text stringlengths 0 2.67k |
|---|
**Justin Garrison:** \[45:56\] Many companies will have a cycle of like "I'm on call this week. I need to be more available this week. The other three weeks, I'm fine, or I can relax." But you have to plan around that. I definitely had moments where I'm like, "I can't do vacation", because I'm on call that week, that s... |
**Autumn Nash:** Well, it's getting even more complicated, because a lot of the open source projects that are going to be able to weather the storm are going to be funded by corporations. So not only are they getting paid, but now you've got different interests, you've got different... You know what I mean? It depends ... |
**Lili Cosic:** But also, open source, I feel like it's shifting into an odd direction, where -- |
**Autumn Nash:** It's so weird. Everything's gotten -- |
**Lili Cosic:** The licensing and everything is shifting into a very interesting direction... Which I understand, from a business point of view; I completely get why you don't want to necessarily -- you put a lot of effort into a company, especially if it's your project that you started, and you want to have it in the ... |
I think Kubernetes and these really large projects work only because there's so many companies involved, and so many large companies involved that there's not really one company we associate it with. But even if Google was the one who initially did the majority of the work, but still. |
**Justin Garrison:** A lot of successful open source is either completely monetized, owned by a single company, they drive direction... Which is great. They could do a lot of stuff. The Hashi stack is one of those examples. Cockroach is another. The other side of it is foundations. And it's like "We have to have this n... |
**Lili Cosic:** Yeah, politically. Yeah, definitely. |
**Justin Garrison:** There's a lot of other things you have to consider. |
**Autumn Nash:** I think before the CNFC and the Linux Foundation were big... Now when these open source projects that people are betting on, \[unintelligible 00:48:39.24\] are being born, and they're starting foundations... You can tell that REST was very intentionally -- their foundation was started, a lot of their d... |
**Lili Cosic:** Yeah, I agree. |
**Justin Garrison:** \[49:19\] I mean, the last 10 years have had a lot of influx in various ways... I just watched the Node.js documentary on YouTube. I'm not a Node developer, but fascinating politics on how they weren't adapting Node to solve some use cases that many large corporations needed... And so they literall... |
**Lili Cosic:** No, definitely not. Yeah. |
**Autumn Nash:** And it's weird, because you want open source to get money, but also this is -- I don't know, it's getting very complicated. |
**Lili Cosic:** Yeah. If it's politics involved straight away. I preferred when I was very naive in the open source world, before realizing a lot of the things... But I do still think that the engineers themselves who do work on Kubernetes or etcd or Prometheus and those kinds of projects are truly the nicest people. I... |
**Justin Garrison:** Well, and even just how intentional they were to set up the community up front. Like, we know we're part of a foundation, but shout-out to Sarah Novotny and Joe Beda and the people that were doing it early on, and just saying "Hey, we want this community to be very explicit about being welcoming. A... |
**Autumn Nash:** The CNFC makes me want to play with Kubernetes just because I want to hang out with the people. |
**Lili Cosic:** Yeah, it's been really nice. I go to KubeCons every once in a while; not as much as I used to, but I still go. I was at the KubeCon EU, and I do like CFP reviews, and things that... But it's really nice to just go there and you still see the same people who work on projects, whether they're still at the... |
**Justin Garrison:** And a lot of times they get that job because they did that work, right? They're like "Oh, we have an initiative at this company to do more of that work, or we want to influence in a way" and they're hiring someone to also bring that influence in and say "Oh, you're already a chair on this SIG. Plea... |
**Lili Cosic:** Definitely. |
**Justin Garrison:** Lily, thank you so much for coming on the show. Where can people find you online? |
**Lili Cosic:** Twitter, or whatever it's called these days... I still have that. Yeah. So it's @lilicosic. |
**Justin Garrison:** Thank you so much for coming on and tell us all about Influx, what you've been doing, and kind of just your career and arc. This has been fascinating. |
**Autumn Nash:** Your career is amazing. You just are so cool. |
**Lili Cosic:** Thank you. I will now hide. Thanks for having me. It's been really fun talking to you. |
**Autumn Nash:** It was so nice meeting you. |
**Justin Garrison:** Alright, we'll talk to you again soon. |
• The 80/20 rule in software development and maintenance |
• Why software engineering can be emotionally draining |
• Tars and compression with Jon Johnson |
• Microsoft Build announcements, including the new Snapdragon chip and data collection concerns |
• AI integration in technology and potential social implications |
• Copilot features on GitHub, Bing, and Windows |
• Concerns about the over-reliance on AI and its potential to take away the fun of writing and learning. |
• The value of using AI as a tool for research, idea generation, and conversation-based learning. |
• Criticisms of Microsoft's climate ambitions being jeopardized by their AI obsession. |
• The issue of AI training consuming massive amounts of power and contributing to climate change. |
• Concerns about the lack of transparency in AI-generated content and the potential for biases in AI decision-making. |
• The importance of verifying sources and not relying solely on AI or human authorities. |
• Concerns over tech companies prioritizing stock prices over innovation |
• Feeling of stagnation and lack of excitement in the industry |
• Mainstreaming of certain technologies vs. cutting-edge innovations |
• Interview with Jon Johnson about container technology and tars |
• Discussion of dad jokes and humor in tech communities |
• Discussion about a person sending $10 to be mentioned in Changelog Plus Plus content |
• Identifying a performance issue with TerraForm state files taking 75% of processing time due to JSON marshalling and unmarshaling |
• The issue was eventually fixed by implementing an in-mem state store that throws away the data anyway, but the original solution took weeks to implement |
• Alternative solutions for fixing the problem include improving marshalling and unmarshaling speed or using a faster disk |
• Discussion about how choosing a language with better JSON handling (such as Java) could be beneficial in certain situations |
• Comparison of performance between Go, Kotlin, and Java in handling JSON data |
• Jon Johnson shares his experience with Gzip and how it relates to his work at Chainguard |
• He explains how APKs (Alpine Package Keeper) are essentially gzipped tarballs in a container format |
• Jon discusses optimizing performance by reducing disk I/O when working with gzipped tarballs |
• Justin Garrison mentions the concept of registries or repositories, where compressed files and metadata are stored for distribution |
• The conversation touches on other formats like CPIOs (an ancient compression format) and zip files, which offer benefits like faster seeking and indexing |
• Differences between zip and tar files |
• Benefits of Targz over zip, including compression and indexing |
• Star-gz file format, which combines compression and indexing |
• eStargz format, an extension of Star-gz with optimized access patterns |
• Comparison of compression efficiency between zip, tar, and Star-gz formats |
• Use cases for Star-gz and eStargz in container images |
• Potential drawbacks to using Star-gz, including rebuild requirements |
• Gzip and Zstandard file compression |
• Deflate algorithm and its use in random access with tar-gz files |
• Productivity and creativity while taking breaks or engaging in mindless tasks |
• Mindset and thinking patterns for problem-solving, including the concept of "slow thinking" vs. "fast thinking" |
• Importance of taking breaks, napping, and being bored to allow the subconscious to work through problems |
• Introduction to Deflate, a compression algorithm |
• Comparison between Gzip and Deflate, with Deflate being the core of Gzip |
• Explanation of how Deflate blocks work, including headers, types (BTYPE), and Huffman encoding |
• Discussion of pointers in Deflate for efficient copying of repeated data |
• Description of how Deflate can "predict the future" by copying sequences of identical characters |
• Deflate compression algorithm basics |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.