 324 |
+
[1008.08 --> 1016.16] But with BMCs, because of the way the standards are written, because of the practices in that era, ultimately sort of the stagnation in that space,
|
| 325 |
+
[1016.16 --> 1022.68] you see things like machines that are shipped with a default username and password for the BMC where it's admin, admin.
|
| 326 |
+
[1023.38 --> 1027.08] And the expectation is that you will change it, but there's no enforcement on that.
|
| 327 |
+
[1027.18 --> 1028.30] Or root Calvin, right?
|
| 328 |
+
[1028.72 --> 1031.04] Root Calvin is, is the Dell iDRAC one.
|
| 329 |
+
[1031.16 --> 1031.24] Yeah.
|
| 330 |
+
[1031.34 --> 1032.08] Who is Calvin?
|
| 331 |
+
[1032.64 --> 1033.24] Do we know this?
|
| 332 |
+
[1033.44 --> 1033.72] No.
|
| 333 |
+
[1033.90 --> 1034.68] Calvin and Hobbes.
|
| 334 |
+
[1035.02 --> 1035.68] Is that what that is?
|
| 335 |
+
[1035.72 --> 1036.20] No, I don't know.
|
| 336 |
+
[1036.24 --> 1036.88] I just made that up.
|
| 337 |
+
[1037.38 --> 1037.70] Yeah.
|
| 338 |
+
[1037.70 --> 1040.08] I, it's a, it's a long lost mystery.
|
| 339 |
+
[1041.46 --> 1049.34] But at least some of the vendors like HPE are starting to do things like generate random eight character passwords for each machine.
|
| 340 |
+
[1049.34 --> 1051.20] And so each machine comes with a password.
|
| 341 |
+
[1051.36 --> 1057.34] But one of the things I also mentioned at the, the talk at OpenSource Firmware Conference was like,
|
| 342 |
+
[1057.42 --> 1063.30] you can rent machines with massive amounts of GPUs and run password crackers on them.
|
| 343 |
+
[1063.30 --> 1069.90] And so like, I actually have two HPE machines in my lab and they have the default passwords and I didn't actually know what they were.
|
| 344 |
+
[1070.34 --> 1074.52] So I ran a password cracker across all of the BMC passwords.
|
| 345 |
+
[1074.92 --> 1078.88] And, you know, those took a little bit longer, but we're talking like 20 hours.
|
| 346 |
+
[1079.42 --> 1085.74] So, I mean, there's so many different ways that you have to have done things right to make sure that somebody isn't getting into your BMC.
|
| 347 |
+
[1085.82 --> 1090.46] And then you look at things like USB Anywhere where I didn't need any credentials.
|
| 348 |
+
[1090.46 --> 1099.60] The, you know, one of the failures found was actually the ability to just use the service as an existing user without actually logging in at all.
|
| 349 |
+
[1099.86 --> 1100.62] Absolutely terrifying.
|
| 350 |
+
[1100.98 --> 1102.92] And, and this is, it's the brainstem of the box.
|
| 351 |
+
[1103.00 --> 1104.04] This controls everything.
|
| 352 |
+
[1104.28 --> 1105.84] So can I ask you, you mentioned Redfish.
|
| 353 |
+
[1106.10 --> 1110.26] Jess, you and I both did a very good job of not exclaiming anything when you said Redfish.
|
| 354 |
+
[1110.50 --> 1112.24] I was more chill than you were.
|
| 355 |
+
[1112.74 --> 1113.04] Okay.
|
| 356 |
+
[1113.20 --> 1114.92] Now you're reflecting that back on me.
|
| 357 |
+
[1114.98 --> 1115.14] Fine.
|
| 358 |
+
[1115.58 --> 1117.06] Redfish does drive me crazy though.
|
| 359 |
+
[1117.34 --> 1118.58] I mean, is this just me?
|
| 360 |
+
[1118.58 --> 1124.32] I think because Redfish says, hey, the way we're going to manage these BMCs is we're going to hang them out over the internet.
|
| 361 |
+
[1124.62 --> 1127.46] And then we discover vulnerability and it's like, or we're hanging them out over the network rather.
|
| 362 |
+
[1127.88 --> 1129.60] And then we discover any of these vulnerabilities.
|
| 363 |
+
[1129.60 --> 1132.96] And it feels like the vendors say like, well, you shouldn't have had your BMC hanging out over the network.
|
| 364 |
+
[1133.04 --> 1135.08] It's like, you told me to do that, to take advantage of this functionality.
|
| 365 |
+
[1135.60 --> 1135.70] Yeah.
|
| 366 |
+
[1135.74 --> 1137.20] There's sort of this back and forth, right?
|
| 367 |
+
[1137.22 --> 1141.66] So keep in mind that IPMI was created by Intel, like I said, back in the nineties.
|
| 368 |
+
[1142.70 --> 1144.52] No standards have intelligent in the name.
|
| 369 |
+
[1145.00 --> 1145.28] Sorry.
|
| 370 |
+
[1145.56 --> 1146.20] It was the nineties.
|
| 371 |
+
[1146.40 --> 1147.16] I guess it was the nineties.
|
| 372 |
+
[1147.16 --> 1147.88] Does that feel intelligent?
|
| 373 |
+
[1147.88 --> 1150.28] It's never going to feel intelligent in retrospect.
|
| 374 |
+
[1150.64 --> 1153.52] But in the nineties, it was intelligent.
|
| 375 |
+
[1153.68 --> 1155.08] I'm not sure it was intelligent in the nineties.
|
| 376 |
+
[1155.42 --> 1159.48] The nineties, there were, anyway, maybe we had the standards of like NT on MIPS.
|
| 377 |
+
[1159.58 --> 1160.00] It was intelligent.
|
| 378 |
+
[1160.26 --> 1160.40] Sorry.
|
| 379 |
+
[1161.86 --> 1162.22] IPMI.
|
| 380 |
+
[1162.46 --> 1162.88] I'm sorry.
|
| 381 |
+
[1163.24 --> 1163.50] Yeah.
|
| 382 |
+
[1163.74 --> 1169.62] So with Redfish, it's actually being developed from DMTF, which is a proper standards organization,
|
| 383 |
+
[1169.62 --> 1175.14] but it kind of suffers the same fate, which is you have a standards group that meets and as
|
| 384 |
+
[1175.14 --> 1179.48] a committee amongst many different companies, they develop the standard for what it should
|
| 385 |
+
[1179.48 --> 1183.64] be able to do as far as system management, but they don't actually produce a reference
|
| 386 |
+
[1183.64 --> 1185.34] implementation of this, right?
|
| 387 |
+
[1185.34 --> 1188.22] They publish the spec and then they wait for people to actually implement it.
|
| 388 |
+
[1188.22 --> 1193.30] So Redfish is actually something where it's been around for years, at least five.
|
| 389 |
+
[1193.66 --> 1196.18] Only now are implementation starting to show up.
|
| 390 |
+
[1196.28 --> 1199.54] And if you look at what the state of the art was when they began the standards process,
|
| 391 |
+
[1199.72 --> 1203.72] it was clear that they just sort of took the checklist of where the industry like had
|
| 392 |
+
[1203.72 --> 1210.72] gone or where the rest of computing had gone to microservices and RESTful APIs and
|
| 393 |
+
[1210.72 --> 1211.42] things like that.
|
| 394 |
+
[1211.42 --> 1216.08] And they were looking at the pitfalls of IPMI, where most of the issues with IPMI were actually
|
| 395 |
+
[1216.08 --> 1219.28] around the security model and the data interchange.
|
| 396 |
+
[1219.64 --> 1222.92] Like you had to write, it was all a custom binary format.
|
| 397 |
+
[1223.60 --> 1227.86] So on one hand, it's like, okay, I understand you're trying to solve the problem by moving
|
| 398 |
+
[1227.86 --> 1231.26] towards standardized interfaces that everybody's like well vetted.
|
| 399 |
+
[1231.82 --> 1236.64] But on the other hand, now my BMC, I mean, it was going to run an HTTP server anyway, but
|
| 400 |
+
[1236.64 --> 1238.14] now I have to run a JSON parser.
|
| 401 |
+
[1238.38 --> 1240.78] There's a whole host of issues that that opens up.
|
| 402 |
+
[1240.78 --> 1246.40] And then now you also have to do things like WebSockets because the IKVM support is not
|
| 403 |
+
[1246.40 --> 1247.72] going to run over a custom protocol.
|
| 404 |
+
[1247.86 --> 1248.80] It's going to run over HTTP.
|
| 405 |
+
[1249.14 --> 1250.08] So you've got to do something.
|
| 406 |
+
[1250.56 --> 1251.74] It's easy to get a lot of that stuff wrong.
|
| 407 |
+
[1251.88 --> 1253.28] It's easy to get it wrong.
|
| 408 |
+
[1253.46 --> 1258.50] And like they're hedging on that they can use existing implementations that have been
|
| 409 |
+
[1258.50 --> 1259.10] well-tested.
|
| 410 |
+
[1259.60 --> 1264.30] The problem is, is that usually those well-tested implementations don't fit on a BMC, right?
|
| 411 |
+
[1264.32 --> 1266.80] You're dealing with a very constrained resource.
|
| 412 |
+
[1266.80 --> 1272.06] The other side of it was they really didn't, the folks that were specking out this protocol
|
| 413 |
+
[1272.06 --> 1277.52] really didn't talk with actual deployment, like operators that are doing real deployments.
|
| 414 |
+
[1277.74 --> 1278.56] Who did they talk with?
|
| 415 |
+
[1278.88 --> 1279.50] Well, they-
|
| 416 |
+
[1279.50 --> 1281.26] This is a question I frequently have when I look at Redfish.
|
| 417 |
+
[1281.34 --> 1282.16] Who did we talk to?
|
| 418 |
+
[1282.22 --> 1285.86] Well, they have a lot of representatives from the actual system manufacturers and system
|
| 419 |
+
[1285.86 --> 1286.30] integrators.
|
| 420 |
+
[1286.30 --> 1287.30] I know they talk with themselves.
|
| 421 |
+
[1287.64 --> 1292.26] And they also, you know, they talk with the BMC chip vendors and they talk with some of
|
| 422 |
+
[1292.26 --> 1295.98] the larger like consulting companies that actually do deployments, right?
|
| 423 |
+
[1296.02 --> 1298.96] There are companies that will just like build a data center for you and then hand it off.
|
| 424 |
+
[1299.26 --> 1302.44] So they talked to some of those, but they didn't really think about or interact with
|
| 425 |
+
[1302.44 --> 1304.70] the people who actively-
|
| 426 |
+
[1304.70 --> 1305.22] Use this stuff.
|
| 427 |
+
[1305.22 --> 1306.62] Operate data centers.
|
| 428 |
+
[1306.86 --> 1311.94] And they weren't really looking at what the existing industry practices were like at the
|
| 429 |
+
[1311.94 --> 1312.66] hyperscale, right?
|
| 430 |
+
[1312.66 --> 1317.06] Where this is solved because IPMI was such a terrible solution, but it's solved where
|
| 431 |
+
[1317.06 --> 1320.68] every vendor like hyperscaler has done their own implementation.
|
| 432 |
+
[1321.22 --> 1325.44] It turns out that one of the things that really drives everybody batty about Redfish is actually
|
| 433 |
+
[1325.44 --> 1327.64] the host to BMC interface.
|
| 434 |
+
[1328.16 --> 1330.88] Because on IPMI, it was well-defined.
|
| 435 |
+
[1330.96 --> 1335.04] You speak the same protocol, but you do it over an interface that makes sense.
|
| 436 |
+
[1335.10 --> 1340.36] So you might use, you know, LPC or I2C or one of these like system buses to actually
|
| 437 |
+
[1340.36 --> 1341.66] talk from the host down.
|
| 438 |
+
[1341.66 --> 1343.56] Well, in Redfish, they're like, well-
|
| 439 |
+
[1343.56 --> 1344.22] Just go over the network.
|
| 440 |
+
[1344.38 --> 1348.48] We built the entire security model and design around an HTTP endpoint.
|
| 441 |
+
[1348.62 --> 1348.80] Yeah.
|
| 442 |
+
[1348.92 --> 1354.50] So clearly, if you want your BIOS on your host system to be able to read the temperature
|
| 443 |
+
[1354.50 --> 1360.54] sensors to show it on the screen when it's booting, you have to implement a USB networking
|
| 444 |
+
[1360.54 --> 1366.54] stack to be able to send an HTTP request and then parse the JSON result that comes back
|
| 445 |
+
[1366.54 --> 1368.14] to show this information.
|
| 446 |
+
[1368.34 --> 1369.64] This does not feel like progress.
|
| 447 |
+
[1369.64 --> 1371.72] This feels like a big step in the wrong direction.
|
| 448 |
+
[1371.80 --> 1375.04] Especially then, I mean, even absent vulnerabilities, it sounds incredibly complicated.
|
| 449 |
+
[1375.10 --> 1375.88] Then you have all these vulnerabilities.
|
| 450 |
+
[1376.46 --> 1376.58] Right.
|
| 451 |
+
[1376.86 --> 1381.70] So the firmware world has gotten very, very complex as we keep shifting more and more
|
| 452 |
+
[1381.70 --> 1386.94] complexity down into system firmware as we are building, you know, bigger and more complex
|
| 453 |
+
[1386.94 --> 1387.34] systems.
|
| 454 |
+
[1387.34 --> 1388.12] All right.
|
| 455 |
+
[1388.12 --> 1391.34] So we are going to take a quick break and then we are going to come back.
|
| 456 |
+
[1391.40 --> 1393.86] And I think Jess is going to want to ask you some more love stories, Rick.
|
| 457 |
+
[1394.32 --> 1396.28] After a quick break and a word from our sponsor.
|
| 458 |
+
[1397.60 --> 1400.84] On the Metal is brought to you by the Oxide Computer Company.
|
| 459 |
+
[1401.04 --> 1402.60] Wait, did you say computer company, Jess?
|
| 460 |
+
[1402.68 --> 1403.36] Yes, indeed.
|
| 461 |
+
[1403.54 --> 1404.18] But wait a minute.
|
| 462 |
+
[1404.24 --> 1405.28] Everyone runs in the public cloud.
|
| 463 |
+
[1405.42 --> 1407.62] Jeff Bezos owns and operates every computer on the planet.
|
| 464 |
+
[1407.78 --> 1409.64] Why would anyone start a computer company?
|
| 465 |
+
[1409.84 --> 1410.90] That is so not true.
|
| 466 |
+
[1410.90 --> 1416.44] I have spent a bunch of time talking to folks who are still running on premises and actually
|
| 467 |
+
[1416.44 --> 1421.66] like the consensus among all of them is just a feeling of neglect because everyone thinks
|
| 468 |
+
[1421.66 --> 1423.96] that like everything is moving to the public cloud, but it's not.
|
| 469 |
+
[1424.24 --> 1427.08] If you're still running on premises, it's because you haven't heard of the cloud, right?
|
| 470 |
+
[1427.30 --> 1433.14] No, there are really good reasons for running on premises still for security, for latency,
|
| 471 |
+
[1433.38 --> 1435.20] strategic reasons for your business.
|
| 472 |
+
[1435.56 --> 1435.82] Wow.
|
| 473 |
+
[1435.90 --> 1439.34] The people running on premises must feel like everyone has ignored them.
|
| 474 |
+
[1439.62 --> 1440.06] They do.
|
| 475 |
+
[1440.06 --> 1440.52] Indeed.
|
| 476 |
+
[1440.52 --> 1446.28] So if this is you, please head on over to our website, Oxide.computer, sign up for our
|
| 477 |
+
[1446.28 --> 1449.50] mailing list and we would love to get in touch and hear your stories.
|
| 478 |
+
[1449.64 --> 1454.06] We acknowledge that you exist and you've got some really hard technical problems that we're
|
| 479 |
+
[1454.06 --> 1454.36] solving.
|
| 480 |
+
[1454.54 --> 1456.08] Oxide.computer, come join us.
|
| 481 |
+
[1458.26 --> 1459.86] All right, we're back.
|
| 482 |
+
[1460.62 --> 1462.38] Jess, love story questions.
|
| 483 |
+
[1462.66 --> 1462.88] Okay.
|
| 484 |
+
[1463.56 --> 1470.18] This one's more, it's on the border of a nightmare and a love story though, but it is the
|
| 485 |
+
[1470.18 --> 1475.38] most interesting or like the weirdest bug you've ever found.
|
| 486 |
+
[1475.38 --> 1477.42] Oh, so many stories here.
|
| 487 |
+
[1477.78 --> 1478.92] Hard to choose again.
|
| 488 |
+
[1479.06 --> 1482.66] I mean, for someone who's been in the places that Rick has been, this is like legitimately
|
| 489 |
+
[1482.66 --> 1483.72] triggering potentially.
|
| 490 |
+
[1484.48 --> 1486.98] It could be, but we're going to stay away from those.
|
| 491 |
+
[1486.98 --> 1487.30] Okay.
|
| 492 |
+
[1487.40 --> 1487.80] That's good.
|
| 493 |
+
[1488.02 --> 1489.52] That's good on doctor's orders.
|
| 494 |
+
[1489.68 --> 1492.78] And I'm going to limit it to the ones that I actually had direct involvement with because
|
| 495 |
+
[1492.78 --> 1496.58] there's one that is absolutely fascinating, but it wasn't actually something that I've
|
| 496 |
+
[1496.58 --> 1497.14] only no second.
|
| 497 |
+
[1497.14 --> 1504.36] Probably my worst one was actually a bug that I inherited.
|
| 498 |
+
[1504.82 --> 1510.12] So I joined a team and I was on their hardware team doing firmware work.
|
| 499 |
+
[1510.42 --> 1514.58] And I got this bug assigned to me because I was now the owner of that system.
|
| 500 |
+
[1514.88 --> 1516.46] And I looked at how old the bug was.
|
| 501 |
+
[1516.56 --> 1517.60] The bug was five years old.
|
| 502 |
+
[1517.78 --> 1521.96] The bug was that the machine would not stay powered off.
|
| 503 |
+
[1522.80 --> 1523.24] What?
|
| 504 |
+
[1523.24 --> 1526.76] So you, you would log into the machine.
|
| 505 |
+
[1526.94 --> 1528.04] You would run shutdown.
|
| 506 |
+
[1528.46 --> 1530.38] It would go through the whole shutdown sequence.
|
| 507 |
+
[1530.58 --> 1532.04] The power would actually turn off.
|
| 508 |
+
[1532.48 --> 1536.28] And about two seconds later, it would turn right back on and go right back up.
|
| 509 |
+
[1536.42 --> 1537.38] That's brutal to the bug.
|
| 510 |
+
[1538.86 --> 1539.70] It's a ghost.
|
| 511 |
+
[1540.12 --> 1540.42] Zombie.
|
| 512 |
+
[1540.46 --> 1543.64] Well, no, you just think about like, how do you go debug something like that?
|
| 513 |
+
[1544.24 --> 1545.16] I mean, that is, that is.
|
| 514 |
+
[1545.18 --> 1545.84] Because it's off.
|
| 515 |
+
[1545.90 --> 1546.78] Well, it's off.
|
| 516 |
+
[1547.24 --> 1548.12] But it's not off.
|
| 517 |
+
[1548.38 --> 1548.84] Air quotes.
|
| 518 |
+
[1548.98 --> 1549.90] It's clearly not off.
|
| 519 |
+
[1550.14 --> 1551.84] Well, and so thankfully.
|
| 520 |
+
[1551.84 --> 1552.52] Or maybe it was off.
|
| 521 |
+
[1552.52 --> 1557.80] Thankfully, the people in the past who had dealt with this bug understood what the actual issue was.
|
| 522 |
+
[1557.86 --> 1558.88] They had done all the hard work.
|
| 523 |
+
[1559.40 --> 1563.56] So I mostly just got to live through the, oh, this is one of those.
|
| 524 |
+
[1565.14 --> 1569.72] It turns out that that particular generation of chipset.
|
| 525 |
+
[1570.00 --> 1575.48] So the actual like Northbridge and Southbridge had a SMBus controller on it, as they usually do.
|
| 526 |
+
[1575.96 --> 1577.42] And SMBus has an alert pin.
|
| 527 |
+
[1577.42 --> 1586.06] And the concept of the alert pin is that your SMBus slave devices can raise the alert pin to cause an interrupt and say, hey, something happened.
|
| 528 |
+
[1586.22 --> 1586.96] All fine and good.
|
| 529 |
+
[1587.22 --> 1592.50] Well, in this particular version of the chipset, that was a non-maskable interrupt.
|
| 530 |
+
[1592.70 --> 1595.22] It will actually wake the system from sleep if it gets asserted.
|
| 531 |
+
[1595.22 --> 1598.78] Now, that normally wouldn't be an issue.
|
| 532 |
+
[1599.28 --> 1602.36] What would actually be causing the alert pin to go high?
|
| 533 |
+
[1602.36 --> 1611.00] Well, it turns out that the employer at that time was also making their own custom power supplies that were built off of, they would run off utility power.
|
| 534 |
+
[1611.10 --> 1614.22] But they also had a battery charger and automated cutover.
|
| 535 |
+
[1614.32 --> 1616.02] So they had sort of a built-in UPS in them.
|
| 536 |
+
[1616.02 --> 1620.62] And it makes perfect sense that if you lose power, you want to have an alert, of course.
|
| 537 |
+
[1620.62 --> 1625.88] You want to set the alert and notify the host system that something's happening, but nothing clears it.
|
| 538 |
+
[1626.04 --> 1636.46] So the very first time you do a cutover from utility power to battery, doesn't matter when that happens, it would set this alert flag and then it would never clear it.
|
| 539 |
+
[1636.46 --> 1641.52] And so when the machine went to sleep, the alert was set high, it would NMI, and it would wake the system back up.
|
| 540 |
+
[1641.88 --> 1644.04] Which is just this whole Rube Goldberg machine, right?
|
| 541 |
+
[1644.36 --> 1649.76] Now, the reason this got to be really bad was you have to do testing on your batteries to actually make sure they work.
|
| 542 |
+
[1650.18 --> 1658.18] So you were guaranteed that at least once a month you were going to go from utility power to battery power, which meant this bug would show up often.
|
| 543 |
+
[1658.52 --> 1658.60] Right.
|
| 544 |
+
[1658.80 --> 1661.72] And I assume this was dead reproducible or nearly.
|
| 545 |
+
[1661.80 --> 1662.20] It should have been.
|
| 546 |
+
[1662.32 --> 1665.90] Once you understood what was happening, it's really easy to reproduce.
|
| 547 |
+
[1665.90 --> 1668.40] Like no machine would stay powered off, it sounds like.
|
| 548 |
+
[1668.48 --> 1669.96] Well, only that particular generation.
|
| 549 |
+
[1670.40 --> 1677.48] For that particular generation, because, you know, I've often thought that bugs may be psychotic or non-reproducible, but not both.
|
| 550 |
+
[1677.84 --> 1678.02] Yeah.
|
| 551 |
+
[1678.24 --> 1680.58] And that's a psychotic bug, but it's dead reproducible.
|
| 552 |
+
[1680.80 --> 1680.90] Yeah.
|
| 553 |
+
[1681.12 --> 1684.18] Because presumably it took an analyzer or, I mean.
|
| 554 |
+
[1684.52 --> 1687.00] Oh, I mean, it took six months for people to figure out.
|
| 555 |
+
[1687.00 --> 1687.28] Right.
|
| 556 |
+
[1687.34 --> 1687.72] Originally.
|
| 557 |
+
[1688.26 --> 1692.92] Now, my favorite part of this is actually recognizing that there is no way to fix this.
|
| 558 |
+
[1693.98 --> 1694.42] Right.
|
| 559 |
+
[1694.42 --> 1701.22] The series of events and design decisions were locked into the hardware and it was always going to do this.
|
| 560 |
+
[1701.88 --> 1703.02] So what is your fix?
|
| 561 |
+
[1703.26 --> 1704.86] Because you do want the machines to shut off.
|
| 562 |
+
[1705.48 --> 1705.72] Anybody?
|
| 563 |
+
[1706.34 --> 1707.22] I don't know.
|
| 564 |
+
[1707.42 --> 1707.60] Yeah.
|
| 565 |
+
[1707.74 --> 1713.14] And I feel that we in low-level software are often asked to make up for hardware sins.
|
| 566 |
+
[1713.40 --> 1716.40] I feel like we're often engaged in kind of a cover-up, honestly.
|
| 567 |
+
[1716.62 --> 1716.94] Always.
|
| 568 |
+
[1717.22 --> 1717.88] A cover-up.
|
| 569 |
+
[1718.00 --> 1718.10] Yeah.
|
| 570 |
+
[1718.10 --> 1719.08] It feels that way.
|
| 571 |
+
[1719.08 --> 1721.96] It feels like, look, we have to like, look, dad's a drunk.
|
| 572 |
+
[1722.12 --> 1722.40] Okay.
|
| 573 |
+
[1722.44 --> 1724.86] So we just have to like, can we pretend like we're a normal family?
|
| 574 |
+
[1725.02 --> 1725.42] Look alive.
|
| 575 |
+
[1725.58 --> 1726.18] Look alive.
|
| 576 |
+
[1726.28 --> 1726.82] It's like, no.
|
| 577 |
+
[1726.94 --> 1728.16] Well, in this case, look dead.
|
| 578 |
+
[1728.88 --> 1729.24] Yeah.
|
| 579 |
+
[1729.60 --> 1730.44] Everybody look dead.
|
| 580 |
+
[1730.44 --> 1736.10] But this is a tough one to cover up because you, yeah.
|
| 581 |
+
[1736.16 --> 1736.80] How did you do it?
|
| 582 |
+
[1736.80 --> 1737.20] I don't know.
|
| 583 |
+
[1737.54 --> 1741.60] You got the, because NMI, I mean, NMI, there's a reason they call it NMI.
|
| 584 |
+
[1741.84 --> 1742.72] It's non-maskable.
|
| 585 |
+
[1742.90 --> 1743.10] Yeah.
|
| 586 |
+
[1743.10 --> 1747.16] But see, it's really caused by that, that power supply flag setting the alert.
|
| 587 |
+
[1747.34 --> 1747.60] Okay.
|
| 588 |
+
[1747.68 --> 1749.42] And it was possible to clear that from software.
|
| 589 |
+
[1749.64 --> 1749.98] Ah, okay.
|
| 590 |
+
[1750.40 --> 1752.92] So a cron job calls out and clears that once a day.
|
| 591 |
+
[1753.38 --> 1754.12] Oh my gosh.
|
| 592 |
+
[1754.16 --> 1754.44] Nice.
|
| 593 |
+
[1754.66 --> 1758.04] And so the bug comes back when somebody actually removes that cron job because they say,
|
| 594 |
+
[1758.12 --> 1759.16] why do you have this cron job?
|
| 595 |
+
[1759.66 --> 1763.18] And so, and so the cron job would set the bit on the power supply.
|
| 596 |
+
[1763.54 --> 1763.70] Yeah.
|
| 597 |
+
[1763.72 --> 1766.84] It would clear the flag that, that power had failed.
|
| 598 |
+
[1767.04 --> 1767.96] That power had failed.
|
| 599 |
+
[1768.38 --> 1768.94] Got it.
|
| 600 |
+
[1768.94 --> 1773.26] And so, so now if you managed, if you powered it off twice in the same day,
|
| 601 |
+
[1773.34 --> 1774.58] wouldn't you also see the problem or no?
|
| 602 |
+
[1774.84 --> 1779.72] Uh, only if you, I mean, if you had lost utility power at some point, that flag would get set.
|
| 603 |
+
[1779.84 --> 1780.02] Right.
|
| 604 |
+
[1780.36 --> 1781.20] And it would stay set.
|
| 605 |
+
[1781.30 --> 1782.62] And so you wouldn't be able to turn off.
|
| 606 |
+
[1782.72 --> 1782.92] Right.
|
| 607 |
+
[1783.08 --> 1785.32] But as soon as the, as soon as the cron job fires,
|
| 608 |
+
[1785.42 --> 1787.94] as soon as the cron job fires, it's going to clear that flag and then you're good to go.
|
| 609 |
+
[1787.96 --> 1790.52] And where is the, is the cron job, where's the cron job running?
|
| 610 |
+
[1790.96 --> 1793.42] Is it running on the BMC?
|
| 611 |
+
[1793.54 --> 1794.18] Oh, no, no, no.
|
| 612 |
+
[1794.30 --> 1795.54] This, this was running on the host.
|
| 613 |
+
[1795.80 --> 1795.98] Yeah.
|
| 614 |
+
[1796.20 --> 1796.76] Okay, nice.
|
| 615 |
+
[1796.76 --> 1797.06] Yeah.
|
| 616 |
+
[1797.06 --> 1800.24] That is one where you'd be like, so this is just like someone logs into this machine.
|
| 617 |
+
[1800.34 --> 1802.24] It's like cron tab minus L and there's that.
|
| 618 |
+
[1802.34 --> 1802.82] There it is.
|
| 619 |
+
[1802.96 --> 1803.04] Right.
|
| 620 |
+
[1803.12 --> 1807.88] I mean, that's like all the machines that I've ever used were obviously pets because people
|
| 621 |
+
[1807.88 --> 1810.98] would log in and be like, oh, there are some cron jobs just set up.
|
| 622 |
+
[1811.42 --> 1813.04] Like, yeah.
|
| 623 |
+
[1813.14 --> 1817.14] It's hard to know if this is a pet or a cattle, but that's a cron job.
|
| 624 |
+
[1817.34 --> 1818.64] You think that like, what was that name?
|
| 625 |
+
[1818.70 --> 1819.90] You think you would want to name it?
|
| 626 |
+
[1819.92 --> 1821.32] Like, do not delete me.
|
| 627 |
+
[1821.38 --> 1821.84] I'm serious.
|
| 628 |
+
[1821.92 --> 1822.50] You've been warned.
|
| 629 |
+
[1822.84 --> 1823.02] Yeah.
|
| 630 |
+
[1823.02 --> 1827.18] No, it was something more like PSU fix, which is always great, right?
|
| 631 |
+
[1827.18 --> 1829.00] Like, I don't even know what this is for.
|
| 632 |
+
[1829.24 --> 1830.46] But it's fixing something.
|
| 633 |
+
[1830.60 --> 1831.18] It's fixing something.
|
| 634 |
+
[1831.26 --> 1832.30] It's like, ah, we don't need that.
|
| 635 |
+
[1832.40 --> 1832.96] I deleted it.
|
| 636 |
+
[1833.00 --> 1833.42] It was fine.
|
| 637 |
+
[1833.54 --> 1833.72] Right.
|
| 638 |
+
[1833.88 --> 1834.84] It was fine.
|
| 639 |
+
[1835.04 --> 1835.24] Yeah.
|
| 640 |
+
[1835.44 --> 1836.42] Systems seem to work.
|
| 641 |
+
[1836.84 --> 1837.78] Until the next month.
|
| 642 |
+
[1837.98 --> 1838.16] Yeah.
|
| 643 |
+
[1838.20 --> 1839.56] That is, uh, that's grisly.
|
| 644 |
+
[1839.64 --> 1842.18] So was the cron job was, uh, was your doing?
|
| 645 |
+
[1842.76 --> 1843.66] Uh, no, actually.
|
| 646 |
+
[1843.72 --> 1844.76] I mean, somebody else had written that.
|
| 647 |
+
[1844.84 --> 1847.86] And so I just came up with better ways of making sure the cron job didn't get removed.
|
| 648 |
+
[1848.00 --> 1848.80] There you go.
|
| 649 |
+
[1849.02 --> 1849.38] Nice.
|
| 650 |
+
[1849.44 --> 1849.70] Nice.
|
| 651 |
+
[1849.70 --> 1850.76] That's good.
|
| 652 |
+
[1851.02 --> 1852.08] That's an important role.
|
| 653 |
+
[1852.88 --> 1857.58] Uh, so, and then how, in terms of, I mean, you have been up, down, and all around.
|
| 654 |
+
[1857.68 --> 1861.88] You've been at, I think, um, I mean, you've done disc controller stuff.
|
| 655 |
+
[1862.04 --> 1867.26] You've done, um, and I assume that at every level you've got horror stories.
|
| 656 |
+
[1867.48 --> 1867.92] Oh, yeah.
|
| 657 |
+
[1868.28 --> 1870.38] I mean, that's, that's the nature of hardware, right?
|
| 658 |
+
[1870.42 --> 1874.44] And, and firmware is, as you said, firmware is making up for, for the mistakes of hardware.
|
| 659 |
+
[1874.52 --> 1877.02] And there's always interesting things that have occurred.
|
| 660 |
+
[1877.02 --> 1881.08] Depending on which field you get into, you bring in more interesting failure modes.
|
| 661 |
+
[1881.26 --> 1884.42] But it's always the case of something will go wrong in the hardware.
|
| 662 |
+
[1884.62 --> 1886.24] And firmware is your first chance of fixing that.
|
| 663 |
+
[1886.34 --> 1891.08] You start moving higher up the stack as you work on situations that have to bring in more
|
| 664 |
+
[1891.08 --> 1894.18] data or more context to actually decide what to do in this failure scenario.
|
| 665 |
+
[1894.58 --> 1899.34] So I'm curious where you've had bugs where the hardware is in danger of physically destroying
|
| 666 |
+
[1899.34 --> 1901.10] itself if software doesn't do the right thing.
|
| 667 |
+
[1901.18 --> 1904.94] So the, at that level, this is not true for, for most of us that are even at the hardware
|
| 668 |
+
[1904.94 --> 1905.68] software interface.
|
| 669 |
+
[1906.18 --> 1908.82] Most of the time, hardware is pretty good at not destroying itself.
|
| 670 |
+
[1909.66 --> 1913.42] But I, I gotta believe that some of the layers you've been, that you, you've seen some things
|
| 671 |
+
[1913.42 --> 1914.48] where it's like, no, no, no.
|
| 672 |
+
[1914.50 --> 1918.66] If we don't actually, you know, certainly if you're writing firmware for a disk controller,
|
| 673 |
+
[1918.82 --> 1920.88] I mean, you can actually, a lot of bad things can happen.
|
| 674 |
+
[1921.74 --> 1922.84] Very bad things can happen.
|
| 675 |
+
[1922.84 --> 1927.92] There, there was actually a mailing list internally at one of my employers that was nothing but
|
| 676 |
+
[1927.92 --> 1931.26] pictures of equipment that had caught fire in the data center.
|
| 677 |
+
[1931.50 --> 1932.86] Whoa, that's dope.
|
| 678 |
+
[1932.86 --> 1934.80] That's like a Reddit that I would subscribe to.
|
| 679 |
+
[1934.98 --> 1935.04] Yeah.
|
| 680 |
+
[1935.16 --> 1935.30] Yeah.
|
| 681 |
+
[1935.32 --> 1938.54] And the thing was, it was like, there was always a new contribution about every week.
|
| 682 |
+
[1938.64 --> 1944.00] I mean, so the rate at which things would just burst into flames, um, was relatively
|
| 683 |
+
[1944.00 --> 1944.32] high.
|
| 684 |
+
[1944.40 --> 1946.88] Now you have to keep in mind, I, I used to work at hyperscalers.
|
| 685 |
+
[1946.88 --> 1953.70] And so like at the volume they work at, the probability of any event occurring is almost
|
| 686 |
+
[1953.70 --> 1957.30] certainty for this to be happening this often is just a consequence of scale.
|
| 687 |
+
[1957.72 --> 1960.44] But yeah, that, that was definitely a thing in the hard drive space.
|
| 688 |
+
[1960.44 --> 1961.70] There's a really interesting one though.
|
| 689 |
+
[1961.70 --> 1967.18] Uh, there was a particular generation of hard drives that there was a mistake in the actual
|
| 690 |
+
[1967.18 --> 1968.30] calculations they did.
|
| 691 |
+
[1968.94 --> 1973.44] So hard drives are really, really, really interesting because internally, right, you've got the platters
|
| 692 |
+
[1973.44 --> 1974.80] spinning at high speeds.
|
| 693 |
+
[1974.80 --> 1974.98] Yeah.
|
| 694 |
+
[1974.98 --> 1979.28] The heads are actually trying to move back on a servo motor, but the heads are actually
|
| 695 |
+
[1979.28 --> 1980.26] airfoils.
|
| 696 |
+
[1980.64 --> 1980.88] Yes.
|
| 697 |
+
[1981.12 --> 1987.02] They're relying on the speed of the air from the platter spinning to actually fly, literally
|
| 698 |
+
[1987.02 --> 1989.40] fly above the surface of the media.
|
| 699 |
+
[1989.40 --> 1994.48] So you have to do these calculations around number of impurities that you might expect
|
| 700 |
+
[1994.48 --> 1998.92] and what those heights would be to decide how high to fly the head, uh, to the disc
|
| 701 |
+
[1998.92 --> 2000.14] of the lifespan of the drive.
|
| 702 |
+
[2000.24 --> 2004.04] And there's a different height that you use for reading versus writing because you need
|
| 703 |
+
[2004.04 --> 2008.14] to be lower when you're writing so that you, uh, change the magnetism over a much smaller
|
| 704 |
+
[2008.14 --> 2008.54] area.
|
| 705 |
+
[2008.54 --> 2013.64] And it turned out that they had just, they had taken the previous generation and like
|
| 706 |
+
[2013.64 --> 2018.92] the calculation from that and, and like multiplied it by two, but they forgot that it was an error
|
| 707 |
+
[2018.92 --> 2020.94] bound like both directions.
|
| 708 |
+
[2021.16 --> 2021.44] Uh-oh.
|
| 709 |
+
[2021.68 --> 2028.08] And so they ended up making the error very tight for the actual fly height of the head.
|
| 710 |
+
[2028.26 --> 2028.56] Wow.
|
| 711 |
+
[2028.98 --> 2029.30] So.
|
| 712 |
+
[2029.48 --> 2034.04] And the fly heights, it should be said, these are very, very small distances.
|
| 713 |
+
[2034.16 --> 2034.38] Yeah.
|
| 714 |
+
[2034.72 --> 2034.96] Yeah.
|
| 715 |
+
[2035.06 --> 2035.78] I mean, we're talking.
|
| 716 |
+
[2035.78 --> 2039.64] The number that was given to me that I honestly still don't believe.
|
| 717 |
+
[2039.74 --> 2043.46] So if you tell me this is false, I, but the number was given to me by someone who works
|
| 718 |
+
[2043.46 --> 2048.14] for a disc manufacturer, the fly height during a write 0.8 nanometers.
|
| 719 |
+
[2048.44 --> 2049.08] Yeah, that's about right.
|
| 720 |
+
[2049.38 --> 2051.08] That is, that's insane.
|
| 721 |
+
[2051.20 --> 2052.22] That's 800 picometers.
|
| 722 |
+
[2053.30 --> 2054.38] That's insanity.
|
| 723 |
+
[2055.08 --> 2055.24] Yeah.
|
| 724 |
+
[2055.60 --> 2058.16] And especially as they've gotten into things like the helium filled drives.
|
| 725 |
+
[2058.28 --> 2058.46] Yeah.
|
| 726 |
+
[2058.54 --> 2058.68] Right.
|
| 727 |
+
[2058.98 --> 2059.18] Yeah.
|
| 728 |
+
[2059.18 --> 2060.28] This is definitely a helium filled drive.
|
| 729 |
+
[2060.46 --> 2063.78] They, because they ran into the problem of, uh, with natural air.
|
| 730 |
+
[2064.30 --> 2064.66] Molecules.
|
| 731 |
+
[2064.66 --> 2070.00] Well, the air turbulent, the turbulence effects of the platter spinning would actually cause
|
| 732 |
+
[2070.00 --> 2071.00] too much variation.
|
| 733 |
+
[2071.00 --> 2073.68] So you couldn't fly any closer because you'd be in the turbulent flow.
|
| 734 |
+
[2074.10 --> 2079.10] And so that's why they switched to helium was actually that helium has lower drag on the
|
| 735 |
+
[2079.10 --> 2079.36] surface.
|
| 736 |
+
[2079.52 --> 2080.78] And so you can actually fly closer.
|
| 737 |
+
[2081.40 --> 2081.54] Yeah.
|
| 738 |
+
[2081.76 --> 2083.88] So this thing was flying too close then.
|
| 739 |
+
[2083.90 --> 2084.82] It was flying too close.
|
| 740 |
+
[2084.82 --> 2085.30] Man.
|
| 741 |
+
[2085.30 --> 2092.14] And so in a hard drive, like you expect surface impurities and you expect to, to encounter
|
| 742 |
+
[2092.14 --> 2099.02] them with the drive head, AKA the head literally slamming into a mountain on the surface of the
|
| 743 |
+
[2099.02 --> 2101.46] disc at 7,200 RPM.
|
| 744 |
+
[2101.86 --> 2102.08] Right.
|
| 745 |
+
[2102.20 --> 2105.10] A mountain, a four nanometer high mountain.
|
| 746 |
+
[2105.34 --> 2105.58] Right.
|
| 747 |
+
[2105.78 --> 2107.78] I mean, just to put this in scale.
|
| 748 |
+
[2107.94 --> 2110.60] But I mean, from a visual perspective, right?
|
| 749 |
+
[2110.60 --> 2110.94] Yeah, exactly.
|
| 750 |
+
[2111.06 --> 2111.30] Absolutely.
|
| 751 |
+
[2111.30 --> 2113.42] The platter is supposed to be perfectly flat.
|
| 752 |
+
[2113.50 --> 2115.66] So these things are big like peaks.
|
| 753 |
+
[2116.16 --> 2116.18] Peaks.
|
| 754 |
+
[2116.30 --> 2118.48] And you're going to have this head smash into it.
|
| 755 |
+
[2118.76 --> 2123.28] And so the failure mode is actually that the head itself becomes deformed from repeated
|
| 756 |
+
[2123.28 --> 2123.84] impacts.
|
| 757 |
+
[2123.84 --> 2130.34] So this was where we had to like ask for electron microscope photography of the heads of failed
|
| 758 |
+
[2130.34 --> 2136.34] drives to root cause that in fact, yes, this really was a case where the fly height had
|
| 759 |
+
[2136.34 --> 2137.20] been miscalculated.
|
| 760 |
+
[2137.42 --> 2141.20] And so these drives were actually destroying their heads faster than anticipated.
|
| 761 |
+
[2141.38 --> 2146.28] But you raise a really interesting point about that you've got a hardware that is designed
|
| 762 |
+
[2146.28 --> 2149.36] for this kind of mechanical impact over and over again.
|
| 763 |
+
[2149.36 --> 2153.92] And it's actually hard to find the software defect, even though if the poor head could
|
| 764 |
+
[2153.92 --> 2155.80] speak, it would say, hey, you know what?
|
| 765 |
+
[2155.82 --> 2157.02] I'm running into a lot of things.
|
| 766 |
+
[2157.18 --> 2158.70] I'm running into more things than you would expect.
|
| 767 |
+
[2158.82 --> 2164.22] I mean, because it's so designed to have, you know, to be able to take another lap effectively
|
| 768 |
+
[2164.22 --> 2166.28] and hit the track that it missed.
|
| 769 |
+
[2166.50 --> 2166.68] Yeah.
|
| 770 |
+
[2166.74 --> 2169.62] I mean, this is kind of the nature of the beast when you're down this low in the firmware
|
| 771 |
+
[2169.62 --> 2174.74] stack is when you're truly interacting with the hardware at this level, you can usually
|
| 772 |
+
[2174.74 --> 2175.64] do dangerous things.
|
| 773 |
+
[2175.64 --> 2179.22] Um, and often the effects are hard to tell.
|
| 774 |
+
[2179.44 --> 2184.24] So it's the same as when you're building like critical systems for industrial control or
|
| 775 |
+
[2184.24 --> 2186.12] spacecraft or whatever, right?
|
| 776 |
+
[2186.18 --> 2191.14] There's always this case where I'm dealing with hardware that has some effect that I can
|
| 777 |
+
[2191.14 --> 2195.42] only measure through my sensors and it has a risk to human safety.
|
| 778 |
+
[2195.62 --> 2197.02] So what do I do?
|
| 779 |
+
[2197.42 --> 2199.32] Um, and how do you design that hardware?
|
| 780 |
+
[2199.32 --> 2205.02] And often the tendency has been to shift more and more control over to the software and assume
|
| 781 |
+
[2205.02 --> 2206.40] that the software can do the right job.
|
| 782 |
+
[2206.52 --> 2210.30] And there's always this back and forth of, well, in some cases that's fine, right?
|
| 783 |
+
[2210.32 --> 2214.36] Like in the hard drive case, yes, having the fly height controlled by the firmware makes
|
| 784 |
+
[2214.36 --> 2214.72] sense.
|
| 785 |
+
[2215.12 --> 2218.46] But when you get it wrong, it's going to have massive effects.
|
| 786 |
+
[2218.58 --> 2222.86] On the other hand, when you have like actual, you know, critical devices, you actually want
|
| 787 |
+
[2222.86 --> 2227.06] to build the hardware in a totally different way so that if there's a failure in the software,
|
| 788 |
+
[2227.06 --> 2229.26] that it's intrinsically safe, right?
|
| 789 |
+
[2229.32 --> 2231.92] It cannot possibly do the thing that would be dangerous.
|
| 790 |
+
[2232.04 --> 2234.40] Unfortunately, these are not actually dangerous to people, but they're definitely dangerous
|
| 791 |
+
[2234.40 --> 2234.88] to the device.
|
| 792 |
+
[2234.96 --> 2236.74] At least the failures you're talking about.
|
| 793 |
+
[2237.02 --> 2237.28] Yeah.
|
| 794 |
+
[2237.36 --> 2240.46] It just happens to be that I've worked on other systems where it was much more about not
|
| 795 |
+
[2240.46 --> 2241.08] hurting humans.
|
| 796 |
+
[2241.30 --> 2241.64] Yeah.
|
| 797 |
+
[2241.86 --> 2242.90] We definitely shouldn't hurt the humans.
|
| 798 |
+
[2243.42 --> 2246.34] Are you familiar, Jesse, are you familiar with the legend of the walking drives?
|
| 799 |
+
[2246.90 --> 2247.26] No.
|
| 800 |
+
[2247.52 --> 2247.70] Do you?
|
| 801 |
+
[2248.00 --> 2248.22] Oh, yeah.
|
| 802 |
+
[2248.22 --> 2248.82] Yeah, right.
|
| 803 |
+
[2248.82 --> 2256.32] So where the drives, when you would have a spindle that was sufficiently large, there's enough
|
| 804 |
+
[2256.32 --> 2257.44] angular momentum on that thing.
|
| 805 |
+
[2257.52 --> 2262.66] You can actually make the drive lurch by seeking to the same location over and over again.
|
| 806 |
+
[2262.84 --> 2265.78] And this is when like a hard drive was the size of a washing machine.
|
| 807 |
+
[2265.94 --> 2266.16] Right.
|
| 808 |
+
[2266.30 --> 2267.10] Oh, geez.
|
| 809 |
+
[2267.72 --> 2271.20] So there is actually, there's a great story in the new hacker's dictionary, which I hate
|
| 810 |
+
[2271.20 --> 2271.40] to.
|
| 811 |
+
[2271.58 --> 2272.82] I have that inside.
|
| 812 |
+
[2272.98 --> 2275.26] I hate to encourage people to buy it because he's such a.
|
| 813 |
+
[2275.52 --> 2276.00] It's good.
|
| 814 |
+
[2276.44 --> 2276.90] It is good.
|
| 815 |
+
[2276.90 --> 2279.58] It's a good, it's a good book written by a bad man.
|
| 816 |
+
[2279.66 --> 2281.80] No, just it's written by he who must not be named.
|
| 817 |
+
[2281.92 --> 2282.84] He who must not be named.
|
| 818 |
+
[2282.84 --> 2287.80] But there's a good story in there about hackers getting into a Xerox machine and making the
|
| 819 |
+
[2287.80 --> 2288.42] drives walk.
|
| 820 |
+
[2289.36 --> 2291.90] So Rick, let's talk about the future of firmware a little bit.
|
| 821 |
+
[2292.12 --> 2295.46] Because, you know, Jess and I were at, and Steve, you missed the open source firmware
|
| 822 |
+
[2295.46 --> 2295.98] conference.
|
| 823 |
+
[2296.40 --> 2297.82] But Jess and I went down there.
|
| 824 |
+
[2297.84 --> 2298.70] It was a lot of fun.
|
| 825 |
+
[2298.98 --> 2299.34] It was dope.
|
| 826 |
+
[2300.04 --> 2302.68] It was like an old school conference.
|
| 827 |
+
[2302.94 --> 2305.14] Like people actually wanted to help each other.
|
| 828 |
+
[2305.26 --> 2306.12] People were nice.
|
| 829 |
+
[2306.34 --> 2308.42] There wasn't like vendor booths everywhere.
|
| 830 |
+
[2308.72 --> 2310.38] Like it was like actually legit.
|
| 831 |
+
[2310.60 --> 2311.22] I thought it was great.
|
| 832 |
+
[2311.22 --> 2312.52] Yeah, I mean, Rick, is that your read too?
|
| 833 |
+
[2312.56 --> 2313.46] I thought that was a lot of fun.
|
| 834 |
+
[2313.68 --> 2314.06] Yeah, yeah.
|
| 835 |
+
[2314.14 --> 2318.48] I mean, it's definitely a conference that is by and for a particular group of developers.
|
| 836 |
+
[2318.96 --> 2322.12] And it's not become a commercial venue.
|
| 837 |
+
[2322.32 --> 2323.74] It's not become a commercial venue.
|
| 838 |
+
[2324.22 --> 2325.02] And I don't know, it did.
|
| 839 |
+
[2325.10 --> 2329.66] I was telling Jess when I said this did kind of date back to an era when you had no other
|
| 840 |
+
[2329.66 --> 2333.84] way of connecting to people in your incredibly small demographic.
|
| 841 |
+
[2334.60 --> 2338.84] Other than, and you would kind of walk in the room like, oh my God, there are 200 people
|
| 842 |
+
[2338.84 --> 2340.86] here that are interested in open source firmware.
|
| 843 |
+
[2340.86 --> 2341.46] I thought it was great.
|
| 844 |
+
[2341.52 --> 2342.06] It was dope.
|
| 845 |
+
[2342.44 --> 2343.58] It was so much fun.
|
| 846 |
+
[2343.78 --> 2348.76] And tons of people, the BMC track had like, I mean, the track that you were in, like 100
|
| 847 |
+
[2348.76 --> 2349.36] people in there.
|
| 848 |
+
[2349.72 --> 2349.98] Yeah.
|
| 849 |
+
[2350.16 --> 2350.38] Yeah.
|
| 850 |
+
[2350.38 --> 2352.16] And a lot of that's all relatively recent.
|
| 851 |
+
[2352.38 --> 2357.96] I mean, the idea of an open source BMC stack actually only came about like two and
|
| 852 |
+
[2357.96 --> 2358.84] a half, three years ago.
|
| 853 |
+
[2359.46 --> 2361.08] And it seems like a lot of interest from a lot of folks.
|
| 854 |
+
[2361.20 --> 2363.18] So what's your thinking on the future of open source firmware?
|
| 855 |
+
[2363.36 --> 2367.38] I feel that we're kind of on the cusp of this becoming real.
|
| 856 |
+
[2367.44 --> 2367.84] What do you think?
|
| 857 |
+
[2367.84 --> 2368.74] I think so.
|
| 858 |
+
[2369.42 --> 2374.62] It's going to be different as we talk about different devices that use firmware.
|
| 859 |
+
[2375.00 --> 2380.48] Like there's EFI for an x86 system and has a large amount of it that's been open since
|
| 860 |
+
[2380.48 --> 2382.54] the very beginning, but portions of it aren't.
|
| 861 |
+
[2382.76 --> 2386.04] And similar to BMCs, BMCs are getting to be more open.
|
| 862 |
+
[2386.40 --> 2391.28] There's this fine line of happening between firmware being open source, but there being
|
| 863 |
+
[2391.28 --> 2395.80] aspects of it that are closed because of the security concerns or the security model around
|
| 864 |
+
[2395.80 --> 2396.02] it.
|
| 865 |
+
[2396.02 --> 2401.68] So you might need a signing key that you can't get unless you have an NDA, which kind of
|
| 866 |
+
[2401.68 --> 2401.92] sucks.
|
| 867 |
+
[2402.52 --> 2406.00] And so there's other things like, you know, can I actually get the data sheets to even
|
| 868 |
+
[2406.00 --> 2406.78] implement the firmware?
|
| 869 |
+
[2406.96 --> 2410.62] So there's, there's issues that the industry is still figuring out how to adopt it, but
|
| 870 |
+
[2410.62 --> 2415.68] they've also definitely seen the success stories of using open source in firmware.
|
| 871 |
+
[2415.68 --> 2420.88] You know, part of the background of the Chromebooks using Coreboot is actually that, you know,
|
| 872 |
+
[2420.94 --> 2425.14] a person was flagged down in the hallway as he was working in that, that building, not
|
| 873 |
+
[2425.14 --> 2428.80] on firmware at all, but he had a past of working on Coreboot.
|
| 874 |
+
[2428.92 --> 2432.40] And they said, Hey, we are trying to work out a firmware situation for these Chromebook
|
| 875 |
+
[2432.40 --> 2432.80] things.
|
| 876 |
+
[2433.32 --> 2435.24] What do you think about trying to port Coreboot to it?
|
| 877 |
+
[2435.28 --> 2438.90] And so they did it and they got it done and it worked better than the actual reference
|
| 878 |
+
[2438.90 --> 2440.34] firmware from, from the vendor.
|
| 879 |
+
[2440.34 --> 2442.88] And it's actually a similar story for OpenBMC.
|
| 880 |
+
[2443.08 --> 2447.50] So I was at Google when that happened and I was managing a team that was starting to
|
| 881 |
+
[2447.50 --> 2454.56] work on BMCs and we got a machine in and we were actually trying to talk with the sales
|
| 882 |
+
[2454.56 --> 2460.50] folks at a major vendor of BMC firmware and just get a quote from them on what it would
|
| 883 |
+
[2460.50 --> 2462.12] cost to license their product.
|
| 884 |
+
[2462.34 --> 2466.30] You would expect that this is something that they would be wanting to do, but it had been
|
| 885 |
+
[2466.30 --> 2468.66] going on long enough that I just said, you know what?
|
| 886 |
+
[2468.66 --> 2472.82] I'm going to port OpenBMC to this and we will see who wins first.
|
| 887 |
+
[2473.04 --> 2474.76] It is actually faster than your sales cycle.
|
| 888 |
+
[2474.86 --> 2478.54] I can port this faster than I can get your sales folks to call me back to license it.
|
| 889 |
+
[2478.56 --> 2484.60] Yeah, I had an entire Linux environment booting on the BMC and then actually causing the host
|
| 890 |
+
[2484.60 --> 2487.24] CPU to power on in about two days.
|
| 891 |
+
[2487.38 --> 2488.14] That's impressive.
|
| 892 |
+
[2488.22 --> 2495.58] And we actually didn't get a quote from said major vendor for another week, at which point
|
| 893 |
+
[2495.58 --> 2498.08] we're just like, I'm sorry, we've got a different solution.
|
| 894 |
+
[2498.08 --> 2499.36] Yeah, thanks to your delay.
|
| 895 |
+
[2499.62 --> 2501.62] We actually have a, that's terrific.
|
| 896 |
+
[2502.00 --> 2507.70] So certainly it seems like open source firmware is alive and well and thriving in the BMC.
|
| 897 |
+
[2508.96 --> 2512.56] And hopefully on the, for the BIOS as well, I mean, can we please?
|
| 898 |
+
[2512.84 --> 2517.62] BIOS is, I mean, as I said, like most of EFI is open source.
|
| 899 |
+
[2518.00 --> 2521.28] You can build EDK2 and get a reference implementation.
|
| 900 |
+
[2521.28 --> 2527.12] What you're missing is a lot of the fit and finish that makes it a production BIOS and the
|
| 901 |
+
[2527.12 --> 2530.96] specific per board configuration stuff that has to happen.
|
| 902 |
+
[2531.16 --> 2532.20] That you actually need to boot.
|
| 903 |
+
[2532.56 --> 2533.40] Yeah, yeah, yeah.
|
| 904 |
+
[2533.40 --> 2539.20] And that's kind of where the difficult divide is in that the silicon vendors tend to feel
|
| 905 |
+
[2539.20 --> 2543.76] that if they give you the raw documentation for how to interact with the system, that they're
|
| 906 |
+
[2543.76 --> 2545.38] handing over too many trade secrets.
|
| 907 |
+
[2546.00 --> 2550.86] So instead they want to hand you some sort of blob that is, you know, just call these
|
| 908 |
+
[2550.86 --> 2554.30] functions in this prebuilt library and it will do the thing.
|
| 909 |
+
[2555.00 --> 2556.36] Do you think that that is well informed?
|
| 910 |
+
[2556.48 --> 2560.10] Do you think that that's just out of embarrassment for you seeing all the dirty laundry?
|
| 911 |
+
[2560.10 --> 2562.04] It's a bit of both.
|
| 912 |
+
[2562.98 --> 2569.42] If you go way back, like these companies were burned by relying on external parties to write
|
| 913 |
+
[2569.42 --> 2575.72] firmware and them having to produce accurate developer information and provide it to folks
|
| 914 |
+
[2575.72 --> 2577.40] in a open-ish way.
|
| 915 |
+
[2578.20 --> 2583.14] And as part of actually having to write the documentation, right, and maintain it as a business
|
| 916 |
+
[2583.14 --> 2583.58] cost.
|
| 917 |
+
[2583.76 --> 2588.30] And also the, it does actually tell a lot about the design of your hardware.
|
| 918 |
+
[2588.30 --> 2592.28] If you tell somebody how to initialize your memory controller, they now know how your
|
| 919 |
+
[2592.28 --> 2593.96] memory controller actually works under the hood.
|
| 920 |
+
[2594.58 --> 2596.80] And that, that can be quite damaging as well.
|
| 921 |
+
[2596.88 --> 2600.68] So there's, there's some validity to it, but there's also a side of, but come on, you're,
|
| 922 |
+
[2600.82 --> 2603.64] the machine that I'm interested in is three generations old.
|
| 923 |
+
[2603.78 --> 2604.00] Right.
|
| 924 |
+
[2604.24 --> 2604.66] You actually.
|
| 925 |
+
[2604.82 --> 2606.56] Why can't you give me information on this?
|
| 926 |
+
[2606.68 --> 2606.88] Right.
|
| 927 |
+
[2606.90 --> 2607.90] Because it might be embarrassing.
|
| 928 |
+
[2608.00 --> 2609.66] Well, I mean, they get embarrassed anyway.
|
| 929 |
+
[2610.02 --> 2610.28] Right.
|
| 930 |
+
[2610.54 --> 2611.22] Yeah, exactly.
|
| 931 |
+
[2611.28 --> 2612.66] They can get embarrassed the easy way or the hard way.
|
| 932 |
+
[2612.86 --> 2613.86] Open compute project.
|
| 933 |
+
[2614.08 --> 2616.10] Open compute seems very vibrant.
|
| 934 |
+
[2616.10 --> 2618.86] It seems like the OCP summit was another one that was, that was exciting for us.
|
| 935 |
+
[2618.94 --> 2619.76] What's your take on that?
|
| 936 |
+
[2620.30 --> 2620.48] Yeah.
|
| 937 |
+
[2620.86 --> 2622.92] OCP has gone an interesting direction.
|
| 938 |
+
[2623.76 --> 2628.50] You know, I, when Facebook first started it, it, it seemed like a unique approach to
|
| 939 |
+
[2628.50 --> 2634.82] getting, stirring up the, the existing ODM manufacturer business to build things that
|
| 940 |
+
[2634.82 --> 2637.92] were different from the, the 19 inch rack standard.
|
| 941 |
+
[2638.18 --> 2641.14] A standard that doesn't like go back to like, isn't that half of a horse's butt or whatever?
|
| 942 |
+
[2641.14 --> 2643.32] Is this going to be another Roman chariots thing?
|
| 943 |
+
[2643.34 --> 2646.20] 19 inches is half, half of a horse's butt.
|
| 944 |
+
[2646.30 --> 2646.74] Is that what you're saying?
|
| 945 |
+
[2646.74 --> 2648.02] Maybe it's like a full horse's butt.
|
| 946 |
+
[2648.10 --> 2648.88] I feel like a horse's butt.
|
| 947 |
+
[2648.98 --> 2649.24] Isn't it?
|
| 948 |
+
[2649.46 --> 2650.58] No, because doesn't this go?
|
| 949 |
+
[2650.78 --> 2651.02] Railroad ties.
|
| 950 |
+
[2651.10 --> 2651.24] Yeah.
|
| 951 |
+
[2651.30 --> 2651.60] This is good.
|
| 952 |
+
[2651.66 --> 2654.64] The railroad ties and the railroad ties ultimately go to like, you have to have two horses.
|
| 953 |
+
[2654.90 --> 2656.18] That was disproven.
|
| 954 |
+
[2656.58 --> 2656.78] Yeah.
|
| 955 |
+
[2656.78 --> 2657.38] I'm so sorry.
|
| 956 |
+
[2657.72 --> 2657.98] I know.
|
| 957 |
+
[2658.00 --> 2661.12] I mean, because horse's butts are like actually pretty large.
|
| 958 |
+
[2661.50 --> 2661.86] Yeah.
|
| 959 |
+
[2661.90 --> 2662.54] It depends on the horse.
|
| 960 |
+
[2662.62 --> 2663.46] Maybe a pony butt.
|
| 961 |
+
[2663.98 --> 2664.60] Like a Shetland pony butt.
|
| 962 |
+
[2664.60 --> 2666.42] Is that where the 24 inch standard came from?
|
| 963 |
+
[2666.42 --> 2666.68] Yeah.
|
| 964 |
+
[2666.68 --> 2667.28] There we go.
|
| 965 |
+
[2667.42 --> 2667.74] Exactly.
|
| 966 |
+
[2668.10 --> 2668.52] Steed butt.
|
| 967 |
+
[2668.52 --> 2668.78] All right.
|
| 968 |
+
[2668.78 --> 2671.18] Sorry to sidetrack us in the butts.
|
| 969 |
+
[2671.30 --> 2672.74] Bring up internet lore.
|
| 970 |
+
[2674.04 --> 2683.74] So over time, as they really brought more people on, it seems like it really gained a lot of movement once Google and Microsoft also joined.
|
| 971 |
+
[2684.28 --> 2690.60] And so now you have sort of the structure of the hyperscalers are feeding into this and making it clear what they want.
|
| 972 |
+
[2690.60 --> 2701.52] They're actually doing a lot of the design work as well and providing reference designs that are what they are using, but they might not be perfectly fit for others to consume.
|
| 973 |
+
[2702.24 --> 2708.88] And so by handing the designs down to OCP, then OCP partners can actually pick that up and produce real designs.
|
| 974 |
+
[2708.88 --> 2710.24] And you have different approaches.
|
| 975 |
+
[2710.24 --> 2716.24] Like Google is well known for doing all of their own server development and working with their manufacturing partners directly.
|
| 976 |
+
[2716.56 --> 2719.96] And, you know, these are very, very custom to them.
|
| 977 |
+
[2720.30 --> 2722.40] This is what Jess calls infrastructure privilege.
|
| 978 |
+
[2723.56 --> 2724.56] I like that term.
|
| 979 |
+
[2724.72 --> 2725.52] It's a good term.
|
| 980 |
+
[2725.68 --> 2726.06] It is.
|
| 981 |
+
[2726.24 --> 2726.46] Yeah.
|
| 982 |
+
[2726.46 --> 2736.06] It's because in talking to folks that work at hyperscalers, we have found that a lot of them don't understand the pain of literally, you know, off the shelf hardware.
|
| 983 |
+
[2736.38 --> 2736.74] Oh, yeah.
|
| 984 |
+
[2736.88 --> 2739.04] They haven't had to touch it in a long time.
|
| 985 |
+
[2739.52 --> 2740.28] Or ever, maybe.
|
| 986 |
+
[2740.40 --> 2740.52] Yeah.
|
| 987 |
+
[2740.56 --> 2745.88] I feel like you've got now a whole generation that's just grown up thinking like everyone has these computers that are this awesome.
|
| 988 |
+
[2745.88 --> 2753.04] At some point, I feel like I got gaslit into thinking that maybe Dell was not that bad by some people who were like, well, it can't be that bad.
|
| 989 |
+
[2753.06 --> 2754.74] It's like, oh, no, like it actually is.
|
| 990 |
+
[2754.74 --> 2755.82] No, it's bad.
|
| 991 |
+
[2755.82 --> 2756.76] But ask Calvin.
|
| 992 |
+
[2757.10 --> 2757.66] Wherever you go.
|
| 993 |
+
[2757.98 --> 2758.88] So you can find that guy.
|
| 994 |
+
[2759.52 --> 2760.42] Calvin and Hobbes.
|
| 995 |
+
[2760.68 --> 2760.98] Maybe.
|
| 996 |
+
[2761.28 --> 2761.48] Yeah.
|
| 997 |
+
[2761.80 --> 2767.44] But then you end up with folks like Facebook who really put out more of an RFQ type thing to the community.
|
| 998 |
+
[2767.68 --> 2770.60] Or they work with a partner to actually design a machine.
|
| 999 |
+
[2771.04 --> 2775.52] And it's very clear that it's been significantly contributed to by the ODM, right?
|
| 1000 |
+
[2775.52 --> 2778.60] So the ODMs are still having a lot of direct involvement in those designs.
|
| 1001 |
+
[2779.26 --> 2781.56] But either way, you end up with a lot of options.
|
| 1002 |
+
[2781.56 --> 2795.22] You see new developments throughout the space in terms of not only like bleeding edge technology developments, but another interesting one is companies that buy older systems that are being decommissioned and refurbish them and sell them.
|
| 1003 |
+
[2795.22 --> 2796.00] Right?
|
| 1004 |
+
[2796.00 --> 2800.40] That was a non-existent market for the high end of computing.
|
| 1005 |
+
[2800.56 --> 2804.32] And not sell them as furniture, even though I think we would all buy them as furniture.
|
| 1006 |
+
[2804.38 --> 2809.40] You know, the last real one was the Mac Pros, the big aluminum cases.
|
| 1007 |
+
[2809.50 --> 2810.36] And those weren't that comfortable.
|
| 1008 |
+
[2811.50 --> 2811.74] Fair.
|
| 1009 |
+
[2811.74 --> 2811.76] Fair.
|
| 1010 |
+
[2812.18 --> 2815.82] So, yeah, I mean, it's interesting to see where this is all going.
|
| 1011 |
+
[2816.08 --> 2823.10] I think OCP is one of those, if you look at it from, well, I should be able to buy this infrastructure and play like the hyperscalers.
|
| 1012 |
+
[2823.60 --> 2825.20] It's not complete, right?
|
| 1013 |
+
[2825.26 --> 2830.80] There's an understanding that the hardware is there, the mechanical infrastructure, et cetera.
|
| 1014 |
+
[2831.04 --> 2832.62] You might be able to purchase it.
|
| 1015 |
+
[2832.82 --> 2834.28] The software story is still complicated.
|
| 1016 |
+
[2834.66 --> 2837.68] I feel like OCP Summit, there was no Kubernetes talk at OCP Summit.
|
| 1017 |
+
[2837.98 --> 2838.20] Thank God.
|
| 1018 |
+
[2838.48 --> 2840.12] Like you act like this is a bad thing.
|
| 1019 |
+
[2840.12 --> 2841.08] It's like, I'm relieved.
|
| 1020 |
+
[2841.36 --> 2849.42] I felt like it was, yeah, no, I was in a Kubernetes free zone, which is kind of ironic because it's like you'd think that these two demographics have a lot of overlap, but it's good.
|
| 1021 |
+
[2849.46 --> 2849.98] It was refreshing.
|
| 1022 |
+
[2850.18 --> 2850.92] It was refreshing.
|
| 1023 |
+
[2851.50 --> 2852.26] Especially for Jess.
|
| 1024 |
+
[2852.42 --> 2853.42] It felt great.
|
| 1025 |
+
[2853.50 --> 2855.38] I was like, wow, I like this conference a lot.
|
| 1026 |
+
[2855.52 --> 2856.52] It was a lot of fun.
|
| 1027 |
+
[2856.78 --> 2859.26] And I mean, who doesn't want to geek out over hardware?
|
| 1028 |
+
[2859.48 --> 2859.94] It's good stuff.
|
| 1029 |
+
[2860.06 --> 2863.52] No, that was, I mean, having the hardware there is like just a treat.
|
| 1030 |
+
[2864.10 --> 2866.78] Yeah, the vendor hall there is a lot more entertaining than other conferences.
|
| 1031 |
+
[2866.96 --> 2867.70] Oh, it was great.
|
| 1032 |
+
[2868.12 --> 2869.08] And so, Rick, what are you excited about?
|
| 1033 |
+
[2869.08 --> 2869.50] Looking forward.
|
| 1034 |
+
[2869.50 --> 2873.52] I mean, you're still having fun at the hardware software interface, clearly.
|
| 1035 |
+
[2874.28 --> 2874.38] Yeah.
|
| 1036 |
+
[2874.50 --> 2876.26] I mean, that's my career in a nutshell.
|
| 1037 |
+
[2876.90 --> 2877.08] Yeah.
|
| 1038 |
+
[2877.26 --> 2883.62] Right now, it's a lot of pushing on the security story around firmware, raising that awareness.
|
| 1039 |
+
[2883.62 --> 2895.06] It's amazing that for all the work that's gone into things like secure boot and verified boot on client-type devices, that in the server space, it's slower to catch on.
|
| 1040 |
+
[2895.06 --> 2900.56] Also, that all of the periphery devices don't get audited nearly the same way.
|
| 1041 |
+
[2900.56 --> 2904.40] How do I know that my hard drive firmware is actually not malicious?
|
| 1042 |
+
[2904.40 --> 2907.90] These are hard questions that the industry hasn't really tackled.
|
| 1043 |
+
[2907.90 --> 2912.60] And talk about there is a domain where I don't think open firmware is ever going to come to the spindle.
|
| 1044 |
+
[2912.74 --> 2912.98] I don't know.
|
| 1045 |
+
[2913.02 --> 2913.70] What do you think on that?
|
| 1046 |
+
[2914.08 --> 2914.22] Yeah.
|
| 1047 |
+
[2914.28 --> 2914.66] I mean, that's...
|
| 1048 |
+
[2915.22 --> 2916.68] That's going to be proprietary forever.
|
| 1049 |
+
[2916.82 --> 2926.20] There's some low-level parts of that that are fundamentally like you would be having to know the mechanical parameters of it and for the design.
|
| 1050 |
+
[2926.20 --> 2926.92] It would just be...
|
| 1051 |
+
[2926.92 --> 2927.66] It'd be awkward.
|
| 1052 |
+
[2928.00 --> 2930.68] Well, optimistic that it's going to get better on the server space, I think.
|
| 1053 |
+
[2930.76 --> 2941.44] We're going to actually get some actual true tested firmware and that the open firmware movement is going to continue to grow and blossom and hopefully OSFC will be even bigger next year.
|
| 1054 |
+
[2941.58 --> 2944.18] But not lose the feel because I love the feel.
|
| 1055 |
+
[2944.18 --> 2944.62] Not use the feel.
|
| 1056 |
+
[2944.72 --> 2946.84] Also, could they change the currency to not be in euros?
|
| 1057 |
+
[2947.02 --> 2947.28] That was weird.
|
| 1058 |
+
[2947.28 --> 2948.00] No, I liked it.
|
| 1059 |
+
[2948.04 --> 2948.30] It's quaint.
|
| 1060 |
+
[2948.30 --> 2949.44] That was so weird.
|
| 1061 |
+
[2949.66 --> 2950.68] It makes it quaint.
|
| 1062 |
+
[2950.90 --> 2952.72] If they change it, then it would not be as weird.
|
| 1063 |
+
[2952.72 --> 2960.76] I mean, my wife was convinced that our credit card was stolen by someone in Berlin and she was right to think that because it was like we were being charged 108 euros.
|
| 1064 |
+
[2960.94 --> 2964.72] I think that it's quaint and if they change it, then it will change the feel of it being...
|
| 1065 |
+
[2965.38 --> 2969.58] It's like, you know, I remember when I could go to this conference in Silicon Valley and pay euros for it.
|
| 1066 |
+
[2969.78 --> 2970.04] Yeah.
|
| 1067 |
+
[2970.44 --> 2971.60] Those are the glory days.
|
| 1068 |
+
[2971.78 --> 2974.36] I mean, Nine Elements is a European company.
|
| 1069 |
+
[2974.36 --> 2976.58] Oh, and I honor their European roots.
|
| 1070 |
+
[2976.74 --> 2977.14] I just...
|
| 1071 |
+
[2977.14 --> 2978.14] It was weird.
|
| 1072 |
+
[2978.72 --> 2979.24] But it was great.
|
| 1073 |
+
[2979.28 --> 2979.84] Great conference.
|
| 1074 |
+
[2980.18 --> 2981.04] Yes, it was amazing.
|
| 1075 |
+
[2981.04 --> 2984.70] And the eggs at Facebook are very good, if anyone was wondering.
|
| 1076 |
+
[2984.80 --> 2985.90] Yeah, did you have the breakfast at Facebook?
|
| 1077 |
+
[2986.08 --> 2986.26] No.
|
| 1078 |
+
[2986.82 --> 2987.80] It was so good.
|
| 1079 |
+
[2987.86 --> 2989.30] I dream about it like every day.
|
| 1080 |
+
[2989.38 --> 2990.74] Those eggs were epic.
|
| 1081 |
+
[2991.70 --> 2995.34] I mean, normally I feel I would moderate Jess on this, but yeah.
|
| 1082 |
+
[2995.66 --> 2996.50] No, they were...
|
| 1083 |
+
[2996.50 --> 2997.74] They were divine.
|
| 1084 |
+
[2997.98 --> 3000.38] Well, Jess is the one that points out that Facebook kind of acts like your parents.
|
| 1085 |
+
[3000.68 --> 3004.44] No, so the second day we were walking in and I was like, let's get egg drunk.
|
| 1086 |
+
[3005.96 --> 3007.14] You did get egg drunk.
|
| 1087 |
+
[3007.38 --> 3007.70] I know.
|
| 1088 |
+
[3007.70 --> 3008.82] You didn't just say let's get egg drunk.
|
| 1089 |
+
[3008.90 --> 3009.96] We actually got egg drunk.
|
| 1090 |
+
[3009.96 --> 3010.32] Still egg drunk.
|
| 1091 |
+
[3010.48 --> 3010.80] I know.
|
| 1092 |
+
[3011.28 --> 3013.06] We actually were strung out on protein.
|
| 1093 |
+
[3013.16 --> 3014.48] But they were like perfectly cooked eggs.
|
| 1094 |
+
[3014.78 --> 3015.58] They really are.
|
| 1095 |
+
[3015.66 --> 3016.66] And then we went to a hackathon.
|
| 1096 |
+
[3016.66 --> 3017.88] We went to a firmware hackathon.
|
| 1097 |
+
[3018.28 --> 3020.92] We were with Rick desoldering a...
|
| 1098 |
+
[3020.92 --> 3021.56] That was the best.
|
| 1099 |
+
[3021.86 --> 3022.38] It was the best.
|
| 1100 |
+
[3022.42 --> 3023.24] It was the best.
|
| 1101 |
+
[3023.38 --> 3024.02] That was so much fun.
|
| 1102 |
+
[3024.22 --> 3027.12] I mean, what's a conference without actually taking apart hardware?
|
| 1103 |
+
[3027.44 --> 3027.70] Oh.
|
| 1104 |
+
[3027.80 --> 3029.36] That was the best part.
|
| 1105 |
+
[3029.62 --> 3030.64] It was so much fun.
|
| 1106 |
+
[3030.72 --> 3031.46] And we still have...
|
| 1107 |
+
[3031.46 --> 3032.48] We got that box.
|
| 1108 |
+
[3032.62 --> 3032.86] It's still ripped.
|
| 1109 |
+
[3032.86 --> 3033.32] It was right over there.
|
| 1110 |
+
[3033.56 --> 3034.10] We're going to debray it.
|
| 1111 |
+
[3034.10 --> 3034.66] It's just hanging out.
|
| 1112 |
+
[3034.84 --> 3035.54] It's hanging out.
|
| 1113 |
+
[3035.82 --> 3036.56] All right, Steve.
|
| 1114 |
+
[3036.56 --> 3038.52] You look like you woke up over there.
|
| 1115 |
+
[3038.70 --> 3038.96] Yeah.
|
| 1116 |
+
[3039.08 --> 3040.26] Do you have any questions for us?
|
| 1117 |
+
[3040.26 --> 3041.12] Or there was an air gap.
|
| 1118 |
+
[3041.52 --> 3041.64] So...
|
| 1119 |
+
[3041.64 --> 3041.88] I hate that.
|
| 1120 |
+
[3042.44 --> 3044.42] I do lots of air gaps.
|
| 1121 |
+
[3044.42 --> 3049.18] So, again, having been operating thousands of machines, you know, less than six months
|
| 1122 |
+
[3049.18 --> 3053.12] ago, A, the BMCs that are hanging out on the internet, terrifying.
|
| 1123 |
+
[3054.20 --> 3055.76] But it doesn't seem like...
|
| 1124 |
+
[3055.76 --> 3060.50] I mean, I wouldn't have thought to go have, you know, our operations team or other teams
|
| 1125 |
+
[3060.50 --> 3065.56] go look into what sorts of vulnerabilities we have in the lower level software systems.
|
| 1126 |
+
[3066.30 --> 3068.70] Like, A, like, are people not worried about this?
|
| 1127 |
+
[3068.76 --> 3069.38] And should they be?
|
| 1128 |
+
[3069.38 --> 3074.30] And B, if they are, where does one even start to figure out how exposed they may be?
|
| 1129 |
+
[3074.54 --> 3080.16] So, a lot of it comes down to really having a security group that thinks through your threat
|
| 1130 |
+
[3080.16 --> 3080.80] model, right?
|
| 1131 |
+
[3080.82 --> 3085.74] So, it may be legitimate for you not to think about this space at all, depending upon if
|
| 1132 |
+
[3085.74 --> 3088.50] you are exposed in a lot of other ways that are much more significant to you.
|
| 1133 |
+
[3088.88 --> 3089.88] The security space...
|
| 1134 |
+
[3089.88 --> 3089.90] Okay.
|
| 1135 |
+
[3090.54 --> 3091.80] Well, the security space...
|
| 1136 |
+
[3091.80 --> 3092.62] You've got bigger problems.
|
| 1137 |
+
[3092.62 --> 3096.14] The entire model of security is like, you can't have perfect security.
|
| 1138 |
+
[3096.28 --> 3096.82] It doesn't exist.
|
| 1139 |
+
[3097.32 --> 3100.66] And so, it's a question of how do you allocate your resources to provide the best defenses
|
| 1140 |
+
[3100.66 --> 3102.86] against what you anticipate your attacks to be.
|
| 1141 |
+
[3103.22 --> 3109.56] So, if you're really a, you know, public-facing website, you probably have an attack model that
|
| 1142 |
+
[3109.56 --> 3111.32] looks at that as the easiest venue.
|
| 1143 |
+
[3111.52 --> 3117.42] And so, someone doing a supply chain attack against your machine is very low on your list,
|
| 1144 |
+
[3117.62 --> 3117.80] right?
|
| 1145 |
+
[3117.84 --> 3119.06] It may be difficult to solve.
|
| 1146 |
+
[3119.18 --> 3122.04] It may be you're actually relying on your vendor to solve that for you.
|
| 1147 |
+
[3122.04 --> 3127.36] So, there are legitimate reasons why, especially smaller companies, you may not...
|
| 1148 |
+
[3127.36 --> 3128.72] It just may not be in your threat model at all.
|
| 1149 |
+
[3129.12 --> 3134.34] Now, why is the space so terrible in terms of the security today?
|
| 1150 |
+
[3134.64 --> 3139.92] Well, I like to think of it as how the industry changed around web security, right?
|
| 1151 |
+
[3139.92 --> 3143.88] If you look back to the early days of the web, like, there was...
|
| 1152 |
+
[3143.88 --> 3145.00] It was completely open.
|
| 1153 |
+
[3145.12 --> 3146.76] Nobody really was thinking through security at all.
|
| 1154 |
+
[3146.90 --> 3148.08] And over time, it evolved.
|
| 1155 |
+
[3148.08 --> 3153.10] But it got reached a point where the security community felt they needed to actually establish,
|
| 1156 |
+
[3153.30 --> 3156.48] these are the 10 most common problems that we see.
|
| 1157 |
+
[3156.58 --> 3158.44] And everybody keeps repeating these mistakes.
|
| 1158 |
+
[3158.94 --> 3163.46] And what we've found is that the complexity of actually building proper implementations
|
| 1159 |
+
[3163.46 --> 3169.46] that avoid all of these risks is so hard that we need to actually call attention to these
|
| 1160 |
+
[3169.46 --> 3172.26] problems that you need to focus on and do the right thing.
|
| 1161 |
+
[3172.26 --> 3174.78] That worked really well in that community, in that space.
|
| 1162 |
+
[3174.78 --> 3181.50] And so, that's driven a lot of what you see in terms of the security developments in the web arena.
|
| 1163 |
+
[3181.72 --> 3185.94] But when you look toward system firmware, because it's a smaller set of companies,
|
| 1164 |
+
[3186.08 --> 3194.18] because the idea of cloud-based system hosting and bare metal clouds and things like that is relatively new,
|
| 1165 |
+
[3194.18 --> 3199.16] you're dealing with a lot of folks who've not been exposed to the mindset of thinking like an attacker.
|
| 1166 |
+
[3199.66 --> 3203.36] So, they're not building the defenses in, right?
|
| 1167 |
+
[3203.40 --> 3205.56] The assumption is the firmware is the first thing that runs.
|
| 1168 |
+
[3205.68 --> 3207.34] It needs to have full access to the machine.
|
| 1169 |
+
[3207.46 --> 3208.62] It needs to do what it does.
|
| 1170 |
+
[3209.12 --> 3211.42] And the only thing it's going to talk to is the OS.
|
| 1171 |
+
[3211.48 --> 3212.64] And I don't need to worry about that.
|
| 1172 |
+
[3212.84 --> 3214.04] That landscape is changing.
|
| 1173 |
+
[3214.04 --> 3217.78] Do you think that Spectre and Meltdown helped open eyes in terms of,
|
| 1174 |
+
[3217.82 --> 3223.14] I think Spectre and Meltdown were such a shock to the system about the bedrock that we've been relying on for so long.
|
| 1175 |
+
[3223.72 --> 3228.68] All of a sudden, the firmware vulnerabilities that have been found exploded after Spectre and Meltdown.
|
| 1176 |
+
[3229.32 --> 3235.58] Do you think that they played any role in kind of shining a bright light down Stack?
|
| 1177 |
+
[3235.78 --> 3237.60] I think it certainly helped gain a lot of awareness.
|
| 1178 |
+
[3237.98 --> 3241.90] I mean, there's been a lot of security vulnerabilities in firmware for a long time.
|
| 1179 |
+
[3241.90 --> 3245.28] Did they receive the same attention from the press?
|
| 1180 |
+
[3245.80 --> 3246.54] Not really.
|
| 1181 |
+
[3247.00 --> 3249.80] Spectre and Meltdown really had the implication of,
|
| 1182 |
+
[3250.22 --> 3252.74] these are bugs that cannot be fixed, right?
|
| 1183 |
+
[3252.98 --> 3254.46] That was kind of the big thing.
|
| 1184 |
+
[3254.64 --> 3257.26] You can mitigate them, but you cannot fix them.
|
| 1185 |
+
[3257.42 --> 3260.68] That was the message that really started to get people to think about,
|
| 1186 |
+
[3260.92 --> 3265.64] my hardware might be broken in a way where I have to take significant performance hits or do something about it.
|
| 1187 |
+
[3265.88 --> 3268.96] And the cycle for mitigating it might be months.
|
| 1188 |
+
[3269.46 --> 3270.50] And that's very uncommon.
|
| 1189 |
+
[3270.50 --> 3275.00] Like, if you think about, if I find a vulnerability in a software package,
|
| 1190 |
+
[3275.38 --> 3278.84] it's likely that there'll be a patch out in days, weeks.
|
| 1191 |
+
[3279.14 --> 3283.20] You know, three months is actually the normal vulnerability disclosure window.
|
| 1192 |
+
[3283.90 --> 3287.02] But in hardware, it could easily be a year.
|
| 1193 |
+
[3287.36 --> 3290.56] Do you think as people start running more cloud-like systems,
|
| 1194 |
+
[3291.32 --> 3294.20] there may emerge a top 10 or a top 5 or a, you know,
|
| 1195 |
+
[3294.20 --> 3300.32] make sure you've checked these aspects of your infrastructure as a best practice?
|
| 1196 |
+
[3300.50 --> 3300.90] Yeah.
|
| 1197 |
+
[3301.06 --> 3306.74] I guess I characterize it as the main thing that happens at the hyperscale and in the cloud place is that the,
|
| 1198 |
+
[3307.14 --> 3310.48] in traditional enterprise, you had a machine owner, right?
|
| 1199 |
+
[3310.52 --> 3316.48] The end owner of the machine was running their application and they sort of owned that whole system top to bottom
|
| 1200 |
+
[3316.48 --> 3322.06] and relied upon your IT department to basically make sure that it was powered on and cooled.
|
| 1201 |
+
[3322.06 --> 3327.08] And if it needed hardware, that they would actually be the hands to go fix it.
|
| 1202 |
+
[3327.08 --> 3336.26] But when you look at the cloud space and how the world of running things like Kubernetes and OpenStack and various systems,
|
| 1203 |
+
[3336.50 --> 3338.74] you're starting to move more to a multi-tenancy.
|
| 1204 |
+
[3339.16 --> 3347.48] You're starting to move toward the owner of the machine is probably more of a operations group that deals with the job scheduling service.
|
| 1205 |
+
[3347.48 --> 3351.82] And they're relying on the physical hardware as a service from the IT department.
|
| 1206 |
+
[3352.04 --> 3355.58] And then you have actual application owners further up the stack.
|
| 1207 |
+
[3355.98 --> 3361.88] And so the assumption that it's the end application owner's responsibility for the whole machine.
|
| 1208 |
+
[3361.88 --> 3366.96] And so we don't need a whole lot of security because they just get the whole machine and they're the only ones that are on it.
|
| 1209 |
+
[3367.32 --> 3368.66] It starts to break down a lot.
|
| 1210 |
+
[3369.28 --> 3375.92] And along with that, you see, well, okay, we had all these security features because we thought somebody might want this someday.
|
| 1211 |
+
[3376.46 --> 3377.66] Do they actually work?
|
| 1212 |
+
[3378.76 --> 3385.70] And a lot of times what we're finding are actually vulnerabilities in the security features where they've been either misconfigured or simply not enabled.
|
| 1213 |
+
[3386.30 --> 3386.44] All right.
|
| 1214 |
+
[3386.46 --> 3389.84] Well, Rick, thank you very much for being with us on On The Metal.
|
| 1215 |
+
[3390.32 --> 3390.90] Yes, thank you.
|
| 1216 |
+
[3390.90 --> 3392.32] A terrific conversation.
|
| 1217 |
+
[3392.50 --> 3393.26] Thank you for having me.
|
| 1218 |
+
[3393.34 --> 3393.62] All right.
|
| 1219 |
+
[3393.66 --> 3394.10] Until next time.
|
| 1220 |
+
[3394.26 --> 3394.38] Thanks, Rick.
|
| 1221 |
+
[3395.42 --> 3399.46] You've been listening to On The Metal, tales from the hardware software interface.
|
| 1222 |
+
[3399.88 --> 3405.88] For show notes, to learn more about our guests, or to sign up for our mailing list, visit us at onthemetal.fm.
|
| 1223 |
+
[3406.34 --> 3408.74] On The Metal is a production of Oxide Computer Company.
|
| 1224 |
+
[3408.96 --> 3412.10] It is recorded in the Oxide Garage in Oakland, California.
|
| 1225 |
+
[3412.76 --> 3415.72] To learn more about Oxide, visit us at oxide.computer.
|
| 1226 |
+
[3416.14 --> 3419.24] On The Metal is hosted by me, Brian Cantrell, along with Jess Frissel.
|
| 1227 |
+
[3419.24 --> 3421.90] And we are frequently joined by our boss, Steve Tuck.
|
| 1228 |
+
[3422.16 --> 3424.42] Our original and awesome theme music is by J.J.
|
| 1229 |
+
[3424.50 --> 3426.04] Wiesler at Pollen Music Group.
|
| 1230 |
+
[3426.30 --> 3427.42] You can learn more about J.J.
|
| 1231 |
+
[3427.50 --> 3430.02] and Pollen at pollenmusicgroup.com.
|
| 1232 |
+
[3430.38 --> 3434.20] We are edited and produced by Chris Hill and his crew at HumblePod.
|
| 1233 |
+
[3434.64 --> 3440.38] From Jess, from Steve, from me, and from all of us at Oxide Computer Company, thanks for listening to On The Metal.
|
| 1234 |
+
[3440.38 --> 3470.36] We'll be right back.
|
| 1235 |
+
[3470.38 --> 3500.36] We'll be right back.
|