text
stringlengths
0
1.49k
[1354.98 --> 1356.90] and virtual interfaces
[1356.90 --> 1358.30] and all these things,
[1358.60 --> 1360.44] which on top of adding latency,
[1360.44 --> 1363.42] also add more points of failure
[1363.42 --> 1365.44] and weird debugging.
[1366.44 --> 1367.60] Yeah, that's, I think,
[1367.64 --> 1370.62] one of the unsolved problems so far.
[1370.82 --> 1371.36] It is just,
[1371.36 --> 1373.48] how can I get you to,
[1373.58 --> 1374.56] how can we tell you
[1374.56 --> 1376.30] what's broken right now?
[1377.32 --> 1379.26] And how do you get out of it?
[1380.06 --> 1380.28] Yeah.
[1380.82 --> 1382.68] I think one of the hardest parts right now
[1382.68 --> 1384.28] is everything's moving so fast
[1384.28 --> 1386.58] and there's so many cool projects popping up
[1386.58 --> 1389.14] is kind of like that analysis paralysis.
[1389.56 --> 1392.46] Like which one of these will be the thing, right?
[1392.84 --> 1394.90] Like you could adopt Istio
[1394.90 --> 1397.44] and in six months,
[1397.54 --> 1398.58] nobody's using Istio.
[1398.72 --> 1400.00] There's some new thing, right?
[1400.00 --> 1402.24] And everything's just moving
[1402.24 --> 1403.50] at the speed of light
[1403.50 --> 1404.90] in the kind of container
[1404.90 --> 1406.40] and orchestration world right now
[1406.40 --> 1407.62] that it gets really hard
[1407.62 --> 1409.58] to settle in
[1409.58 --> 1411.68] and just commit to something.
[1411.98 --> 1412.70] Because especially
[1412.70 --> 1414.46] if you're a large infrastructure, right?
[1414.52 --> 1416.06] Like if you pick Envoy
[1416.06 --> 1417.88] or Istio or something today,
[1419.00 --> 1420.38] you know, swapping that out later
[1420.38 --> 1422.40] could be a huge effort.
[1423.20 --> 1424.32] Yeah, and you probably already
[1424.32 --> 1426.00] have solutions in place, right?
[1426.00 --> 1428.38] So most of these companies
[1428.38 --> 1429.48] probably already have
[1429.48 --> 1432.42] like a library like GoKit
[1432.42 --> 1433.40] in their infrastructure
[1433.40 --> 1435.90] that allows you to encapsulate
[1435.90 --> 1437.44] all of those calling patterns.
[1437.98 --> 1440.04] Like call all the services at once,
[1440.28 --> 1441.64] drop all the requests on lists
[1441.64 --> 1443.50] when I have the first response
[1443.50 --> 1444.10] and so on.
[1444.82 --> 1447.08] That's essentially what's being captured there.
[1447.88 --> 1448.06] Yeah.
[1448.06 --> 1449.78] So anyway,
[1449.90 --> 1451.34] so it's going to be interesting
[1451.34 --> 1452.16] to see it play out.
[1452.22 --> 1453.24] I'm going to be a spectator
[1453.24 --> 1453.84] in this one
[1453.84 --> 1455.06] because
[1455.06 --> 1460.18] this whole go swagger thing
[1460.18 --> 1461.32] keeps you fairly busy.
[1461.32 --> 1461.66] So
[1461.66 --> 1464.30] in between my work.
[1464.94 --> 1465.74] Is that what you work on
[1465.74 --> 1466.40] most of the time?
[1467.48 --> 1467.88] Actually,
[1468.04 --> 1469.02] I used to.
[1469.18 --> 1469.46] So
[1469.46 --> 1471.08] I used to work on it
[1471.08 --> 1471.82] most of the time.
[1472.06 --> 1472.96] I wrote it originally
[1472.96 --> 1474.06] just to prove a point.
[1474.06 --> 1476.66] But then
[1476.66 --> 1478.70] it became fairly useful
[1478.70 --> 1480.02] and VMware started
[1480.02 --> 1481.04] adopting it as well.
[1481.18 --> 1481.30] So
[1481.30 --> 1482.88] now
[1482.88 --> 1483.94] I have somebody else
[1483.94 --> 1485.56] working on it
[1485.56 --> 1486.08] who
[1486.08 --> 1487.04] is steadily
[1487.04 --> 1488.00] fixing bugs.
[1488.12 --> 1489.60] So I got corporate sponsorship
[1489.60 --> 1490.08] finally.
[1491.44 --> 1492.12] And
[1492.12 --> 1493.10] I am personally
[1493.10 --> 1494.84] much more
[1494.84 --> 1495.90] interested in
[1495.90 --> 1496.66] distributed
[1496.66 --> 1497.78] decentralized
[1497.78 --> 1499.06] distributed databases
[1499.06 --> 1499.48] or
[1499.48 --> 1500.26] having
[1500.26 --> 1502.76] another
[1502.76 --> 1503.78] unsolved problem
[1503.78 --> 1504.56] that exists
[1504.56 --> 1504.90] but
[1504.90 --> 1506.04] I don't know
[1506.04 --> 1506.90] how many people
[1506.90 --> 1508.68] are actually
[1508.68 --> 1509.62] confronted with it
[1509.62 --> 1510.38] is
[1510.38 --> 1511.02] gossip