[0.00 --> 3.10] 3, 2, 1 [3.10 --> 9.56] This is Ship It [9.56 --> 12.18] with Justin Garrison and Autumn Nash. [12.56 --> 13.76] If you like this show, [13.96 --> 15.38] you will love The Change Log. [15.86 --> 17.40] It's software news on Mondays, [17.76 --> 19.94] deep technical interviews on Wednesdays, [19.98 --> 20.72] and on Fridays, [21.16 --> 22.34] an awesome talk show [22.34 --> 23.98] for your weekend enjoyment. [24.52 --> 26.52] Find it by searching for The Change Log [26.52 --> 28.08] wherever you get your podcasts. [28.08 --> 30.74] Ship It is brought to you by Fly.io. [31.24 --> 33.56] Launch your app in 5 minutes or less. [33.92 --> 35.98] Learn how at Fly.io. [52.92 --> 53.82] What's up, friends? [53.86 --> 54.90] I'm here with Dave Rosenthal, [55.04 --> 56.28] CTO of Century. [56.78 --> 57.74] So Dave, when I look at Century, [57.74 --> 59.16] I see you driving towards [59.16 --> 61.18] full application health. [61.46 --> 62.90] Error monitoring, where things began. [63.30 --> 64.04] Session replay, [64.26 --> 66.50] being able to replay a view of the interface [66.50 --> 68.22] a user had going on [68.22 --> 69.78] when they experienced an issue [69.78 --> 71.34] with full tracing, full data. [71.68 --> 73.28] The advancements you're making with tracing [73.28 --> 74.56] and profiling, [74.92 --> 75.60] Chrome monitoring, [75.80 --> 76.50] co-coverage, [76.90 --> 77.54] user feedback, [78.10 --> 79.94] and just tons of integrations. [80.36 --> 83.04] Give me a glimpse into the inevitable future. [83.32 --> 84.14] What are you driving towards? [84.14 --> 86.06] Yeah, one of the things that we're seeing [86.06 --> 87.74] is that in the past, [88.06 --> 89.36] people had separate systems [89.36 --> 91.10] where they had like logs on servers, [91.22 --> 91.90] written files. [92.50 --> 94.10] They were maybe sending some metrics [94.10 --> 95.88] to Datadog or something like that [95.88 --> 96.84] or some other system. [97.20 --> 98.38] They were monitoring for errors [98.38 --> 99.06] with some product, [99.16 --> 99.78] maybe it was Century. [100.18 --> 101.18] But more and more, [101.24 --> 102.36] what we see is people want [102.36 --> 104.32] all of these sources of telemetry [104.32 --> 106.28] logically tied together somehow. [106.82 --> 108.94] And that's really what we're pursuing [108.94 --> 110.10] at Century now. [110.10 --> 111.88] We have this concept of a trace ID, [112.20 --> 113.26] which is kind of a key [113.26 --> 114.52] that ties together [114.52 --> 116.26] all of the pieces of data [116.26 --> 118.00] that are associated with the user action. [118.42 --> 119.70] So if user loads a web page, [119.96 --> 121.10] we want to tie together [121.10 --> 122.88] all the server requests that happened, [123.06 --> 124.42] any errors that happened, [124.86 --> 126.18] any metrics that were collected. [126.70 --> 128.58] And what that allows on the backend, [129.08 --> 130.26] you don't just have to look at [130.26 --> 131.50] like three different graphs [131.50 --> 133.08] and sort of line them up in time [133.08 --> 135.04] and try to draw your own conclusions. [135.32 --> 136.62] You can actually like analyze [136.62 --> 138.24] and slice and dice the data [138.24 --> 140.42] and say, hey, what did this metric look like [140.42 --> 142.06] for people with this operating system [142.06 --> 143.22] versus this metric look like [143.22 --> 144.70] for people with this operating system [144.70 --> 146.60] and actually get into those details. [146.94 --> 148.26] So this kind of idea [148.26 --> 151.64] of tying all of the telemetry data together [151.64 --> 153.94] using this concept of a trace ID [153.94 --> 155.30] or basically some key, [155.52 --> 157.46] I think is a big win [157.46 --> 159.38] for developers trying to diagnose [159.38 --> 161.28] and debug real world systems [161.28 --> 162.54] and something that is, [162.88 --> 164.26] we're kind of charged the path [164.26 --> 165.08] for that for everybody. [165.48 --> 165.72] Okay. [166.08 --> 166.96] Let's see you get there. [167.14 --> 168.14] Let's see you get there tomorrow. [168.48 --> 168.58] Yeah. [168.70 --> 169.02] Perfectly. [169.28 --> 170.50] How will systems be different? [170.74 --> 172.80] How will teams be different as a result? [173.14 --> 173.40] Yeah. [173.50 --> 174.66] I mean, I guess, again, [174.92 --> 176.12] I just keep saying it maybe, [176.26 --> 177.60] but I think it kind of goes back [177.60 --> 179.10] to this debuggability experience. [179.26 --> 180.92] When you are digging into an issue, [181.30 --> 182.56] you know, having a sort of [182.56 --> 184.16] a richer data model that's, [184.28 --> 185.80] you know, your logs are structured. [185.92 --> 187.42] They're sort of this hierarchical structure [187.42 --> 188.12] with spans. [188.12 --> 189.80] And not only is it just the spans [189.80 --> 190.42] that are structured, [190.56 --> 191.60] they're tied to errors, [191.72 --> 192.78] they're tied to other things. [193.06 --> 194.48] So when you have the data model [194.48 --> 195.66] that's kind of interconnected, [195.66 --> 198.18] it opens up all different kinds [198.18 --> 200.66] of analysis that were just kind of [200.66 --> 202.42] either very manual before, [202.66 --> 204.74] kind of guessing that maybe this log [204.74 --> 205.42] was, you know, [205.46 --> 206.56] happened at the same time [206.56 --> 207.34] as this other thing, [207.54 --> 208.50] or we're just impossible. [208.72 --> 210.24] We get excited not only about [210.24 --> 211.46] the new kinds of issues [211.46 --> 212.28] that we can detect [212.28 --> 213.70] with that interconnected data model, [213.70 --> 215.12] but also just for every issue [215.12 --> 215.92] that we do detect, [216.00 --> 217.00] how easy it is to get [217.00 --> 217.74] to the bottom of it. [217.74 --> 218.54] I love it. [218.62 --> 219.86] Okay, so they mean it [219.86 --> 220.96] when they say code breaks, [221.20 --> 222.50] fix it faster with Sentry. [222.74 --> 224.72] More than 100,000 growing teams [224.72 --> 226.32] use Sentry to find problems fast, [226.50 --> 227.46] and you can too. [227.92 --> 230.40] Learn more at Sentry.io. [230.54 --> 234.30] That's S-E-N-T-R-Y.io. [234.90 --> 235.92] And use our code, [236.18 --> 236.70] changelog, [236.80 --> 238.94] get $100 off the team plan. [239.24 --> 240.56] That's almost four months free [240.56 --> 241.90] for you to try out Sentry. [242.20 --> 242.82] Once again, [243.24 --> 244.72] Sentry.io. [244.72 --> 244.96] Sentry.io. [256.98 --> 258.70] Hello, and welcome to Ship It, [258.88 --> 260.20] the podcast all about everything [260.20 --> 261.42] after Get Pushed. [261.50 --> 262.62] I'm your host, Justin Garrison, [262.84 --> 264.78] and with me as always is Autumn Nash. [264.86 --> 265.40] How's it going, Autumn? [266.06 --> 267.72] I'm very happy to be here. [267.96 --> 268.84] Slightly caffeinated, [269.08 --> 270.12] like almost there. [271.14 --> 271.64] Getting there. [271.78 --> 272.40] Getting there. [272.40 --> 273.04] A little more coffee. [273.04 --> 275.66] Are we telling people [275.66 --> 276.30] about your new job? [276.92 --> 278.06] Yeah, we're good to go now. [278.36 --> 279.48] I mean, the world knows [279.48 --> 279.96] at this point. [281.20 --> 282.02] That's right, you announced [282.02 --> 282.36] it on LinkedIn. [282.46 --> 282.82] We're good. [283.00 --> 284.22] So what's your new job? [284.66 --> 287.02] So I am the product manager [287.02 --> 289.68] for Azure Linux. [290.18 --> 292.34] So the Azure Linux distribution [292.34 --> 294.20] at Microsoft. [294.38 --> 295.10] Just started this week. [295.18 --> 295.58] Congratulations. [295.90 --> 296.40] Well, actually, [296.64 --> 298.60] the security product manager. [298.60 --> 300.12] I'm sure there's a lot of PMs [300.12 --> 301.34] on Azure Linux. [301.34 --> 302.78] So I am the one that like [302.78 --> 305.60] works on the security vision [305.60 --> 307.86] for Azure Linux. [308.06 --> 309.38] So anyone listening to the podcast, [309.48 --> 310.06] if you have problems [310.06 --> 311.34] with security on Azure, [311.46 --> 312.00] you know what. [312.34 --> 313.02] Rude, Justin. [313.44 --> 313.90] Rude. [314.10 --> 314.66] Don't lie. [314.80 --> 315.80] The DMs are going to be [315.80 --> 317.44] from you making different accounts. [317.96 --> 317.98] Like, [319.06 --> 319.98] on my bot army. [321.30 --> 321.98] Because what? [322.50 --> 323.12] It's going to be [323.12 --> 324.76] Justin's bot army being like, [324.92 --> 326.50] can you fix this for me? [326.50 --> 328.26] I found a bug. [329.12 --> 330.96] For any long-time listeners [330.96 --> 331.46] of the show, [331.60 --> 332.50] you'll know our guest. [332.72 --> 333.58] How's it going, Gerhard? [334.22 --> 335.14] It's going really well. [335.48 --> 337.16] I'm very happy to be back. [337.50 --> 338.48] This feels very cozy. [338.86 --> 340.12] I'm so excited to meet you. [340.30 --> 341.30] I feel like I'm geeking out. [341.64 --> 341.96] Likewise. [342.54 --> 343.38] I was a long-time listener [343.38 --> 343.76] of the show. [343.86 --> 344.56] I thought it was great. [344.82 --> 345.48] Can you bring us back? [345.54 --> 347.10] Like, why did you start ShipIt? [347.52 --> 349.26] It started with all the work [349.26 --> 349.92] that we were doing [349.92 --> 351.26] on Changelog [351.26 --> 352.38] with Adam and Jared. [352.38 --> 353.02] I mean, [353.06 --> 354.04] there was a lot of [354.04 --> 355.80] infra work [355.80 --> 357.64] and setting everything up [357.64 --> 358.90] and going through [358.90 --> 359.60] all the motions [359.60 --> 360.84] that you normally do [360.84 --> 361.80] when you take an application [361.80 --> 362.46] to production. [362.78 --> 363.78] And we've been doing that [363.78 --> 364.28] for, [364.64 --> 365.66] I don't know how many years [365.66 --> 366.70] before ShipIt started, [366.76 --> 367.48] but it's been years [367.48 --> 367.96] in the making. [368.58 --> 369.48] And there were blog posts [369.48 --> 370.26] before that. [370.96 --> 372.52] And one day, [372.82 --> 373.56] we realized, [373.74 --> 374.12] actually, [374.42 --> 375.84] there's so much here [375.84 --> 377.02] that we could [377.02 --> 378.06] start a podcast, [378.42 --> 379.30] start a new show [379.30 --> 380.66] if you'd be up for it. [381.24 --> 382.26] And the rest is [382.26 --> 382.58] history. [383.16 --> 384.14] And you carried that on [384.14 --> 385.46] for, I think it was 90 episodes, [385.62 --> 386.22] which was awesome. [386.86 --> 387.56] And then also, [387.68 --> 388.04] we have, [388.28 --> 388.46] like, [388.54 --> 389.44] going full circle [389.44 --> 390.44] from you stopping [390.44 --> 391.18] at the 90th episode, [391.26 --> 391.78] we have some news [391.78 --> 392.64] to share everyone else [392.64 --> 393.66] that ShipIt, [393.74 --> 394.56] as this podcast [394.56 --> 395.40] on the Changelog, [395.80 --> 396.66] is going to stop [396.66 --> 397.36] at the end of the year. [397.44 --> 399.02] So at the end of December 2024, [399.88 --> 400.40] don't know when you're [400.40 --> 401.10] listening to this now, [401.22 --> 402.76] but we're stopping [402.76 --> 403.46] the podcast. [403.82 --> 404.18] Again, [404.26 --> 404.56] for you, [404.56 --> 405.60] first time for me [405.60 --> 406.48] and Adam to stop it. [407.08 --> 407.10] Well, [407.34 --> 408.46] that was all news [408.46 --> 409.14] to me as well [409.14 --> 410.64] when we scheduled [410.64 --> 412.76] this conversation. [413.40 --> 413.86] We didn't know [413.86 --> 414.42] about that. [414.92 --> 416.36] And I'm glad [416.36 --> 417.16] that I was able [417.16 --> 418.10] to come back [418.10 --> 418.86] one more time [418.86 --> 420.72] before the original [420.72 --> 421.32] ShipIt, [421.42 --> 422.12] in this form, [422.62 --> 424.36] will be put on pause. [424.46 --> 425.26] I always like to say [425.26 --> 425.86] it's on pause. [426.28 --> 427.06] Maybe indefinitely, [427.84 --> 428.74] most likely indefinitely, [429.14 --> 430.12] but you're right. [430.28 --> 431.18] It's like history [431.18 --> 431.92] repeating itself. [432.38 --> 432.54] Yeah. [432.70 --> 432.90] So, [432.90 --> 433.92] and for anyone, [434.06 --> 434.28] you know, [434.32 --> 434.56] listening, [435.06 --> 435.82] sorry about the news [435.82 --> 436.44] breaking it to you. [436.66 --> 437.46] This is a decision [437.46 --> 438.66] for Changelog [438.66 --> 439.58] as the network. [439.72 --> 441.22] They're stripping down, [441.70 --> 442.04] not stripping, [442.12 --> 443.30] they're not going to do [443.30 --> 444.14] a lot of the extra [444.14 --> 445.02] podcasts they were doing. [445.12 --> 446.60] I think GoTime [446.60 --> 447.94] and JS Party, [448.40 --> 449.14] they want to focus [449.14 --> 449.58] on the main [449.58 --> 450.40] Changelog podcast. [450.78 --> 451.40] And that makes [451.40 --> 452.00] total sense to me. [452.06 --> 452.58] I think we have, [452.58 --> 453.26] they're up to seven [453.26 --> 453.80] right now. [454.04 --> 455.08] I came in, [455.26 --> 455.36] and, [455.46 --> 455.62] you know, [455.70 --> 456.48] when we restarted [456.48 --> 456.68] ShipIt, [456.72 --> 456.98] we were just like, [457.04 --> 457.64] let's just see [457.64 --> 458.56] what happens. [459.14 --> 459.84] And Autumn and I [459.84 --> 460.98] have been doing this [460.98 --> 461.96] for almost a full year. [462.50 --> 463.36] And they wanted [463.36 --> 464.06] to trim it back. [464.10 --> 464.46] And that makes [464.46 --> 464.80] total sense. [464.94 --> 465.34] Autumn and I [465.34 --> 466.38] are planning on [466.38 --> 467.46] continuing on [467.46 --> 469.50] with some form [469.50 --> 470.26] of this podcast, [470.46 --> 471.28] at least for a little while. [471.34 --> 472.04] We still have [472.04 --> 473.30] a bunch of amazing [473.30 --> 474.16] people to interview [474.16 --> 475.80] about all these [475.80 --> 476.32] different topics [476.32 --> 476.82] that we're just like, [476.86 --> 477.30] you know what? [477.40 --> 477.54] Like, [477.60 --> 478.34] this already has [478.34 --> 478.86] some momentum. [479.22 --> 479.62] We already [479.62 --> 481.04] appreciate everyone [481.04 --> 481.60] that's listening [481.60 --> 483.58] and talking to us [483.58 --> 484.30] and telling us, [484.64 --> 485.20] giving us feedback [485.20 --> 485.62] and telling us [485.62 --> 486.12] what they like [486.12 --> 486.74] about the show. [487.00 --> 487.68] So we want to [487.68 --> 488.16] continue that. [488.24 --> 488.54] We want to, [488.66 --> 489.12] we think there is [489.12 --> 489.96] space for this [489.96 --> 491.86] in the podcast [491.86 --> 492.26] universe. [492.80 --> 493.58] And it's a passion [493.58 --> 494.10] that, [494.10 --> 494.56] you know, [494.62 --> 495.12] Autumn and I [495.12 --> 495.80] and a lot of people [495.80 --> 496.94] share just about [496.94 --> 498.32] infrastructure and technology [498.32 --> 498.96] and just running, [499.12 --> 499.38] you know, [499.40 --> 500.42] responsibility of running [500.42 --> 502.08] software in general. [502.58 --> 503.00] And it's awesome [503.00 --> 503.96] meeting all the people [503.96 --> 504.58] that run, [504.74 --> 505.64] like that maintain [505.64 --> 506.46] and run software [506.46 --> 507.12] and infrastructure. [507.60 --> 508.34] And the variety. [508.86 --> 509.68] Like the variety [509.68 --> 510.16] of people that we, [510.24 --> 511.22] everything from 3D [511.22 --> 511.92] printer software, [512.04 --> 512.64] like the Octoprint [512.64 --> 513.68] stuff to stuff [513.68 --> 514.32] in space. [514.32 --> 515.78] And it's been awesome [515.78 --> 516.56] just learning all [516.56 --> 517.62] of the things [517.62 --> 518.24] that are different [518.24 --> 519.06] and the challenges [519.06 --> 519.92] in each space, [519.94 --> 520.88] but also all the things [520.88 --> 521.46] that are the same. [521.82 --> 521.88] Yeah, [521.92 --> 522.76] that are very much [522.76 --> 523.16] the same [523.16 --> 523.98] in the most [523.98 --> 524.94] hilarious of ways. [525.52 --> 526.34] We should also [526.34 --> 527.30] see if listeners [527.30 --> 528.08] want to send us [528.08 --> 528.66] some ideas [528.66 --> 529.24] for the name [529.24 --> 530.16] of our new podcast [530.16 --> 530.70] because that would [530.70 --> 531.10] be neat. [531.30 --> 531.68] That's going to be [531.68 --> 532.36] unhinged. [532.54 --> 532.76] Yeah, [532.86 --> 534.82] but that's [534.82 --> 535.26] when you get [535.26 --> 536.04] the best stuff. [538.04 --> 538.62] So yeah, [538.64 --> 539.22] so this episode, [539.32 --> 540.04] I think we have [540.04 --> 541.06] three more episodes [541.06 --> 542.02] after this one [542.02 --> 543.08] to finish out [543.08 --> 543.54] the year. [544.02 --> 544.92] And hopefully [544.92 --> 545.98] on that last episode, [545.98 --> 546.76] we will have [546.76 --> 548.58] some more formal [548.58 --> 549.22] announcement about [549.22 --> 550.26] where you can find us, [550.78 --> 551.66] where this is going forward. [552.28 --> 552.88] Jared and Adam [552.88 --> 553.54] have been great [553.54 --> 554.36] and they are [554.36 --> 555.40] encouraging us [555.40 --> 555.92] to continue [555.92 --> 556.94] and allowing us [556.94 --> 558.18] to keep doing this [558.18 --> 558.76] so they might keep [558.76 --> 559.46] some sort of redirect [559.46 --> 560.40] up for people [560.40 --> 560.92] that are listening [560.92 --> 561.52] to this later [561.52 --> 562.66] than the end [562.66 --> 563.26] of 2024. [564.20 --> 564.40] But yeah, [564.42 --> 564.92] we want to be able [564.92 --> 565.90] to keep that going [565.90 --> 566.38] for some people [566.38 --> 566.82] and make it as [566.82 --> 567.66] seamless as possible, [567.80 --> 568.38] but also like [568.38 --> 568.68] you're probably [568.68 --> 569.04] going to have to [569.04 --> 569.72] add a new feed [569.72 --> 571.30] in your podcast listener. [571.92 --> 572.52] I'm really excited [572.52 --> 572.76] though. [572.86 --> 573.20] I feel like [573.20 --> 575.00] this just allows [575.00 --> 576.00] for like a new [576.00 --> 577.16] evolution of ShipIt. [577.52 --> 577.80] Yeah. [578.22 --> 578.52] I mean, [578.52 --> 579.56] you had 40 episodes, [579.72 --> 579.84] right? [579.94 --> 580.44] Nine months. [580.96 --> 582.00] More than 40 actually [582.00 --> 582.54] at this point, [582.64 --> 583.30] close to 50. [583.94 --> 584.60] How would you [584.60 --> 585.24] summarize that [585.24 --> 585.92] in a few words? [586.60 --> 587.56] All the episodes [587.56 --> 588.56] that you've done [588.56 --> 589.24] so far [589.24 --> 591.52] in this format? [591.88 --> 592.46] I think kind of [592.46 --> 593.10] going back to what [593.10 --> 593.50] Justin said, [593.54 --> 594.62] it's amazing to see [594.62 --> 595.76] like you can be [595.76 --> 596.58] running a satellite [596.58 --> 597.26] in space [597.26 --> 597.94] or you can be [597.94 --> 599.70] running pipelines [599.70 --> 601.80] and platform teams [601.80 --> 603.04] and it's so much [603.04 --> 603.78] that is different [603.78 --> 604.42] but so much [604.42 --> 605.12] of it is the same. [605.26 --> 605.86] So much of the [605.86 --> 606.46] new technology [606.46 --> 607.22] that we've built [607.22 --> 608.32] to make infrastructure [608.32 --> 609.72] easier is also [609.72 --> 610.84] just reminiscent [610.84 --> 612.66] of like the past, [612.82 --> 613.04] you know, [613.10 --> 614.22] and it just makes it, [614.98 --> 615.22] I don't know, [615.30 --> 616.44] it's like all the [616.44 --> 616.98] different ways [616.98 --> 617.66] that you can solve [617.66 --> 618.90] this awesome big puzzle [618.90 --> 619.60] in it. [619.98 --> 621.74] I think sometimes [621.74 --> 622.76] tech gets really weird, [622.98 --> 623.28] you know, [623.34 --> 624.44] and this podcast [624.44 --> 625.34] has made me [625.34 --> 626.80] remember why [626.80 --> 627.92] I love what we do [627.92 --> 630.24] and kept me loving it [630.24 --> 632.08] even in like the last year, [632.32 --> 632.64] you know? [633.30 --> 633.98] I think like for me [633.98 --> 634.36] some of the, [634.82 --> 635.52] my favorite episodes [635.52 --> 636.36] were the throwbacks, [636.48 --> 636.62] right? [636.64 --> 637.46] Like talking to Rich [637.46 --> 637.96] and Mandy [637.96 --> 638.72] and people that were like, [638.98 --> 640.08] this is what it was like [640.08 --> 641.52] to run the AOL chat rooms. [641.60 --> 641.76] I'm like, [641.78 --> 642.34] that was awesome, [642.40 --> 642.56] right? [642.58 --> 643.20] It was just like, [643.50 --> 644.54] it was basically [644.54 --> 645.08] the same thing [645.08 --> 645.60] we're doing now [645.60 --> 646.64] just with tools [646.64 --> 647.04] that everyone's like, [647.08 --> 647.16] oh, [647.18 --> 647.68] you shouldn't use [647.68 --> 648.00] those anymore. [648.02 --> 648.20] I'm like, [648.24 --> 649.32] that ran the internet [649.32 --> 650.28] for years and years [650.28 --> 650.62] and years. [650.62 --> 651.66] Like we can't just [651.66 --> 652.66] throw out all the old [652.66 --> 653.40] stuff that was, [653.40 --> 654.54] was super functional [654.54 --> 656.04] because we don't like [656.04 --> 656.52] it anymore. [656.86 --> 657.54] And so those were [657.54 --> 658.28] really cool to me. [659.00 --> 659.04] Also, [660.24 --> 661.04] it's wild. [661.22 --> 662.54] Like the amount [662.54 --> 663.44] of people like [663.44 --> 664.38] that we've met [664.38 --> 665.06] and they were just like, [665.10 --> 665.64] we were doing [665.64 --> 666.68] this cool thing [666.68 --> 667.42] and I found it [667.42 --> 668.34] and I start doing it [668.34 --> 668.88] and then it leads [668.88 --> 669.64] to this job [669.64 --> 670.76] and this whole career. [670.90 --> 672.20] Like learning [672.20 --> 673.46] how you run Linux [673.46 --> 674.32] and different things [674.32 --> 674.86] in space [674.86 --> 675.98] is just wild to me [675.98 --> 676.54] and how you have [676.54 --> 677.14] to make sure [677.14 --> 679.16] that it can be updated [679.16 --> 680.64] and just all the [680.64 --> 682.00] thought that goes into it. [682.14 --> 683.28] But the people [683.28 --> 684.76] that we've met [684.76 --> 685.86] are almost cooler [685.86 --> 686.78] than the technology. [687.60 --> 687.78] Absolutely. [688.22 --> 688.40] I mean, [688.42 --> 688.56] yeah, [688.60 --> 690.18] the people in their journey [690.18 --> 690.68] into it [690.68 --> 691.84] have been really fun [691.84 --> 692.18] to learn from. [692.24 --> 693.02] In almost every case, [693.06 --> 693.90] it was like someone just, [694.06 --> 694.20] well, [694.20 --> 694.80] I just stepped up [694.80 --> 695.54] and learned a thing. [695.66 --> 696.40] That's what I'm saying. [696.54 --> 698.30] But how many jobs [698.30 --> 699.04] can you make [699.04 --> 700.80] this type of impact [700.80 --> 701.50] on the world, [701.96 --> 702.96] this type of money [702.96 --> 704.28] and this type of [704.28 --> 705.30] community [705.30 --> 706.78] and just because [706.78 --> 707.66] you thought something [707.66 --> 708.12] was cool [708.12 --> 708.78] and you nerded out [708.78 --> 709.12] about it, [709.20 --> 711.16] that is the essence [711.16 --> 711.88] of what makes [711.88 --> 713.38] us still want to do this [713.38 --> 714.20] at the end of the day. [714.52 --> 714.66] You know? [715.16 --> 715.46] Yeah. [715.46 --> 716.80] I think my favorite [716.80 --> 718.00] ideas start with [718.00 --> 719.14] this can't be done. [719.64 --> 719.84] Yeah. [720.24 --> 721.10] This is too crazy. [721.36 --> 722.08] This is like no way. [722.12 --> 723.24] This is never going to work. [723.94 --> 725.24] And going through the cycles [725.24 --> 727.20] to either realize, [727.44 --> 727.68] indeed, [727.80 --> 728.64] this will never work [728.64 --> 729.66] the way I thought it would. [730.42 --> 731.70] But the learnings [731.70 --> 732.98] and the relationships [732.98 --> 733.56] that you make [733.56 --> 734.10] along the way, [734.30 --> 735.14] those are the ones [735.14 --> 735.98] that will take you [735.98 --> 737.22] wherever you're going next. [737.90 --> 738.22] So, [738.94 --> 739.48] it's all [739.48 --> 740.46] little steps, [740.68 --> 741.26] some is steps, [741.26 --> 742.52] and usually the missteps [742.52 --> 743.00] are the ones [743.00 --> 743.74] that teach you the most. [744.12 --> 744.86] That would be one [744.86 --> 746.96] of my takeaways, [747.44 --> 747.74] I think, [747.88 --> 749.10] from ShipIt [749.10 --> 750.34] and from all the work [750.34 --> 750.94] that you do [750.94 --> 752.18] in this industry. [752.78 --> 753.82] Learning from mistakes. [754.36 --> 755.14] So powerful. [755.68 --> 756.42] So true. [756.56 --> 757.44] Because I think it's, [757.70 --> 757.84] like, [757.88 --> 758.74] I used to be, like, [758.76 --> 760.16] really scared of making mistakes [760.16 --> 761.02] and wanted it to be, like, [761.08 --> 761.94] perfect at everything, [762.06 --> 762.80] which was my toxic [762.80 --> 763.60] engineer trait. [764.40 --> 765.08] And, like, [765.16 --> 766.30] I think at, like, [766.34 --> 767.00] a certain point [767.00 --> 767.68] when you've done [767.68 --> 768.48] so much stuff [768.48 --> 769.18] in, like, production [769.18 --> 770.06] and just, like, [770.12 --> 771.14] worked in this industry [771.14 --> 771.80] for so long, [771.88 --> 772.60] like, you're no longer [772.60 --> 773.40] scared of, like, [773.90 --> 774.58] making mistakes. [774.58 --> 775.38] Like, you just kind of [775.38 --> 776.30] almost have to, like, [776.82 --> 778.94] get joy in the ambiguity, [779.28 --> 779.58] you know, [779.64 --> 781.24] and the doing hard things [781.24 --> 781.92] because you have [781.92 --> 782.66] no other choice. [782.96 --> 783.48] Well, you have to be [783.48 --> 784.04] given the freedom [784.04 --> 784.62] to do that, right? [784.66 --> 785.76] Like, the number one [785.76 --> 786.52] contributing factor [786.52 --> 788.24] to, like, [788.32 --> 789.26] good performance teams [789.26 --> 790.46] is psychological safety. [790.94 --> 791.04] Right? [791.06 --> 791.96] Like, being able to say, [792.04 --> 793.30] like, I don't know [793.30 --> 794.60] or I made a mistake [794.60 --> 795.12] and everyone's like, [795.44 --> 795.70] great, [795.84 --> 796.88] what do we learn from that? [796.94 --> 797.76] Where are we going forward? [798.20 --> 799.18] And that's okay [799.18 --> 800.68] to be able, you know, [800.68 --> 801.60] to have that freedom [801.60 --> 802.48] to make the mistakes [802.48 --> 804.46] and there's a lot of privilege [804.46 --> 806.04] in that for some of us. [806.04 --> 806.62] I guess, like, [806.64 --> 807.60] a white dude in tech. [807.76 --> 808.24] Like, I've been given [808.24 --> 809.00] the benefit of the doubt [809.00 --> 810.30] more than I should have been [810.30 --> 811.32] throughout my career [811.32 --> 812.08] to be able to say, [812.16 --> 813.04] actually, I don't know that [813.04 --> 813.80] or I messed up. [813.92 --> 815.50] Sorry, I'll fix it next time, [815.62 --> 816.56] which I know a lot of people [816.56 --> 817.08] don't give that, [817.14 --> 818.34] but also a lot of companies [818.34 --> 819.02] don't give that [819.02 --> 819.76] because they're just like, [820.14 --> 821.42] we hire senior people [821.42 --> 822.52] and senior people [822.52 --> 823.42] know what they're doing, right? [823.46 --> 823.90] Like, no. [824.12 --> 824.72] Like, senior people [824.72 --> 825.88] don't know what they're doing either. [826.22 --> 826.92] They've just taken down [826.92 --> 828.14] production before, right? [828.14 --> 830.12] It's just like the only real difference [830.12 --> 831.08] of, like, the junior people [831.08 --> 831.86] are, like, terrified [831.86 --> 832.66] to take down production [832.66 --> 833.72] and the senior people [833.72 --> 834.02] are like, [834.08 --> 834.90] oh, no, this is going to be all right. [835.06 --> 835.96] But that's why I think [835.96 --> 837.36] this podcast is important [837.36 --> 839.82] and that's why I'm proud [839.82 --> 841.28] of the last 50 episodes [841.28 --> 841.82] that we did [841.82 --> 843.70] because I feel like [843.70 --> 844.82] there's a lot of podcasts [844.82 --> 845.38] that are, like, [845.42 --> 846.42] big on, like, tech [846.42 --> 846.76] and, like, [846.78 --> 848.26] they're very technically deep depth, [848.38 --> 848.94] but, like, [849.66 --> 850.80] I appreciate the way [850.80 --> 851.52] that we talk about [851.52 --> 852.30] making mistakes [852.30 --> 853.48] and the way that we talk about, [853.48 --> 854.68] like, the people aspect [854.68 --> 855.88] and how you have to have [855.88 --> 856.66] that safe environment [856.66 --> 858.42] and we can talk about diversity [858.42 --> 859.76] and all these different things [859.76 --> 861.14] because, like, [861.20 --> 862.42] I think people really think [862.42 --> 862.92] that, like, [863.06 --> 864.74] diversity or safe places [864.74 --> 865.20] or, like, [865.22 --> 865.94] all these things [865.94 --> 866.44] are, like, [866.52 --> 867.58] an added bonus [867.58 --> 868.50] to technology, [868.50 --> 870.22] but you can't make good tech [870.22 --> 871.50] without thinking [871.50 --> 872.22] about the people, [872.32 --> 873.26] without thinking about [873.26 --> 874.66] how to make a better environment. [874.66 --> 875.30] So, like, [875.96 --> 876.92] whatever we can do [876.92 --> 878.22] to use whatever privilege [878.22 --> 878.80] we have [878.80 --> 879.82] to influence [879.82 --> 881.10] and to make things better [881.10 --> 881.84] and to, like, [881.88 --> 883.00] help people know [883.00 --> 883.40] that, like, [883.58 --> 884.34] they can get started [884.34 --> 885.28] and also just to talk [885.28 --> 886.00] to people that are [886.00 --> 887.86] really good technically [887.86 --> 889.04] but are come from [889.04 --> 890.08] all these different backgrounds. [890.20 --> 891.00] Like, look at all the people [891.00 --> 891.96] that we've had on the show, [892.22 --> 892.52] you know? [892.76 --> 892.96] Like, [893.38 --> 894.40] so I just think it's cool [894.40 --> 895.46] to be able to use [895.46 --> 896.86] the privilege that we have [896.86 --> 899.06] to try to make it better [899.06 --> 900.08] and make other people, [900.08 --> 900.36] like, [900.46 --> 900.72] seen [900.72 --> 901.84] and to also, like, [901.92 --> 902.62] show that, like, [902.82 --> 903.58] you can be different [903.58 --> 905.12] and still very technically deep, [905.36 --> 905.66] you know? [906.14 --> 906.96] Well, I guess we'll just [906.96 --> 907.88] transition right into, [908.12 --> 908.32] Gerhard, [908.36 --> 909.10] what have you been doing [909.10 --> 911.28] since you kind of left Ship It, [911.38 --> 912.54] since you left Changelog, [912.54 --> 913.66] what have you been working on? [913.76 --> 914.98] What software [914.98 --> 916.10] have you been responsible for? [916.86 --> 917.26] So, [917.42 --> 918.86] it feels like [918.86 --> 920.26] I never really left Changelog [920.26 --> 921.04] because, [921.20 --> 921.94] first of all, [921.98 --> 922.96] the Kaizen episodes [922.96 --> 924.72] and of all the [924.72 --> 925.84] infrastructure improvements [925.84 --> 927.02] that we are still driving [927.02 --> 928.50] and they're still [928.50 --> 929.38] very much present. [930.08 --> 931.20] Trying all the things [931.20 --> 931.70] that we did [931.70 --> 932.74] over the years [932.74 --> 933.92] and taking it [933.92 --> 935.00] to a place where it is now [935.00 --> 935.92] and continuing the journey, [936.04 --> 936.78] that has been [936.78 --> 938.88] a long-term, [939.40 --> 940.58] very satisfying journey. [940.98 --> 941.52] That's the way [941.52 --> 942.08] I would put it. [942.50 --> 943.16] And I'm very happy [943.16 --> 944.06] that that is continuing [944.06 --> 945.44] and we figured out [945.44 --> 947.08] a way to make that work [947.08 --> 948.02] with Adam and Gerard. [948.26 --> 948.28] So, [948.88 --> 949.80] that is [949.80 --> 952.22] personally a very satisfying thing [952.22 --> 953.04] and also professionally [953.04 --> 954.10] a very satisfying thing. [954.76 --> 955.56] After Ship It, [956.06 --> 957.08] and the reason why [957.08 --> 957.42] for me, [957.52 --> 958.30] even like back then [958.30 --> 958.96] when I was [958.96 --> 959.88] the last episode, [960.02 --> 960.30] 19, [960.42 --> 961.16] Bracing Change, [961.76 --> 962.88] it was a [962.88 --> 963.52] priorities, [964.52 --> 965.56] like, [965.68 --> 966.52] I had to reshuffle [966.52 --> 967.28] a bunch of priorities, [967.40 --> 967.70] basically, [967.70 --> 968.46] and I had to give [968.46 --> 969.28] more time [969.28 --> 970.72] to my main job, [970.96 --> 971.62] which at the time [971.62 --> 972.14] was Dagger, [972.48 --> 972.84] a Dagger, [972.96 --> 973.96] and it still is a Dagger [973.96 --> 974.82] even to this day. [975.82 --> 977.66] And I think that was [977.66 --> 978.64] one of the big changes [978.64 --> 979.38] that happened [979.38 --> 981.10] between me starting Ship It [981.10 --> 981.42] and then, [981.44 --> 981.60] you know, [981.64 --> 982.38] having to [982.38 --> 984.06] part ways [984.06 --> 985.50] at that point. [986.68 --> 987.12] 2021 [987.12 --> 988.78] was a very interesting year [988.78 --> 989.62] and it was not [989.62 --> 990.22] because of COVID, [990.56 --> 991.26] but that obviously, [991.36 --> 991.84] that did, [991.96 --> 992.72] that did make it [992.72 --> 993.84] interesting for everybody. [994.32 --> 995.16] But for me personally, [995.16 --> 996.08] I was transitioning [996.08 --> 998.00] into a startup again. [998.38 --> 998.82] I went from [998.82 --> 999.72] a large enterprise [999.72 --> 1000.62] at the time [1000.62 --> 1001.30] that was VMware [1001.30 --> 1003.16] and at VMware [1003.16 --> 1004.14] I have been working [1004.14 --> 1005.26] on RabbitMQ [1005.26 --> 1006.42] for, [1006.62 --> 1006.84] I think, [1006.94 --> 1007.90] six or seven years [1007.90 --> 1008.76] and I went through [1008.76 --> 1009.58] like different [1009.58 --> 1011.00] types of teams [1011.00 --> 1011.80] until eventually [1011.80 --> 1012.54] end up on the [1012.54 --> 1014.06] core RabbitMQ team. [1014.42 --> 1015.22] So you get all [1015.22 --> 1015.94] the Erlang, [1016.56 --> 1017.62] you get all the Make [1017.62 --> 1018.90] and there's a story [1018.90 --> 1019.64] there because it [1019.64 --> 1020.58] connects to Dagger [1020.58 --> 1022.32] and you just get [1022.32 --> 1023.70] to see a lot of [1023.84 --> 1025.10] really important systems, [1025.76 --> 1026.98] distributed systems, [1027.14 --> 1028.30] distributed systems problems [1028.30 --> 1029.92] and you realize [1029.92 --> 1030.60] how important [1030.60 --> 1031.48] the kernel is [1031.48 --> 1032.52] even when you're [1032.52 --> 1033.52] not using containers. [1034.14 --> 1035.10] So little differences [1035.10 --> 1036.38] between the different kernels [1036.38 --> 1037.56] can have a huge impact [1037.56 --> 1038.66] on how something [1038.66 --> 1039.48] like the Erlang VM [1039.48 --> 1040.00] behaves [1040.00 --> 1041.06] and these are [1041.06 --> 1041.68] really important [1041.68 --> 1042.22] applications [1042.22 --> 1043.34] like think for banks, [1043.50 --> 1044.54] financial institutions, [1045.32 --> 1046.50] GPS trackers [1046.50 --> 1048.50] and you may be thinking [1048.50 --> 1049.20] food deliveries [1049.20 --> 1049.98] but there's also [1049.98 --> 1051.22] some other GPS trackers [1051.22 --> 1052.08] which are really important [1052.08 --> 1053.32] they work correctly. [1053.84 --> 1054.60] Tills, [1054.92 --> 1055.94] payment systems, [1056.10 --> 1057.16] it's all over the place. [1057.66 --> 1058.36] At some point, [1058.46 --> 1059.58] we didn't realize this, [1060.22 --> 1061.22] some cars, [1061.62 --> 1062.60] the doors wouldn't open [1062.60 --> 1064.00] and RabbitMQ [1064.00 --> 1064.90] was in that stack. [1065.30 --> 1065.98] Like I was not [1065.98 --> 1066.76] expecting that. [1067.00 --> 1067.36] Like you know, [1067.72 --> 1068.40] you would honestly [1068.40 --> 1069.24] not expect that. [1069.40 --> 1069.84] It is wild [1069.84 --> 1070.46] like to see [1070.46 --> 1071.56] where tech ends up [1071.56 --> 1072.24] and how it ends up [1072.24 --> 1072.74] being used. [1073.30 --> 1074.14] A lot of the time [1074.14 --> 1074.98] it gets used wrong [1074.98 --> 1076.10] so having those [1076.10 --> 1076.64] conversations [1076.64 --> 1077.46] and going through [1077.46 --> 1078.00] those cycles [1078.00 --> 1078.58] when you have [1078.58 --> 1079.34] big teams [1079.34 --> 1080.62] and big budgets [1080.62 --> 1081.46] and big enterprises [1081.46 --> 1082.04] is fun [1082.04 --> 1083.08] but also [1083.08 --> 1084.12] it's a certain [1084.12 --> 1084.78] type of game. [1085.38 --> 1085.92] So after playing [1085.92 --> 1086.34] that game [1086.34 --> 1087.10] for like six, [1087.24 --> 1087.76] seven years [1087.76 --> 1088.84] something like that [1088.84 --> 1089.82] I said, [1089.94 --> 1090.26] you know what, [1090.28 --> 1091.02] it's time to go back [1091.02 --> 1091.94] to the startup world [1091.94 --> 1093.44] because I did start [1093.44 --> 1094.22] like on that journey [1094.22 --> 1095.52] before getting to VMware. [1095.98 --> 1096.88] We were a small startup. [1097.16 --> 1098.10] We were Cloud Credo. [1098.44 --> 1099.66] We were consultants [1099.66 --> 1101.34] for Cloud Foundry [1101.34 --> 1102.04] at the time [1102.04 --> 1102.62] and Bosch [1102.62 --> 1103.76] for those that [1103.76 --> 1105.42] remember Bosch [1105.42 --> 1106.56] maybe a few listeners [1106.56 --> 1106.92] will [1106.92 --> 1109.62] and Chef [1109.62 --> 1110.22] wasn't working [1110.22 --> 1111.16] for those systems [1111.16 --> 1112.32] so a team [1112.32 --> 1113.52] of 20 something people [1113.52 --> 1114.40] then we became [1114.40 --> 1114.82] Pivotal [1114.82 --> 1116.02] as in we were [1116.02 --> 1117.00] acquired by Pivotal [1117.00 --> 1117.82] but in my mind [1117.82 --> 1118.90] we took over Pivotal [1118.90 --> 1119.46] in some way [1119.46 --> 1120.88] because of that [1120.88 --> 1121.72] craziness [1121.72 --> 1122.50] the crazy spirit [1122.50 --> 1123.08] that we had [1123.08 --> 1124.04] and that worked [1124.04 --> 1124.48] really well [1124.48 --> 1125.42] so being part of [1125.42 --> 1126.20] Pivotal was great [1126.20 --> 1127.16] and pair programming [1127.16 --> 1128.14] and extreme programming [1128.14 --> 1129.46] that was at the core [1129.46 --> 1129.76] of it [1129.76 --> 1130.88] and then Pivotal [1130.88 --> 1131.72] eventually got acquired [1131.72 --> 1132.22] by VMware [1132.22 --> 1133.12] so those transitions [1133.12 --> 1134.68] from 20 to 2000 [1134.68 --> 1135.78] to 40,000 [1135.78 --> 1136.70] were huge jumps [1136.70 --> 1138.40] and huge changes [1138.40 --> 1140.02] so I wrote all of that [1140.02 --> 1140.36] and I said [1140.36 --> 1140.72] you know what [1140.72 --> 1141.54] it's time to go back [1141.54 --> 1142.34] to the startup world [1142.34 --> 1143.48] and that's where [1143.48 --> 1144.18] Dagger enters [1144.18 --> 1145.98] so Dagger was interesting [1145.98 --> 1147.26] because I was fascinated [1147.26 --> 1147.82] by Docker [1147.82 --> 1149.50] and I was working [1149.50 --> 1150.42] with Docker [1150.42 --> 1151.20] and using Docker [1151.20 --> 1152.58] but I haven't helped [1152.58 --> 1153.60] build Docker [1153.60 --> 1154.70] so Dagger [1154.70 --> 1155.66] was in the moment [1155.66 --> 1156.88] where I could try that [1156.88 --> 1158.20] and I took it [1158.20 --> 1159.70] and three years later [1159.70 --> 1160.48] here I am [1160.48 --> 1161.30] for anyone [1161.30 --> 1161.80] that doesn't know [1161.80 --> 1162.42] what Dagger is [1162.42 --> 1162.86] describe [1162.86 --> 1164.16] like what [1164.16 --> 1164.92] what are you doing [1164.92 --> 1165.54] or what does Dagger [1165.54 --> 1166.06] actually make [1166.06 --> 1166.52] as a product [1166.52 --> 1167.00] as a startup [1167.00 --> 1167.40] that's like [1167.40 --> 1168.26] hey we're going [1168.26 --> 1168.96] to change the world [1168.96 --> 1169.78] for this thing [1169.78 --> 1170.36] what is that [1170.36 --> 1171.96] so Dagger [1171.96 --> 1174.28] is what happens [1174.28 --> 1175.16] when you get tired [1175.16 --> 1175.88] of all the YAML [1175.88 --> 1177.34] when you get tired [1177.34 --> 1177.96] of all the YAML [1177.96 --> 1178.84] in your pipelines [1178.84 --> 1181.04] especially your CICD pipelines [1181.04 --> 1182.28] or when you get tired [1182.28 --> 1183.34] of your Jenkins file [1183.34 --> 1184.76] or when you get tired [1184.76 --> 1185.50] of your scripts [1185.50 --> 1187.36] you want something [1187.36 --> 1189.34] that scales [1189.34 --> 1190.52] with teams [1190.52 --> 1191.70] and with ideas [1191.70 --> 1192.76] that can [1192.76 --> 1194.14] it's really hard [1194.14 --> 1194.66] to capture them [1194.66 --> 1195.04] in YAML [1195.04 --> 1196.78] if you are finding [1196.78 --> 1197.48] yourself starting [1197.48 --> 1198.60] to template YAML [1198.60 --> 1200.16] for GitHub Actions [1200.16 --> 1201.02] or CircleCI [1201.02 --> 1202.56] or any CICD system [1202.56 --> 1204.20] you know you need Dagger [1204.20 --> 1205.64] the other option [1205.64 --> 1206.70] is to go towards [1206.70 --> 1207.22] Bazel [1207.22 --> 1208.06] and to go into [1208.06 --> 1208.62] that world [1208.62 --> 1209.64] but for anyone [1209.64 --> 1210.40] that knows [1210.40 --> 1210.86] that world [1210.86 --> 1211.32] and experiences [1211.32 --> 1211.84] that world [1211.84 --> 1212.52] knows that is [1212.52 --> 1213.52] a very heavyweight [1213.52 --> 1214.92] enterprise world [1214.92 --> 1217.20] so Dagger [1217.20 --> 1218.84] takes all the scripts [1218.84 --> 1220.06] and all the YAML [1220.06 --> 1221.76] and it allows you [1221.76 --> 1222.98] to capture that in code [1222.98 --> 1224.56] so what that means [1224.56 --> 1225.36] is that imagine [1225.36 --> 1226.76] writing your automation [1226.76 --> 1228.92] it can be a make file [1228.92 --> 1230.36] it can be your [1230.36 --> 1231.28] GitHub Actions YAML [1231.28 --> 1232.40] it can be your [1232.40 --> 1233.24] CircleCI config [1233.24 --> 1234.20] your Jenkins file [1234.20 --> 1235.12] all those things [1235.12 --> 1235.68] you can take [1235.68 --> 1237.44] and you can put them [1237.44 --> 1238.10] in the code [1238.10 --> 1239.84] that you are familiar with [1239.84 --> 1241.22] whether it's Python [1241.22 --> 1242.38] whether it's Go [1242.38 --> 1243.48] whether it's TypeScript [1243.48 --> 1245.92] and some more [1245.92 --> 1247.16] legacy languages [1247.16 --> 1247.96] which are still [1247.96 --> 1248.64] very much present [1248.64 --> 1249.34] like PHP [1249.34 --> 1251.04] or Elixir for example [1251.04 --> 1251.72] Rust [1251.72 --> 1252.42] some newer ones [1252.42 --> 1253.78] any of these languages [1253.78 --> 1254.50] you can use [1254.50 --> 1256.20] to write your automation [1256.20 --> 1258.24] and you can package it [1258.24 --> 1259.36] in something called modules [1259.36 --> 1260.72] you can distribute [1260.72 --> 1261.54] these modules [1261.54 --> 1262.20] as you would [1262.20 --> 1262.90] any package [1262.90 --> 1264.76] and you can assemble them [1264.76 --> 1266.02] just in time [1266.02 --> 1267.20] and you can combine them [1267.20 --> 1268.00] with other modules [1268.00 --> 1269.54] so what that means [1269.54 --> 1270.40] is that now [1270.40 --> 1271.86] you writing automation [1271.86 --> 1273.16] you integrating with CI CD [1273.16 --> 1274.52] is just a matter of [1274.52 --> 1276.10] calling the right function [1276.10 --> 1277.06] from the right module [1277.06 --> 1278.44] and making sure [1278.44 --> 1279.18] that that function [1279.18 --> 1280.38] gets wired [1280.38 --> 1281.22] with everything else [1281.22 --> 1282.10] that it needs [1282.10 --> 1284.38] so for example [1284.38 --> 1285.70] you have your tests [1285.70 --> 1285.98] right [1285.98 --> 1286.72] and build [1286.72 --> 1287.16] and you say [1287.16 --> 1287.68] okay but [1287.68 --> 1288.94] you could do this [1288.94 --> 1289.62] with make files [1289.62 --> 1290.74] or you could use [1290.74 --> 1291.72] a just file [1291.72 --> 1292.46] or you could use [1292.46 --> 1293.42] anything like that [1293.42 --> 1294.48] and that is true [1294.48 --> 1295.46] you can [1295.46 --> 1297.04] but what ends up [1297.04 --> 1297.74] happening with that [1297.74 --> 1298.52] is that [1298.52 --> 1299.94] there will be [1299.94 --> 1300.68] assumptions [1300.68 --> 1302.24] about the context [1302.24 --> 1303.82] in which that automation [1303.82 --> 1304.26] runs [1304.26 --> 1305.58] Dagger [1305.58 --> 1306.80] the only assumption [1306.80 --> 1307.38] which it makes [1307.38 --> 1307.96] is that there will be [1307.96 --> 1308.40] an engine [1308.40 --> 1309.40] which is a container [1309.40 --> 1309.80] runtime [1309.80 --> 1311.76] and in that container [1311.76 --> 1312.08] runtime [1312.08 --> 1313.22] you always have to specify [1313.22 --> 1314.18] hey which container [1314.18 --> 1315.22] image do you want to run [1315.22 --> 1316.42] which one do you want [1316.42 --> 1317.14] to start with [1317.14 --> 1319.04] so all these functions [1319.04 --> 1320.34] they always have a context [1320.34 --> 1321.90] and the context [1321.90 --> 1322.66] is the same [1322.66 --> 1323.84] regardless where you run [1323.84 --> 1325.38] which means [1325.38 --> 1326.12] that if you want [1326.12 --> 1326.98] to run this locally [1326.98 --> 1327.94] it will run [1327.94 --> 1329.08] exactly the same [1329.08 --> 1330.02] as it runs [1330.02 --> 1331.26] on any CI platform [1331.26 --> 1332.48] anywhere in the world [1332.48 --> 1334.28] you can run Dagger [1334.28 --> 1334.68] in Jenkins [1334.68 --> 1335.26] if you want [1335.26 --> 1336.44] it is an option [1336.44 --> 1337.64] I've been doing [1337.64 --> 1338.50] containers and make files [1338.50 --> 1339.40] for a long time [1339.40 --> 1339.96] right like that [1339.96 --> 1340.82] that assumption [1340.82 --> 1341.78] that I can [1341.78 --> 1343.18] I can run make [1343.18 --> 1344.48] in Jenkins [1344.48 --> 1345.66] or my local machine [1345.66 --> 1346.44] and it runs the same [1346.44 --> 1347.18] I don't need Dagger [1347.18 --> 1347.82] for that right [1347.82 --> 1348.68] correct [1348.68 --> 1349.82] it's like I can do [1349.82 --> 1350.84] I've been doing it [1350.84 --> 1351.06] right like [1351.06 --> 1351.98] all of [1351.98 --> 1352.96] yeah a lot of stuff [1352.96 --> 1353.42] just works like [1353.42 --> 1354.12] oh just execute [1354.12 --> 1354.62] the container [1354.62 --> 1355.52] here's some arguments [1355.52 --> 1356.06] for it like [1356.06 --> 1356.90] you know variables [1356.90 --> 1357.56] to the make file [1357.56 --> 1358.00] that's fine [1358.00 --> 1358.76] so that side of it [1358.76 --> 1359.58] doesn't really [1359.58 --> 1361.18] change how I've [1361.18 --> 1361.66] been doing things [1361.66 --> 1362.02] at least I know [1362.02 --> 1362.40] it does change [1362.40 --> 1362.92] how a lot of people [1362.92 --> 1363.24] are doing it [1363.24 --> 1364.56] because bash is prevalent [1364.56 --> 1366.00] the second thing here [1366.00 --> 1366.38] though I think [1366.38 --> 1367.22] that is interesting [1367.22 --> 1368.72] to me Dagger [1368.72 --> 1369.58] has always [1369.58 --> 1371.14] been almost [1371.14 --> 1372.64] the dagger [1372.64 --> 1373.32] if you will [1373.32 --> 1374.84] the blade [1374.84 --> 1375.44] for like [1375.44 --> 1376.22] the DevOps team [1376.22 --> 1377.74] the fact that [1377.74 --> 1378.46] how DevOps [1378.46 --> 1379.50] used to work [1379.50 --> 1379.88] for me [1379.88 --> 1380.74] at large enterprises [1380.74 --> 1381.78] was [1381.78 --> 1383.40] application teams [1383.40 --> 1383.76] would go write [1383.76 --> 1384.38] a bunch of code [1384.38 --> 1385.18] and the DevOps team [1385.18 --> 1385.48] would come in [1385.48 --> 1386.04] and like drop [1386.04 --> 1387.20] a Jenkins file [1387.20 --> 1387.64] in right [1387.64 --> 1388.28] here's your PR [1388.28 --> 1389.20] for your Jenkins file [1389.20 --> 1390.54] now this is all [1390.54 --> 1390.88] going to work [1390.88 --> 1391.26] magically [1392.02 --> 1392.46] team [1392.46 --> 1393.46] at Disney Plus [1393.46 --> 1393.84] that was like [1393.84 --> 1395.02] a Lib Jenkins team [1395.02 --> 1395.74] like we wrote [1395.74 --> 1396.32] libraries [1396.32 --> 1397.28] I didn't do it [1397.28 --> 1397.96] I hate Jenkins file [1397.96 --> 1398.66] but they wrote [1398.66 --> 1399.36] they wrote libraries [1399.36 --> 1401.80] for this groovy script [1401.80 --> 1402.60] so that everyone [1402.60 --> 1403.34] in their Jenkins file [1403.34 --> 1404.02] would import [1404.02 --> 1405.52] that team's Lib Jenkins [1405.52 --> 1406.40] and then it would do [1406.40 --> 1406.88] a bunch of stuff [1406.88 --> 1407.24] for them [1407.24 --> 1407.88] by default [1407.88 --> 1409.22] but that's so required [1409.22 --> 1410.20] there was some [1410.20 --> 1410.98] other team [1410.98 --> 1412.16] doing some other [1412.16 --> 1412.62] thing [1412.62 --> 1414.02] in some other language [1414.02 --> 1415.10] that was external [1415.10 --> 1416.12] to the application team [1416.12 --> 1417.90] and my sense [1417.90 --> 1418.34] with Dagger [1418.34 --> 1419.18] is the fact that [1419.18 --> 1420.36] Dagger requires [1420.36 --> 1421.38] that the application team [1421.38 --> 1423.12] now owns CICD [1423.12 --> 1424.70] as like they are the ones [1424.70 --> 1425.46] because like the [1425.46 --> 1426.84] familiarity of code [1426.84 --> 1428.12] doesn't matter [1428.12 --> 1429.04] if you're not the one [1429.04 --> 1429.80] writing the code [1429.80 --> 1430.28] right like if you're [1430.28 --> 1431.16] some external person [1431.16 --> 1431.58] you're going to have [1431.58 --> 1432.14] your own [1432.14 --> 1433.64] pep formatting [1433.64 --> 1434.26] for Python [1434.26 --> 1435.06] and your own [1435.06 --> 1435.76] you know modules [1435.76 --> 1436.28] and formatting [1436.28 --> 1436.76] so that's like [1436.76 --> 1437.30] it's not going to [1437.30 --> 1437.86] jive well [1437.86 --> 1438.76] with the application team [1438.76 --> 1439.42] so in this case [1439.42 --> 1441.74] Dagger makes the most sense [1441.74 --> 1442.56] when the team [1442.56 --> 1443.36] writing the code [1443.36 --> 1444.98] is also the one [1444.98 --> 1446.02] doing the CICD [1446.02 --> 1446.54] is that right? [1446.54 --> 1447.36] yes [1447.36 --> 1448.78] that is a valid take [1448.78 --> 1449.48] for sure [1449.48 --> 1450.98] we see [1450.98 --> 1452.50] different teams [1452.50 --> 1454.26] and different companies [1454.26 --> 1454.82] use Dagger [1454.82 --> 1455.56] in different ways [1455.56 --> 1457.04] at this point [1457.04 --> 1457.42] we've seen [1457.42 --> 1458.32] every which way [1458.32 --> 1459.54] and they're all valid [1459.54 --> 1460.84] the point is [1460.84 --> 1461.60] that it forces [1461.60 --> 1463.48] the different perspectives [1463.48 --> 1464.88] to come together [1464.88 --> 1465.78] as code [1465.78 --> 1467.26] so forget like [1467.26 --> 1467.94] a make file [1467.94 --> 1469.06] or Jenkins file [1469.06 --> 1469.76] or a script [1469.76 --> 1470.28] or anything [1470.28 --> 1470.58] just [1470.58 --> 1471.44] we will be writing [1471.44 --> 1472.02] the code [1472.02 --> 1473.30] that our company [1473.30 --> 1474.72] is most familiar with [1474.72 --> 1476.52] whether you're a DevOps person [1476.52 --> 1478.34] whether you're someone [1478.34 --> 1478.92] in the community [1478.92 --> 1479.76] that wrote a module [1479.76 --> 1480.64] it doesn't really matter [1480.64 --> 1481.68] the point is [1481.68 --> 1482.70] we will all be looking [1482.70 --> 1483.42] at the same code [1483.42 --> 1485.04] we understand the same code [1485.04 --> 1486.70] we can contribute [1486.70 --> 1487.44] to the same code [1487.44 --> 1488.52] and all the automation [1488.52 --> 1489.82] ends up being code [1489.82 --> 1490.90] that we can run locally [1490.90 --> 1493.14] from wherever it is [1493.14 --> 1494.36] so just to give you [1494.36 --> 1494.80] an example [1494.80 --> 1496.30] of how powerful this is [1496.30 --> 1497.66] let's say that [1497.66 --> 1498.40] you will take [1498.40 --> 1499.60] the Dagger repository [1499.60 --> 1500.60] as is today [1500.60 --> 1502.26] the Dagger repository [1502.26 --> 1503.26] has a module [1503.26 --> 1504.58] for the entire repository [1504.58 --> 1507.08] which encapsulates [1507.08 --> 1507.92] all the things [1507.92 --> 1508.68] that can happen [1508.68 --> 1509.54] in that repository [1509.54 --> 1511.48] so without you knowing [1511.48 --> 1512.08] anything [1512.08 --> 1513.18] about [1513.18 --> 1514.32] how it runs [1514.32 --> 1515.08] or what's needed [1515.08 --> 1516.38] you can [1516.38 --> 1517.50] start discovering [1517.50 --> 1518.48] what is possible [1518.48 --> 1519.08] in this module [1519.08 --> 1520.18] for example [1520.18 --> 1521.52] build me the docs [1521.52 --> 1522.88] but also [1522.88 --> 1523.70] serve the docs [1523.70 --> 1524.58] you have one command [1524.58 --> 1525.44] that will build [1525.44 --> 1526.42] lint [1526.42 --> 1527.46] serve the docs [1527.46 --> 1528.74] on your local machine [1528.74 --> 1529.62] the Dagger docs [1529.62 --> 1530.46] without you knowing [1530.46 --> 1531.18] anything [1531.18 --> 1532.68] or having to install [1532.68 --> 1533.16] anything [1533.16 --> 1534.42] apart from the Dagger CLI [1534.42 --> 1536.10] in the same way [1536.10 --> 1537.26] you could build yourself [1537.26 --> 1538.12] a Dagger CLI [1538.12 --> 1539.02] if you wanted to [1539.02 --> 1539.94] once you discover [1539.94 --> 1540.90] what the command is [1540.90 --> 1542.42] and it's all self-documenting [1542.42 --> 1543.02] it's all there [1543.02 --> 1543.90] so it provides [1543.90 --> 1545.22] a very nice way [1545.22 --> 1546.20] of consuming things [1546.20 --> 1547.34] without you knowing [1547.34 --> 1548.02] much about [1548.02 --> 1549.04] what this piece [1549.04 --> 1549.86] of software is [1549.86 --> 1550.72] it's almost like [1550.72 --> 1551.86] an API to code [1551.86 --> 1553.50] but an API [1553.50 --> 1554.74] to consuming code [1554.74 --> 1555.84] to consuming resources [1555.84 --> 1557.02] I don't care [1557.02 --> 1557.62] whether it's Python [1557.62 --> 1558.36] whether it's PHP [1558.36 --> 1559.50] what I want [1559.50 --> 1560.22] is the artifact [1560.22 --> 1561.02] or what I want [1561.02 --> 1561.64] is the docs [1561.64 --> 1562.26] or what I want [1562.26 --> 1563.64] is the auto-completion [1563.64 --> 1564.64] whatever the case may be [1564.64 --> 1565.98] so how do you [1565.98 --> 1566.56] encapsulate that [1566.56 --> 1567.10] in a way [1567.10 --> 1568.70] that others can understand [1568.70 --> 1569.36] and consume it [1569.36 --> 1569.96] in an easy way [1569.96 --> 1571.40] well but there's [1571.40 --> 1572.50] a separation there [1572.50 --> 1572.92] of like [1572.92 --> 1574.36] understanding something [1574.36 --> 1575.20] and using something [1575.20 --> 1575.76] without needing [1575.76 --> 1576.46] to understand it [1576.46 --> 1576.58] right [1576.58 --> 1577.00] because again [1577.00 --> 1577.30] the [1577.30 --> 1578.64] I can [1578.64 --> 1579.64] I've written [1579.64 --> 1580.24] plenty of [1580.24 --> 1581.44] things that were [1581.44 --> 1581.72] just like [1581.72 --> 1582.14] you just run [1582.14 --> 1582.82] make docs [1582.82 --> 1583.66] and docs are there [1583.66 --> 1583.98] for you [1583.98 --> 1584.10] right [1584.10 --> 1584.84] like make docs [1584.84 --> 1585.22] dev [1585.22 --> 1585.64] and they're like [1585.64 --> 1585.96] it's like [1585.96 --> 1586.78] they don't need to know [1586.78 --> 1587.32] what's behind [1587.32 --> 1588.26] the make file [1588.26 --> 1588.98] and it's running [1588.98 --> 1589.56] containers still [1589.56 --> 1590.34] it's doing that stuff [1590.34 --> 1591.34] the thing that I think [1591.34 --> 1592.06] is really interesting [1592.06 --> 1592.96] here is the fact [1592.96 --> 1593.38] that like [1593.38 --> 1594.22] those modules [1594.22 --> 1594.98] are shareable [1594.98 --> 1596.14] and the modules [1596.14 --> 1598.10] are something [1598.10 --> 1598.80] really powerful [1598.80 --> 1599.60] that Terraform [1599.60 --> 1600.06] did for us [1600.06 --> 1600.16] right [1600.16 --> 1600.60] like Terraform [1600.60 --> 1601.20] modules were [1601.20 --> 1601.56] powerful [1601.56 --> 1602.06] because you're [1602.06 --> 1602.32] just like [1602.32 --> 1603.18] you don't have [1603.18 --> 1603.64] to know [1603.64 --> 1604.66] behind the scenes [1604.66 --> 1605.70] and granted [1605.70 --> 1606.24] at some point [1606.24 --> 1606.66] you might need [1606.66 --> 1607.60] to escape the module [1607.60 --> 1608.04] you might need [1608.04 --> 1608.84] to override the module [1608.84 --> 1609.18] you might need [1609.18 --> 1609.84] to go build [1609.84 --> 1610.50] your own module [1610.50 --> 1611.28] but you can get [1611.28 --> 1612.28] started with [1612.28 --> 1612.72] something that [1612.72 --> 1613.62] has some opinions [1613.62 --> 1616.06] on how we think [1616.06 --> 1616.46] you should be [1616.46 --> 1617.12] doing this [1617.12 --> 1618.40] and in the dagger [1618.40 --> 1618.78] sense [1618.78 --> 1619.82] and in the Terraform [1619.82 --> 1620.08] sense [1620.08 --> 1621.16] most of the time [1621.16 --> 1621.88] those things [1621.88 --> 1622.38] are just going [1622.38 --> 1623.12] to work for you [1623.12 --> 1623.82] without needing [1623.82 --> 1624.24] to care [1624.24 --> 1624.56] right [1624.56 --> 1625.02] so it's like [1625.02 --> 1625.38] I can [1625.38 --> 1626.44] dagger [1626.44 --> 1626.94] and it [1626.94 --> 1627.30] or whatever [1627.30 --> 1628.08] I can start [1628.08 --> 1628.66] off with the module [1628.66 --> 1629.02] like oh [1629.02 --> 1629.96] this is the thing [1629.96 --> 1630.40] I wanted [1630.40 --> 1631.38] this looks right [1631.38 --> 1632.20] I'm going to go [1632.20 --> 1633.42] with the defaults [1633.42 --> 1633.84] and if I need [1633.84 --> 1634.28] to change it [1634.28 --> 1634.86] I can [1634.86 --> 1636.42] so essentially [1636.42 --> 1637.28] can you get rid [1637.28 --> 1638.18] of the make file [1638.18 --> 1638.94] and all of that [1638.94 --> 1640.10] with dagger [1640.10 --> 1641.20] and just have [1641.20 --> 1642.78] like say [1642.78 --> 1643.60] you were [1643.60 --> 1645.08] building a startup [1645.08 --> 1645.86] or application [1645.86 --> 1646.80] and [1646.80 --> 1648.32] you didn't [1648.32 --> 1648.70] have [1648.70 --> 1649.16] the [1649.16 --> 1650.66] experience [1650.66 --> 1651.34] that Justin [1651.34 --> 1651.74] has [1651.74 --> 1652.52] right [1652.52 --> 1653.14] and you [1653.14 --> 1653.52] just needed [1653.52 --> 1654.10] to figure out [1654.10 --> 1654.56] how to make [1654.56 --> 1655.40] your CI, [1655.48 --> 1655.64] CD [1655.64 --> 1656.12] and like [1656.12 --> 1656.70] your DevOps [1656.70 --> 1657.94] like just [1657.94 --> 1658.74] whole realm [1658.74 --> 1659.30] work [1659.30 --> 1659.96] and you don't [1659.96 --> 1660.74] have that experience [1660.74 --> 1661.32] does this now [1661.32 --> 1661.86] enable you [1661.86 --> 1662.36] to skip [1662.36 --> 1663.60] all of those [1663.60 --> 1664.38] extra files [1664.38 --> 1665.28] in different ways [1665.28 --> 1665.88] and just have [1665.88 --> 1666.46] your whole team [1666.46 --> 1667.24] learn how to [1667.24 --> 1668.18] do your infrastructure [1668.18 --> 1668.92] in that way [1668.92 --> 1669.64] using dagger [1669.64 --> 1670.80] yes [1670.80 --> 1671.92] if you would [1671.92 --> 1672.80] go to the modules [1672.80 --> 1673.36] that Justin [1673.36 --> 1674.24] if so first of all [1674.24 --> 1674.76] it would require [1674.76 --> 1675.84] Justin to take time [1675.84 --> 1677.32] to write the modules [1677.32 --> 1678.40] and to share them [1678.40 --> 1679.02] so that others [1679.02 --> 1679.94] can discover them [1679.94 --> 1681.24] it's just a matter [1681.24 --> 1681.66] of basically [1681.66 --> 1682.24] putting him [1682.24 --> 1683.02] on his [1683.02 --> 1683.78] GitHub repository [1683.78 --> 1685.36] the convention [1685.36 --> 1686.08] is Daggerverse [1686.08 --> 1687.90] so many people [1687.90 --> 1688.84] have the Daggerverse [1688.84 --> 1689.12] repo [1689.12 --> 1690.24] which is a [1690.24 --> 1691.04] collection [1691.04 --> 1691.60] of the different [1691.60 --> 1692.08] modules [1692.08 --> 1692.84] that people [1692.84 --> 1693.52] use and wrote [1693.52 --> 1694.92] so at this point [1694.92 --> 1696.58] there's I think [1696.58 --> 1697.36] five or six [1697.36 --> 1698.30] implementations [1698.30 --> 1699.32] of the Go [1699.32 --> 1700.08] module [1700.08 --> 1701.40] which does all [1701.40 --> 1701.94] things around [1701.94 --> 1702.68] Go applications [1702.68 --> 1703.38] testing [1703.38 --> 1704.14] building [1704.14 --> 1704.74] linting [1704.74 --> 1705.42] all those things [1705.42 --> 1706.50] so you're right [1706.50 --> 1706.88] in the sense [1706.88 --> 1707.28] that you would [1707.28 --> 1708.30] need to figure out [1708.30 --> 1709.26] how to write [1709.26 --> 1709.86] that automation [1709.86 --> 1710.66] just a matter [1710.66 --> 1711.70] of using it [1711.70 --> 1712.66] you have a Go app [1712.66 --> 1713.20] great [1713.20 --> 1714.04] this is how [1714.04 --> 1714.42] you provide [1714.42 --> 1714.92] the source [1714.92 --> 1716.02] and this is [1716.02 --> 1716.76] what options [1716.76 --> 1717.20] are available [1717.20 --> 1717.62] to you [1717.62 --> 1718.80] and you can [1718.80 --> 1719.22] try running [1719.22 --> 1719.62] it locally [1719.62 --> 1720.82] it works [1720.82 --> 1721.30] great [1721.30 --> 1721.92] so how do I [1721.92 --> 1722.50] run this in CI [1722.50 --> 1723.80] the same commands [1723.80 --> 1724.86] the same commands [1724.86 --> 1725.28] that you'd run [1725.28 --> 1725.62] locally [1725.62 --> 1726.42] you'd put in CI [1726.42 --> 1726.92] and it would [1726.92 --> 1727.48] work exactly [1727.48 --> 1727.90] the same [1727.90 --> 1729.24] no more figuring [1729.24 --> 1729.88] of YAML [1729.88 --> 1730.56] no more figuring [1730.56 --> 1731.22] of a lot of [1731.22 --> 1731.52] things like [1731.52 --> 1731.88] caching [1731.88 --> 1732.56] for example [1732.56 --> 1733.20] I mean [1733.20 --> 1734.00] we haven't even [1734.00 --> 1734.68] like unpacked [1734.68 --> 1736.18] this aspect [1736.18 --> 1736.60] of Dagger [1736.60 --> 1737.60] how it caches [1737.60 --> 1739.12] how it for example [1739.12 --> 1740.32] sends open [1740.32 --> 1741.12] telemetry traces [1741.12 --> 1742.28] for every single [1742.28 --> 1742.96] operation which [1742.96 --> 1743.84] happens inside [1743.84 --> 1744.10] of it [1744.10 --> 1744.46] there's like [1744.46 --> 1745.10] so much [1745.10 --> 1745.38] here [1745.38 --> 1746.60] at a surface [1746.60 --> 1747.46] it looks like [1747.46 --> 1748.34] it's a replacement [1748.34 --> 1749.10] for your scripts [1749.10 --> 1750.92] a way of [1750.92 --> 1751.86] embedding this [1751.86 --> 1752.30] knowledge and [1752.30 --> 1752.86] sharing it with [1752.86 --> 1753.20] others [1753.20 --> 1754.44] but other tools [1754.44 --> 1754.94] have done this [1754.94 --> 1755.34] before [1755.34 --> 1756.60] so how is this [1756.60 --> 1756.92] special [1756.92 --> 1757.86] it's all the [1757.86 --> 1758.68] other things [1758.68 --> 1759.74] which we haven't [1759.74 --> 1760.34] gotten into [1760.34 --> 1761.18] which makes it [1761.18 --> 1763.18] a very [1763.18 --> 1764.84] comprehensive way [1764.84 --> 1766.32] of putting [1766.32 --> 1767.26] automation in [1767.26 --> 1767.52] code [1767.52 --> 1768.34] sharing it with [1768.34 --> 1768.64] others [1768.64 --> 1769.44] and letting [1769.44 --> 1770.46] others reuse it [1770.46 --> 1771.00] rather than [1771.00 --> 1771.88] everyone having to [1771.88 --> 1773.08] write the same [1773.08 --> 1773.56] thing in their [1773.56 --> 1774.62] own specific way [1774.62 --> 1775.36] I think the [1775.36 --> 1775.82] code piece for [1775.82 --> 1776.24] me is always [1776.24 --> 1776.72] kind of that [1776.72 --> 1777.64] barrier for a [1777.64 --> 1778.08] lot of people [1778.08 --> 1779.94] because saying [1779.94 --> 1780.48] you get to [1780.48 --> 1781.04] write all your [1781.04 --> 1781.76] automation in [1781.76 --> 1782.20] the code you're [1782.20 --> 1782.80] familiar with [1782.80 --> 1784.38] is a lockout [1784.38 --> 1784.84] for a lot of [1784.84 --> 1785.22] people because [1785.22 --> 1785.42] a lot of [1785.42 --> 1785.80] people don't [1785.80 --> 1786.04] they're not [1786.04 --> 1786.50] comfortable with [1786.50 --> 1786.74] code [1786.74 --> 1787.26] they're not [1787.26 --> 1787.92] all the people [1787.92 --> 1788.58] writing YAML [1788.58 --> 1789.30] are expert [1789.30 --> 1790.38] YAML engineers [1790.38 --> 1792.46] they know how [1792.46 --> 1793.62] the GitLab [1793.62 --> 1794.26] YAML is going [1794.26 --> 1794.92] to do something [1794.92 --> 1795.66] different on [1795.66 --> 1796.58] something that [1796.58 --> 1798.08] has a list [1798.08 --> 1798.72] versus an array [1798.72 --> 1799.48] whatever it is [1799.48 --> 1799.76] right like [1799.76 --> 1800.12] they're like [1800.12 --> 1800.92] I know this [1800.92 --> 1801.88] thing but once [1801.88 --> 1802.46] you ask me to [1802.46 --> 1802.92] like go write [1802.92 --> 1803.68] some go code [1803.68 --> 1805.00] I'm not as [1805.00 --> 1805.60] familiar with [1805.60 --> 1806.02] that and I [1806.02 --> 1806.58] feel a little [1806.58 --> 1807.72] bit out of [1807.72 --> 1808.90] place if I'm [1808.90 --> 1809.44] the one that [1809.44 --> 1809.98] is like an [1809.98 --> 1810.84] external person [1810.84 --> 1812.04] going into do [1812.04 --> 1812.64] and maintain [1812.64 --> 1813.54] and do this [1813.54 --> 1814.16] thing for another [1814.16 --> 1815.12] team that's an [1815.12 --> 1815.64] application team [1815.64 --> 1815.88] right because [1815.88 --> 1817.02] like a lot of [1817.02 --> 1817.36] companies I've [1817.36 --> 1817.70] worked at the [1817.70 --> 1818.32] app teams were [1818.32 --> 1818.90] always this like [1818.90 --> 1819.74] high level like [1819.74 --> 1820.24] you don't mess [1820.24 --> 1820.86] with them like [1820.86 --> 1821.24] they're the ones [1821.24 --> 1821.94] making the money [1821.94 --> 1823.14] and all you other [1823.14 --> 1823.62] people are just [1823.62 --> 1824.82] overhead and [1824.82 --> 1825.30] you're just like [1825.30 --> 1825.84] oh you write the [1825.84 --> 1826.30] YAML everyone [1826.30 --> 1827.02] else writes code [1827.02 --> 1827.80] right and like that [1827.80 --> 1828.34] I think is a [1828.34 --> 1829.06] barrier for a lot [1829.06 --> 1829.48] of people [1829.48 --> 1830.60] I think it's also [1830.60 --> 1831.46] perspective though [1831.46 --> 1832.06] because think about [1832.06 --> 1832.64] it when you go to [1832.64 --> 1833.46] school for computer [1833.46 --> 1834.36] science or you [1834.36 --> 1835.30] first get into [1835.30 --> 1836.48] like computers [1836.48 --> 1837.64] you're writing [1837.64 --> 1838.92] Python or Java [1838.92 --> 1840.00] right and nobody [1840.00 --> 1840.64] tells you about [1840.64 --> 1841.36] DevOps nobody [1841.36 --> 1841.96] tells you about [1841.96 --> 1842.94] scripting unless [1842.94 --> 1844.54] you somehow just [1844.54 --> 1845.36] stumble on that so [1845.36 --> 1846.78] I think for people [1846.78 --> 1847.76] that maybe started [1847.76 --> 1848.94] in systems or [1848.94 --> 1849.62] started with [1849.62 --> 1851.00] scripting yeah but [1851.00 --> 1852.40] it's a huge barrier [1852.40 --> 1853.38] for people who just [1853.38 --> 1854.36] started writing the [1854.36 --> 1855.62] high level code and [1855.62 --> 1856.28] then all of a sudden [1856.28 --> 1856.96] they're thrown into [1856.96 --> 1857.46] a production [1857.46 --> 1858.28] environment well now [1858.28 --> 1858.96] they have to manage [1858.96 --> 1860.56] infrastructure like I [1860.56 --> 1861.24] think you have more [1861.24 --> 1862.18] people coming from [1862.18 --> 1863.38] the like writing [1863.38 --> 1864.34] code because like [1864.34 --> 1864.92] when you think about [1864.92 --> 1865.56] it when you hear [1865.56 --> 1867.62] boot camps or school [1867.62 --> 1868.98] or you go to these [1868.98 --> 1870.32] different like ways [1870.32 --> 1871.06] that people are being [1871.06 --> 1872.16] educated to like get [1872.16 --> 1873.14] into the tech industry [1873.14 --> 1874.42] everybody tells you [1874.42 --> 1875.12] about the code and [1875.12 --> 1875.92] nobody tells you about [1875.92 --> 1877.06] the scripting and the [1877.06 --> 1878.98] version control and [1878.98 --> 1881.50] the DevOps and CICD [1881.50 --> 1883.10] right so like I think [1883.10 --> 1884.94] this is a great tool [1884.94 --> 1886.10] for those people and [1886.10 --> 1886.74] making it more [1886.74 --> 1887.94] accessible to get into [1887.94 --> 1889.22] DevOps and CICD and [1889.22 --> 1890.30] to be able to maintain [1890.30 --> 1891.02] your infrastructure [1891.02 --> 1892.38] because everybody [1892.38 --> 1893.18] talks about making [1893.18 --> 1894.22] this really cool app [1894.22 --> 1894.96] and nobody talks about [1894.96 --> 1895.90] maintaining that really [1895.90 --> 1896.78] cool app and releasing [1896.78 --> 1898.04] it you know that's why [1898.04 --> 1898.56] we're here that's what [1898.56 --> 1899.30] this podcast goes for [1899.30 --> 1900.96] yeah exactly but and [1900.96 --> 1901.64] I think that's a great [1901.64 --> 1902.30] point where it's like [1902.30 --> 1903.34] it's a it's a barrier [1903.34 --> 1904.78] both ways there's the [1904.78 --> 1905.66] person that's only ever [1905.66 --> 1906.90] written web apps with [1906.90 --> 1907.82] spring boots like [1907.82 --> 1908.74] they're like I don't I [1908.74 --> 1909.12] don't know what this [1909.12 --> 1910.08] Docker thing is like [1910.08 --> 1911.80] don't don't like you [1911.80 --> 1912.24] know I'm not going to [1912.24 --> 1913.50] go into the weeds of [1913.50 --> 1914.34] something that's not my [1914.34 --> 1915.26] thing it's like actually [1915.26 --> 1916.76] it's all your thing and [1916.76 --> 1917.34] same thing for the [1917.34 --> 1918.14] system people like I [1918.14 --> 1919.70] don't know how the JVM [1919.70 --> 1920.38] is going to do that [1920.38 --> 1920.94] thing like well you [1920.94 --> 1921.74] better learn because [1921.74 --> 1922.84] that's also your thing [1922.84 --> 1923.68] because I feel like [1923.68 --> 1924.34] when they like they'll [1924.34 --> 1925.10] be like okay you have [1925.10 --> 1926.02] to learn front end but [1926.02 --> 1926.66] then you have to learn [1926.66 --> 1927.56] the back end and then [1927.56 --> 1928.08] you have to learn how [1928.08 --> 1929.34] to fix how to make that [1929.34 --> 1930.12] connect to a database [1930.12 --> 1931.20] but nobody's ever [1931.20 --> 1932.24] talking about how to [1932.24 --> 1934.32] like keep all of this [1934.32 --> 1935.56] working together and [1935.56 --> 1936.74] maintaining it and [1936.74 --> 1938.24] yeah you know so it's [1938.24 --> 1940.30] like I felt like all the [1940.30 --> 1941.30] whole time during school [1941.30 --> 1941.98] and going through [1941.98 --> 1943.22] training at AWS and [1943.22 --> 1943.84] going through different [1943.84 --> 1944.82] boot camps and everything [1944.82 --> 1945.94] like I was like why are [1945.94 --> 1947.52] we always like we [1947.52 --> 1948.54] talk about the main [1948.54 --> 1950.00] like the big main [1950.00 --> 1950.96] products but nobody [1950.96 --> 1951.76] talks about all the [1951.76 --> 1952.56] secrets you need to [1952.56 --> 1953.36] make all of it work [1953.36 --> 1954.34] together you know [1954.34 --> 1956.10] one analogy that I [1956.10 --> 1956.68] think works really [1956.68 --> 1958.12] well when you're [1958.12 --> 1958.90] trying to approach [1958.90 --> 1959.44] dagger and you're [1959.44 --> 1960.06] trying just like to [1960.06 --> 1960.74] figure out like where [1960.74 --> 1961.74] does this thing fit in [1961.74 --> 1963.82] imagine that what [1963.82 --> 1964.78] you're building is a [1964.78 --> 1966.46] software factory right [1966.46 --> 1967.68] your startup your [1967.68 --> 1968.84] company your team [1968.84 --> 1969.60] whatever it is you [1969.60 --> 1970.54] are a software factory [1970.54 --> 1972.44] you are delivering value [1972.44 --> 1973.56] to your users in the [1973.56 --> 1974.40] form of software or [1974.40 --> 1975.84] through software great [1975.84 --> 1977.24] where's the thing [1977.24 --> 1978.82] that helps you [1978.82 --> 1979.82] maintain your factory [1979.82 --> 1980.88] where's the thing [1980.88 --> 1981.80] that helps you do all [1981.80 --> 1982.70] the things that you [1982.70 --> 1984.58] normally I mean there's [1984.58 --> 1985.60] not that much value in [1985.60 --> 1986.24] figuring out for [1986.24 --> 1987.16] example how to run your [1987.16 --> 1988.36] tests right or how to [1988.36 --> 1989.82] cache them properly or [1989.82 --> 1993.12] how to maybe lint [1993.12 --> 1994.26] that's another one how [1994.26 --> 1995.90] to package I mean how [1995.90 --> 1997.00] many hours do you want [1997.00 --> 1997.92] to spend figuring out [1997.92 --> 1999.66] how to package a JVM [1999.66 --> 2000.50] container or Java [2000.50 --> 2002.82] container seriously I [2002.82 --> 2004.16] mean sure there's like [2004.16 --> 2005.02] recipes that you can [2005.02 --> 2006.08] copy and AI is very [2006.08 --> 2007.98] helpful but what if [2007.98 --> 2009.22] some people that were [2009.22 --> 2010.10] really passionate about [2010.10 --> 2010.98] this and had the time [2010.98 --> 2012.86] they encoded this in [2012.86 --> 2013.86] Java let's say using [2013.86 --> 2015.14] Java and then just get [2015.14 --> 2016.60] to use it wouldn't that [2016.60 --> 2018.20] be nice but not just [2018.20 --> 2019.12] that when you have AI [2019.12 --> 2020.22] do it it's very hard to [2020.22 --> 2021.40] get the context but if [2021.40 --> 2022.40] somebody else on your [2022.40 --> 2023.64] team and that or [2023.64 --> 2024.62] somebody else that works [2024.62 --> 2025.52] at the same company as [2025.52 --> 2026.40] you that you can ping [2026.40 --> 2027.34] and be like hey I want [2027.34 --> 2028.34] to reuse this but I [2028.34 --> 2029.08] don't understand this [2029.08 --> 2030.58] part you know yeah and [2030.58 --> 2031.40] if you give people make [2031.40 --> 2032.88] files or Jenkins files or [2032.88 --> 2034.34] YAML what do they say [2034.34 --> 2035.16] mean when was the last [2035.16 --> 2035.90] time you received like [2035.90 --> 2036.66] a YAML package and [2036.66 --> 2037.36] you said wow this is [2037.36 --> 2038.20] cool I'm going to use [2038.20 --> 2039.68] this this is amazing [2039.68 --> 2040.66] when did someone get [2040.66 --> 2041.46] excited about Jenkins [2041.46 --> 2043.46] file rarely not just [2043.46 --> 2044.50] that but it never works [2044.50 --> 2045.52] the same for everybody [2045.52 --> 2048.64] exactly like everyone's [2048.64 --> 2049.58] like use this and it's [2049.58 --> 2050.48] so easy and you're like [2050.48 --> 2053.26] is it though I mean [2053.26 --> 2054.02] same thing can be said [2054.02 --> 2056.08] for a lot of like code [2056.08 --> 2056.80] though right like like [2056.80 --> 2058.90] so many NPM modules and [2058.90 --> 2059.92] Python things I've tried [2059.92 --> 2060.70] to solve like this is [2060.70 --> 2061.80] broken right like what [2061.80 --> 2062.76] what is the expectations [2062.76 --> 2063.60] of this thing to work [2063.60 --> 2064.20] and I think that [2064.20 --> 2066.02] encapsulating that in a [2066.02 --> 2068.66] holistic container that [2068.66 --> 2069.48] can do some of that has [2069.48 --> 2070.76] has been the solve for a [2070.76 --> 2071.54] lot of those things but [2071.54 --> 2072.34] once I'm like I need to [2072.34 --> 2073.64] go write some software [2073.64 --> 2075.26] around this thing but I [2075.26 --> 2076.56] can't even run this thing [2076.56 --> 2078.96] is a problem yeah and [2078.96 --> 2079.44] when it comes to [2079.44 --> 2080.42] introspecting it's okay [2080.42 --> 2081.44] so great so I have this [2081.44 --> 2083.02] command which I run great [2083.02 --> 2083.88] I'll figure out how to [2083.88 --> 2084.64] run it let's see it's in [2084.64 --> 2086.10] the make file so what is [2086.10 --> 2086.94] actually happening in [2086.94 --> 2087.60] this thing when it [2087.60 --> 2089.30] executes how can I [2089.30 --> 2090.92] visualize the execution [2090.92 --> 2092.64] of the different steps [2092.64 --> 2094.50] how do I know when a [2094.50 --> 2095.32] step runs or doesn't [2095.32 --> 2096.20] run I mean I used to [2096.20 --> 2097.64] love make I still have [2097.64 --> 2098.66] make files around which [2098.66 --> 2099.82] I still use to this day [2099.82 --> 2102.04] and they're great just [2102.04 --> 2103.30] files same thing only [2103.30 --> 2104.50] recently on the Kaizen we [2104.50 --> 2105.64] talked about just files [2105.64 --> 2107.76] and how just makes more [2107.76 --> 2108.94] sense in the context of [2108.94 --> 2110.42] changelog that's great [2110.42 --> 2111.76] you know if you're happy [2111.76 --> 2112.50] with your make file if [2112.50 --> 2113.00] you're happy with your [2113.00 --> 2114.06] just file any automation [2114.06 --> 2114.92] that you have that works [2114.92 --> 2116.92] for you keep it it's a [2116.92 --> 2117.78] good thing it's an asset [2117.78 --> 2119.58] it's not a liability but [2119.58 --> 2120.68] when that thing stops [2120.68 --> 2122.32] working when you get [2122.32 --> 2124.04] frustrated when you get [2124.04 --> 2124.92] all the issues with [2124.92 --> 2126.56] YAML and all the things [2126.56 --> 2127.54] that you know you've been [2127.54 --> 2128.56] maybe toiling away for [2128.56 --> 2130.30] years and years when you [2130.30 --> 2131.44] will consider something [2131.44 --> 2132.90] better have a look and [2132.90 --> 2133.80] see if that makes sense [2133.80 --> 2135.88] and then discover the web [2135.88 --> 2138.28] UI discover the traces [2138.28 --> 2140.08] discover all the things [2140.08 --> 2141.16] which are available and [2141.16 --> 2141.80] which are getting better [2141.80 --> 2143.04] discover the shell I [2143.04 --> 2143.74] haven't even talked about [2143.74 --> 2144.46] the shell that's by the [2144.46 --> 2145.20] way that's like a hidden [2145.20 --> 2147.22] experimental feature that [2147.22 --> 2148.16] is coming in a future [2148.16 --> 2149.70] dagger release that's [2149.70 --> 2150.74] been shipped silently [2150.74 --> 2152.30] for a while but if [2152.30 --> 2153.26] you are interested you [2153.26 --> 2154.00] can join the community [2154.00 --> 2155.00] calls and you can find [2155.00 --> 2156.40] out more but enough [2156.40 --> 2157.58] about dagger we can [2157.58 --> 2158.46] talk about infrastructure [2158.46 --> 2159.78] if you want I want to [2159.78 --> 2160.48] lean into like all the [2160.48 --> 2161.00] other things you're [2161.00 --> 2161.42] talking about right [2161.42 --> 2162.12] because like I have a [2162.12 --> 2163.32] plenty of make files I've [2163.32 --> 2164.28] never loved make I've [2164.28 --> 2164.82] always thought it was [2164.82 --> 2166.12] arcane and hard to learn [2166.12 --> 2167.34] in a kind of a in a [2167.34 --> 2168.98] very gatekeepy way right [2168.98 --> 2169.72] I was like this is like [2169.72 --> 2171.36] someone learned it once [2171.36 --> 2172.58] 18 years ago and they're [2172.58 --> 2173.22] like yeah I wrote that [2173.22 --> 2174.40] make file I have no idea [2174.40 --> 2175.52] what it does and it's [2175.52 --> 2176.44] always been really hard to [2176.44 --> 2177.50] get back into it you're [2177.50 --> 2178.82] like this is all obscure [2178.82 --> 2180.22] old docs that aren't [2180.22 --> 2181.90] relevant anymore and but [2181.90 --> 2182.76] it works right like if [2182.76 --> 2183.46] it's the thing if it [2183.46 --> 2184.90] keeps working cool it's [2184.90 --> 2185.66] probably you a you [2185.66 --> 2186.56] problem if it doesn't [2186.56 --> 2188.00] work it works somewhere [2188.00 --> 2189.72] but you talk about like [2189.72 --> 2190.64] dagger is a drop-in [2190.64 --> 2191.44] replacement for some of [2191.44 --> 2192.24] those things but what [2192.24 --> 2193.86] are the other things like [2193.86 --> 2195.56] on the edges there we're [2195.56 --> 2196.52] like actually what are we [2196.52 --> 2197.86] missing out on by not [2197.86 --> 2199.28] doing this by not having [2199.28 --> 2201.54] a newer tool to be able [2201.54 --> 2202.02] to do that you've [2202.02 --> 2202.50] mentioned open [2202.50 --> 2203.56] telemetry you've [2203.56 --> 2204.26] mentioned the shell you [2204.26 --> 2206.12] mentioned modules like [2206.12 --> 2207.72] those are all pieces but [2207.72 --> 2209.34] it's hard to understand [2209.34 --> 2210.84] like why do I need those [2210.84 --> 2212.00] things or what what what [2212.00 --> 2212.94] can't I do today [2212.94 --> 2216.54] okay so modules as a [2216.54 --> 2218.00] category it's a way to [2218.00 --> 2219.38] package the code that you [2219.38 --> 2220.58] wrote and share it with [2220.58 --> 2222.50] others think about it like [2222.50 --> 2223.62] an atomic pieces of code [2223.62 --> 2224.66] that go well together for [2224.66 --> 2226.66] example the go module [2226.66 --> 2228.50] would encapsulate all the [2228.50 --> 2229.68] code for writing with go [2229.68 --> 2231.22] apps there's something for [2231.22 --> 2232.30] node.js there's something [2232.30 --> 2233.86] for helm there's something [2233.86 --> 2235.94] for even kubernetes if [2235.94 --> 2236.82] you want to figure out [2236.82 --> 2238.36] how to run k3s inside of [2238.36 --> 2239.36] dagger that is possible [2239.36 --> 2241.94] many things like this are [2241.94 --> 2242.94] in the dagger verse [2242.94 --> 2244.42] dagger verse.dev is the [2244.42 --> 2246.36] place to go to check what [2246.36 --> 2247.46] modules are available what [2247.46 --> 2249.00] can I pick and choose so [2249.00 --> 2250.02] those are those are the [2250.02 --> 2251.60] modules the open [2251.60 --> 2254.74] telemetry is how we [2254.74 --> 2256.62] capture what happens [2256.62 --> 2258.12] inside of a dagger call [2258.12 --> 2259.84] when basically dagger runs [2259.84 --> 2262.64] and we are sending all that [2262.64 --> 2264.42] information to dagger cloud [2264.42 --> 2266.64] using these traces so that [2266.64 --> 2268.68] we can visualize what [2268.68 --> 2271.96] happens in your run and [2271.96 --> 2272.98] think of like the network [2272.98 --> 2275.32] graph so in the browser if [2275.32 --> 2275.90] you were to open the [2275.90 --> 2276.84] network graph and you would [2276.84 --> 2278.54] see how long resources take [2278.54 --> 2280.50] to load the spans the [2280.50 --> 2281.92] traces it's exactly the same [2281.92 --> 2285.10] concept so that gives you a [2285.10 --> 2287.42] very deep insight into what [2287.42 --> 2288.48] happens in your automation [2288.48 --> 2290.64] and you can see which are the [2290.64 --> 2291.90] steps which take a long time [2291.90 --> 2294.06] or which are the steps which [2294.06 --> 2295.48] for example don't cash well [2295.48 --> 2297.92] all that information would be [2297.92 --> 2299.02] conveyed in this case in [2299.02 --> 2302.36] dagger cloud so the way to do [2302.36 --> 2303.46] that you just basically connect [2303.46 --> 2305.32] your CLI to your dagger cloud [2305.32 --> 2306.50] account you have to create one [2306.50 --> 2309.14] for it's basically free for [2309.14 --> 2311.50] individuals for teams it is a [2311.50 --> 2313.56] paid plan but as an individual [2313.56 --> 2315.28] you can try to see what this [2315.28 --> 2317.54] looks like and you know you can [2317.54 --> 2319.16] maybe bring your team members [2319.16 --> 2320.42] over your shoulder to look or [2320.42 --> 2321.50] share your screen to see what [2321.50 --> 2322.62] it looks like everyone can do [2322.62 --> 2324.70] this so that gives you an [2324.70 --> 2326.16] insight and appreciation of all [2326.16 --> 2327.16] the things that happen in your [2327.16 --> 2329.18] automation and then the shell [2329.18 --> 2332.10] the third thing is a way to [2332.10 --> 2335.26] put yourself in a context where [2335.26 --> 2337.30] you're trying to discover what [2337.30 --> 2339.78] automation is available and how [2339.78 --> 2341.52] to stitch the different functions [2341.52 --> 2343.42] together it's exactly how it [2343.42 --> 2346.44] use pipes how you basically get [2346.44 --> 2348.70] functionality from different parts [2348.70 --> 2351.00] and try experimenting with it to [2351.00 --> 2353.58] see what makes sense so what is [2353.58 --> 2355.86] the right arrangement for this for [2355.86 --> 2359.88] example pipeline that caches well [2359.88 --> 2362.20] that works well that the expensive [2362.20 --> 2363.80] steps happen first and you can do [2363.80 --> 2365.64] that in the dagger shell so it's a [2365.64 --> 2368.50] way to interactively discover and work [2368.50 --> 2370.72] with your automation and the [2370.72 --> 2373.46] perspective is the functions that are [2373.46 --> 2376.56] declared in dagger so that is the [2376.56 --> 2378.76] starting point i think that's really [2378.76 --> 2380.16] important because we don't have [2380.16 --> 2382.36] enough ops availability and like [2382.36 --> 2385.08] actual like insight into our [2385.08 --> 2386.90] automation like automation is a great [2386.90 --> 2388.70] tool and it makes it where you can [2388.70 --> 2391.46] scale and take a lot of the human [2391.46 --> 2393.32] area out but if you don't have a deep [2393.32 --> 2394.94] understanding of your automation it [2394.94 --> 2397.52] makes it really hard to maintain and to [2397.52 --> 2400.00] scale it and just to use it in general [2400.00 --> 2401.84] when it breaks you're kind of out of [2401.84 --> 2403.74] luck all these things put together [2403.74 --> 2406.62] it's about an experience which is a [2406.62 --> 2408.70] bit more visual it's an experience [2408.70 --> 2409.98] which is a little bit more curated [2409.98 --> 2412.82] right it's almost like what would you [2412.82 --> 2416.40] do if you had to do automation [2416.40 --> 2418.70] properly what would you do if you had [2418.70 --> 2423.82] to not reinvent but i would say rethink [2423.82 --> 2426.06] how make files and how your jenkins [2426.06 --> 2429.92] files and how your scripts should work in [2429.92 --> 2432.42] a container first world and containers [2432.42 --> 2433.46] are important because it's that [2433.46 --> 2436.72] immutable thing right because those [2436.72 --> 2438.58] are like having something immutable [2438.58 --> 2440.22] having something that caches having an [2440.22 --> 2443.08] actual layer is able to speed things up [2443.08 --> 2445.44] in a way that's difficult to do [2445.44 --> 2448.42] otherwise make file is local right how [2448.42 --> 2450.92] do you distribute in make file [2450.92 --> 2453.30] resolution it has a dag but how do you [2453.30 --> 2455.54] distribute the dag and that's something [2455.54 --> 2457.32] which today for example in dagger [2457.32 --> 2459.34] getting very close to that being [2459.34 --> 2459.82] possible [2459.82 --> 2461.26] what do you mean by close like what's [2461.26 --> 2461.72] missing there [2461.72 --> 2465.52] so how do you have a cache like we [2465.52 --> 2467.08] tried a couple of iterations and we [2467.08 --> 2468.76] know how this fails how do you have a [2468.76 --> 2473.26] remote cache that you can safely store [2473.26 --> 2475.82] operations in at scale so imagine every [2475.82 --> 2477.60] single step that runs right it has some [2477.60 --> 2480.46] inputs it does a function and then it [2480.46 --> 2482.86] has some outputs if you're able to cache [2482.86 --> 2485.66] those outputs put them somewhere like a [2485.66 --> 2488.08] cdn or an object storage by the way the [2488.08 --> 2489.62] object storage is what is what we have [2489.62 --> 2492.36] and we had and then when a pipeline [2492.36 --> 2493.92] runs again or the same call runs again [2493.92 --> 2496.12] doesn't matter where you call it from as [2496.12 --> 2498.44] long as the engine is connected to this [2498.44 --> 2501.42] object store it can retrieve the steps [2501.42 --> 2502.94] right it can retrieve the layers doesn't [2502.94 --> 2505.44] have to recompute them sometimes it's a [2505.44 --> 2507.36] lot more efficient to pull down these [2507.36 --> 2508.62] layers rather than recompute the [2508.62 --> 2511.28] operation how do you do that safely at [2511.28 --> 2516.22] scale in a way that is easy to use and [2516.22 --> 2517.92] it's easy to operate that is the hard [2517.92 --> 2519.58] part caches should always be invisible [2519.58 --> 2521.44] right like you just like the easiest [2521.44 --> 2522.74] cache is the one you don't know you're [2522.74 --> 2525.20] using that's a local one locally it [2525.20 --> 2527.04] works well when you go distributed that's [2527.04 --> 2528.20] when problems start that's when you have [2528.20 --> 2529.98] race conditions that's when you have [2529.98 --> 2531.58] pruning for example that's when you have [2531.58 --> 2532.72] all sorts of things that you have to [2532.72 --> 2535.54] deal with when you're dealing with many [2535.54 --> 2537.62] terabytes like hundreds of terabytes of [2537.62 --> 2540.22] this data it becomes a hard problem and [2540.22 --> 2542.88] sometimes depending on network i wouldn't [2542.88 --> 2546.24] say even like a network conditions it [2546.24 --> 2548.28] can be cheaper to recompute it locally [2548.28 --> 2550.84] yeah how much of that can you rely on [2550.84 --> 2552.56] like what docker does for caching right [2552.56 --> 2554.48] because i can have my build x cache [2554.48 --> 2556.36] somewhere which is basically the same [2556.36 --> 2558.28] thing of i get this layer that has a [2558.28 --> 2560.54] shah and i can say oh i'm gonna get [2560.54 --> 2561.84] the same shah just give me the data [2561.84 --> 2564.82] right like how much of that is is using [2564.82 --> 2566.56] what's under the hood by relying on [2566.56 --> 2568.98] containers versus building something that [2568.98 --> 2572.50] do so that makes sense when the inputs [2572.50 --> 2575.02] don't change that frequently in this [2575.02 --> 2577.56] case an input is source code and source [2577.56 --> 2580.46] code churns a lot so how can you still [2580.46 --> 2583.62] have a good cache hit ratio when your [2583.62 --> 2585.96] input is something that changes like on a [2585.96 --> 2588.64] like with every commit so how do you [2588.64 --> 2591.02] sequence right your source code how do [2591.02 --> 2593.02] you compare compartmentalize it so that you [2593.02 --> 2595.74] know which functions depend on which [2595.74 --> 2598.64] source code so that you get like [2598.64 --> 2601.42] invalidations working properly and you [2601.42 --> 2603.14] don't bust the cache too often and that [2603.14 --> 2603.84] is a hard problem [2603.84 --> 2612.78] what's up nerds i'm here with kurt mackie [2612.78 --> 2616.02] co-founder and ceo of fly you know we [2616.02 --> 2617.90] love fly so kurt i want to talk to you [2617.90 --> 2620.56] about the magic of the cloud you have [2620.56 --> 2622.86] thoughts on this right right i think it's [2622.86 --> 2624.46] valuable to understand the magic line of [2624.46 --> 2626.08] cloud because you can build better [2626.08 --> 2628.14] features for users basically if you [2628.14 --> 2629.64] understand that you can do a lot of [2629.64 --> 2631.62] stuff particularly now that people are [2631.62 --> 2633.44] doing llm stuff but you can do a lot of [2633.44 --> 2635.32] stuff if you get that and can be creative [2635.32 --> 2637.56] with it so when you say clouds aren't [2637.56 --> 2639.80] magic because you're building a public [2639.80 --> 2641.94] cloud for developers and you go on to [2641.94 --> 2644.70] explain exactly how it works what does [2644.70 --> 2646.40] that mean to you in some ways it means [2646.40 --> 2648.32] these all came from somewhere like there [2648.32 --> 2650.38] was a simpler time before clouds where [2650.38 --> 2652.22] we'd get a server at rack shack and we'd [2652.22 --> 2656.30] ssh or telnet into it even and put files [2656.30 --> 2658.26] somewhere and run the web servers [2658.26 --> 2660.18] ourselves to serve them up to users [2660.18 --> 2662.38] clouds are not magic on top of that [2662.38 --> 2663.84] they're just more complicated ways of [2663.84 --> 2665.84] doing those same things in a way that [2665.84 --> 2667.44] meets the needs of a lot of people [2667.44 --> 2669.16] instead of just one one of the things i [2669.16 --> 2671.22] think that people miss out on and a lot [2671.22 --> 2674.00] of this is actually because aws and gcp [2674.00 --> 2676.12] have created such big black box [2676.12 --> 2678.02] abstractions like lambda is really [2678.02 --> 2679.92] black boxy you can't like pick apart [2679.92 --> 2681.04] lambda and see how it works from the [2681.04 --> 2682.72] outside you have to sort of just use [2682.72 --> 2684.08] what's there but the reality is like [2684.08 --> 2686.02] lambda is not all that complicated it's [2686.02 --> 2687.90] just a modern way to launch little vms [2687.90 --> 2690.64] and serve some requests from them and [2690.64 --> 2693.02] let them like kind of pause and resume [2693.02 --> 2695.46] and free up like physical compute time [2695.46 --> 2697.10] the interesting thing about understanding [2697.10 --> 2698.78] how clouds work is it lets you build [2698.78 --> 2700.50] kind of features for your users you [2700.50 --> 2702.10] never would expect it and our canonical [2702.10 --> 2703.96] version of this for us is that like [2703.96 --> 2705.38] when we looked at how we wanted to [2705.38 --> 2707.22] isolate user code we decided to just [2707.22 --> 2709.62] expose this machines concept which is a [2709.62 --> 2711.10] much lower level abstraction lambda [2711.10 --> 2712.84] that you could use to build lambda on [2712.84 --> 2714.54] top of and what machines are is just [2714.54 --> 2717.32] these vms that are designed to start [2717.32 --> 2719.14] really fast or designed to stop and [2719.14 --> 2720.80] restart really fast or designed to [2720.80 --> 2722.70] suspend sort of like your laptop does [2722.70 --> 2724.58] when it closes and resume really fast [2724.58 --> 2726.16] when you tell them to and what we [2726.16 --> 2727.70] found is that giving people as [2727.70 --> 2729.50] primitive is actually there's like new [2729.50 --> 2731.06] apps being built that couldn't be built [2731.06 --> 2733.48] before specifically because we went so [2733.48 --> 2736.34] low level and made such a minimal [2736.34 --> 2738.82] abstraction on top of generally like [2738.82 --> 2740.62] linux kernel features a lot of our [2740.62 --> 2742.52] platform is actually just exposing a [2742.52 --> 2744.58] nice ux around linux kernel features [2744.58 --> 2746.22] which i think is is kind of interesting [2746.22 --> 2747.56] but like you still need to understand [2747.56 --> 2749.14] what they're doing to get the most use [2749.14 --> 2751.68] out of them very cool okay so experience [2751.68 --> 2754.82] the magic of fly and get told the [2754.82 --> 2756.96] secrets of fly because that's what they [2756.96 --> 2758.38] want you to do they want to share all [2758.38 --> 2760.22] the secrets behind the magic of the fly [2760.22 --> 2761.94] cloud the cloud for productive [2761.94 --> 2764.22] developers the cloud for developers who [2764.22 --> 2765.78] ship learn more and get started for [2765.78 --> 2770.48] free at fly.io again fly.io [2770.48 --> 2783.20] i was digging around in daggerverse while [2783.20 --> 2784.46] you were you were talking about things [2784.46 --> 2787.40] and i didn't you have a shell sdk is [2787.40 --> 2789.86] that like a thing like is that yeah so [2789.86 --> 2792.04] the dagger shell which i'm talking [2792.04 --> 2794.90] about yes so the shell sdk used to be [2794.90 --> 2797.20] a thing i mean okay so let me just like [2797.20 --> 2799.74] unpack a little bit the the power of [2799.74 --> 2803.34] dagger one of the qualities of dagger [2803.34 --> 2807.24] is that it puts a graphql api it exposes [2807.24 --> 2811.06] a graphql api that all sdks talk to so [2811.06 --> 2812.56] the engine itself which is where the work [2812.56 --> 2814.30] happens and i can think of it like the [2814.30 --> 2816.70] server and the way to interact with it [2816.70 --> 2820.54] is via this graphql api the sdks all they [2820.54 --> 2824.44] do they are graphql clients that expose [2824.44 --> 2826.80] all the operations and all the all the [2826.80 --> 2830.88] resources from the graphql api in a [2830.88 --> 2833.34] language specific way with java by the [2833.34 --> 2835.30] way there's a java sdk with its shell [2835.30 --> 2838.46] in this case so if you're able to model [2838.46 --> 2841.26] the interactions with the graphql api [2841.26 --> 2843.54] through shell functions it would work [2843.54 --> 2846.12] one of the things that that dagger [2846.12 --> 2848.14] shipped a while ago was the ability to [2848.14 --> 2851.08] mix and match the sdks right so i can [2851.08 --> 2854.40] use a module that's written in go but my [2854.40 --> 2856.88] apps in python and i can write my [2856.88 --> 2859.54] function on top of it in python and that [2859.54 --> 2863.04] seems like a an outcome of everything [2863.04 --> 2865.10] talks to the graphql api right like [2865.10 --> 2866.72] everything's just like okay it's like we [2866.72 --> 2868.24] all just talked to an api i don't care how [2868.24 --> 2869.94] you got here if you're doing curl commands [2869.94 --> 2872.80] you can get to this graphql do something [2872.80 --> 2875.70] and then the next step is on you to write [2875.70 --> 2877.62] that in whatever language you want that's [2877.62 --> 2878.98] also awesome because it makes it [2878.98 --> 2881.38] accessible for multiple teams working at [2881.38 --> 2883.96] the same startup or enterprise that are [2883.96 --> 2886.44] writing in multiple languages because it [2886.44 --> 2888.10] gets to the point where enterprises want [2888.10 --> 2890.82] to just start making everybody use two [2890.82 --> 2892.40] languages and you're like dude this [2892.40 --> 2894.50] language does not work for the things [2894.50 --> 2896.40] that you want it to do like i mean it [2896.40 --> 2898.18] can but it's not the right you know like it [2898.18 --> 2899.12] just gets to the point where you're [2899.12 --> 2901.42] using a hammer for every project and [2901.42 --> 2902.46] when they want they want to optimize [2902.46 --> 2904.16] shareability right and they're just like [2904.16 --> 2907.10] hey by by making everyone write kotlin [2907.10 --> 2909.32] then everything's gonna go smoother [2909.32 --> 2910.84] don't lie it's usually a typescript we [2910.84 --> 2912.96] all know like it is typescript yeah but [2912.96 --> 2914.32] it's it's you know it goes to the two [2914.32 --> 2915.58] languages right like everything front [2915.58 --> 2917.08] and you know that and maybe some back [2917.08 --> 2918.18] in scotland or something like that it's [2918.18 --> 2919.54] just like okay everyone writes these [2919.54 --> 2920.72] languages now because we have to be [2920.72 --> 2921.94] able to share this stuff and we have [2921.94 --> 2923.90] this devops team over in the corner and [2923.90 --> 2925.52] they're out there writing yaml and [2925.52 --> 2926.50] they're like well i'm just gonna plug in [2926.50 --> 2927.72] this yaml thing it's like the team's [2927.72 --> 2930.16] like no i don't want that but the if if [2930.16 --> 2932.44] the the kotlin team is writing a dagger [2932.44 --> 2934.80] module and the the typescript team wants [2934.80 --> 2937.86] to also use that module they can because [2937.86 --> 2940.14] you just said our lowest common denominator [2940.14 --> 2943.08] is the api if you can call the api i [2943.08 --> 2944.52] really don't care what language you call [2944.52 --> 2946.62] it in we're going to give you the same [2946.62 --> 2949.20] functions and you can every language has [2949.20 --> 2952.26] a way to call a web sockets or or some [2952.26 --> 2954.00] way to like call an api somewhere that's [2954.00 --> 2955.98] external and say like here's data give me [2955.98 --> 2958.58] back something you're enabling the dev team [2958.58 --> 2960.70] to do it in the language they're more [2960.70 --> 2962.70] comfortable with right and you're allowing [2962.70 --> 2966.24] them to fit it into whatever they've already [2966.24 --> 2968.54] built which means that you're not trying to [2968.54 --> 2971.42] completely um this is what i'm looking for [2971.42 --> 2973.24] like there's so many times where they're just [2973.24 --> 2975.70] like go restructure all this code to fit this [2975.70 --> 2978.20] one api and then you break six things it [2978.20 --> 2980.74] doesn't work right like you know like it any [2980.74 --> 2982.78] way that you can make your life easier without [2982.78 --> 2986.38] having to do like a huge restructure is enabling [2986.38 --> 2988.28] enforcing are kind of the same thing right [2988.28 --> 2990.14] because not all dev teams want this they're [2990.14 --> 2991.70] like actually i just liked it when the external [2991.70 --> 2993.52] team was responsible for the thing and when [2993.52 --> 2995.22] it broke i just sent them an email and i went [2995.22 --> 2997.56] to lunch right like that was that was how a [2997.56 --> 2999.92] lot of devs liked it and it's true but in this [2999.92 --> 3002.16] time when we're having we have less resources [3002.16 --> 3003.58] and we're trying to do less with more [3003.58 --> 3006.52] i don't know if people are going to always have a [3006.52 --> 3008.04] whole nother team you know what i mean like [3008.04 --> 3009.94] think about the restructuring of enterprises [3009.94 --> 3011.98] and just the fact of what you have to work lean [3011.98 --> 3015.12] with a startup you may not have the option to [3015.12 --> 3016.78] have that whole other team like there's a reason [3016.78 --> 3021.50] why cloud services and managed products and sas [3021.50 --> 3023.72] products people pay so much money for it because [3023.72 --> 3026.68] it got rid of your dbas it got rid of parts of [3026.68 --> 3028.62] your ops teams and you could have a smaller ops [3028.62 --> 3030.78] team because at the end of the day people want [3030.78 --> 3033.00] us to do all the things with the least amount of [3033.00 --> 3035.92] resources and you know i mean and i would also [3035.92 --> 3037.92] say like if you went to rds and got rid of your [3037.92 --> 3040.48] dbas you made a mistake right like that's not [3040.48 --> 3043.22] it's not necessarily the thing you thought it was [3043.22 --> 3045.58] going to be i also think that these things evolve [3045.58 --> 3048.48] right so like instead of needing a dba now a lot [3048.48 --> 3052.28] of people need data architects right so like maybe [3052.28 --> 3054.32] you don't need somebody running around a data center [3054.32 --> 3057.78] like in doing the dba in the traditional sense but [3057.78 --> 3060.62] just because you got rid of a dba you need a data [3060.62 --> 3062.94] architect to now tell you how to like do your access [3062.94 --> 3065.52] patterns and how to optimize your database but [3065.52 --> 3068.44] one is easier to contract and one you need every day [3068.44 --> 3071.10] like so it's just it's they're all trade-offs [3071.10 --> 3074.88] right like they're it's all like what works for your [3074.88 --> 3076.96] business and what works for your use case [3076.96 --> 3079.88] but i think the problem that we have is that people [3079.88 --> 3082.76] don't know enough about the differences and we just [3082.76 --> 3085.04] tell them like it's this new shiny thing so they don't [3085.04 --> 3087.36] they're not enabled with the right information to make [3087.36 --> 3090.58] those decisions they both have value but what has the [3090.58 --> 3092.14] most value for what you're trying to do [3092.14 --> 3095.72] you know a lot of that also to me at least in my experience has been all [3095.72 --> 3102.08] people that were going after promotions and you see bigger you know you like hey [3102.08 --> 3104.58] guess what i'm going to rewrite all of our make files and dagger [3104.58 --> 3108.92] and it's going to have impact on the business and someone else over here like but why [3108.92 --> 3111.32] didn't you learn why we had the make files in the first place like why [3111.32 --> 3114.20] didn't you learn he's like that's hard right like that that side of the [3114.20 --> 3117.60] business is a lot harder and also never get you a promotion i guess we're like i [3117.60 --> 3122.50] understand make now has never been on the promo doc for like any engineer yeah [3122.50 --> 3128.24] i would not recommend that by the way if you if you have a make file what i would [3128.24 --> 3133.18] say don't rewrite it in dagger try running it in dagger first that would be a much [3133.18 --> 3139.86] smarter first step and always focus on documentation first ship it episode 44 [3139.86 --> 3145.14] can you say that louder like because i think people think that automation means we [3145.14 --> 3149.48] no longer have to document things and it hurts my entire soul so badly [3149.48 --> 3156.18] automation is not documentation i'm going to emphasize this in a couple of ways [3156.18 --> 3161.52] first of all uh ship it episode 44 was a very important moment in my career [3161.52 --> 3168.06] when um we sat down with kelsey and we went through all the things that are [3168.06 --> 3170.76] important for understanding how complex systems work [3170.76 --> 3176.46] and documentation is the first step that you should do before you touch before [3176.46 --> 3181.62] you even think about automation because when you document you realize about all [3181.62 --> 3187.50] the inefficiencies and you realize that if you were to change your automation at [3187.50 --> 3193.06] any point as back to your um what you're mentioning justin make file rewriting that [3193.06 --> 3197.96] in dagger what you really want is the blueprint for the make file and the [3197.96 --> 3203.62] blueprint is the document that you don't have i was one of those people i remember [3203.62 --> 3209.48] daniel fedotov i will not forget him we were on the rabbit mq together and he was [3209.48 --> 3213.52] asking me for the documentation i said hey daniel you don't need documentation i [3213.52 --> 3218.48] wrote this beautiful make file it does everything for you it's self-documenting it [3218.48 --> 3222.94] tells you what the targets are what do you mean where's the documentation this is the [3222.94 --> 3229.42] documentation and um it took me many years to understand how wrong i was in that [3229.42 --> 3236.50] moment and i did it right so when i joined dagger one of the first things which i did [3236.50 --> 3243.12] i made sure that the releasing process is documented so how we released dagger we [3243.12 --> 3249.14] started with a document that document has been updated almost every other week for the [3249.14 --> 3255.68] last three years and that is the blueprint which we use for all the automation at dagger [3255.68 --> 3262.00] that makes my heart so happy releasing md so i've learned my lesson and i hope that you will too [3262.00 --> 3269.80] dear listener write the documentation first keep it up to date it keep refining it keep working on it [3269.80 --> 3276.74] keep sharing it's not done it's never done and then add your automation did you guys hear him like [3276.74 --> 3284.70] just replay i just yes because like it blows my mind like and then people like so you know how you [3284.70 --> 3290.30] were saying like well people want to do that for promotion where they rewrite everything or people [3290.30 --> 3294.60] are like we've been doing it in bash the same way for 20 years and we're never going to change [3294.60 --> 3300.50] anything and there's so many new ways to do it but let's just do it this way forever and you're like [3300.50 --> 3307.22] not that like i do think sometimes if it's really simple bash sometimes is just the way to go right [3307.22 --> 3313.08] but like never thinking about how you can onboard new people and share knowledge and make it easier [3313.08 --> 3320.02] for everybody like you're just doing the same thing forever and the comments aren't docs too right like [3320.02 --> 3322.90] that's the thing that a lot of developers like well it has a bunch of comments we know no no no [3322.90 --> 3328.46] you pull those out somewhere else you make it searchable for someone else that isn't in the code to be [3328.46 --> 3333.18] able to find i've had people tell me not to put comments though because it the script is self [3333.18 --> 3338.90] explanatory no they'll be like it the automation is explanatory if you write it clean enough and i'm [3338.90 --> 3345.84] like no no it's not years ago yeah yeah i know i know now i'm just like when you're new and you've [3345.84 --> 3352.34] never seen this before it's not self-explanatory like every job i've worked at the docs i've written [3352.34 --> 3357.32] have lasted longer than the code i've written and like if you want to like do the lasting impact like [3357.32 --> 3361.84] the docs are the thing that's but we don't incentivize docs we incentivize yeah that's what [3361.84 --> 3366.88] i'm saying so like like and it's just it doesn't make sense because like what is the number one rule [3366.88 --> 3372.22] in school or wherever when they when you learn to code what do they always say write it down plan it [3372.22 --> 3376.36] out and then you start coding but i don't know where we missed that with automation we were just [3376.36 --> 3381.08] like oh but like that's only for code not for scripting or anything else that's important or release [3381.08 --> 3386.64] processes when dude release processes and scripting and like doing the infrastructure usually takes [3386.64 --> 3390.84] longer than writing the code that you want to release so like why would you not do your due [3390.84 --> 3395.70] diligence to make sure well i mean let's be honest the meetings about writing the code take oh god [3395.70 --> 3400.66] yes like you know but that's what just like kills me when they're just like engineers just need to be [3400.66 --> 3405.18] able to write really good code and be technical and i'm like that is such a small part of being a good [3405.18 --> 3410.62] engineer a lot of people that are afraid of ai being able to write code is like you're just looking at [3410.62 --> 3416.38] just a small small part of what the engineers are doing those are the parts that we incentivize and [3416.38 --> 3423.32] reward to be this 10x developer who can write and build things super fast but like who else who [3423.32 --> 3427.58] like but think about those are the people that job hop every two years and somebody else has to go fix [3427.58 --> 3434.02] the like 10x developer stuff that you put together and duct taped it nobody knows what it does nobody's [3434.02 --> 3439.04] documented and you're ruining someone else's on-call life adam you could just drop names if you want this is [3439.04 --> 3446.54] i was a little triggered for a minute there like i'm so tired of tech bros [3446.54 --> 3455.46] so where do you think this goes gearhard like what what is the end goal here for something like [3455.46 --> 3462.40] dagger something that is is making this a i i feel like dagger is the is the productized make right [3462.40 --> 3468.58] if i if i erased everything and said what if this was a really good product and we put it in here and [3468.58 --> 3474.54] we gave them good experience and good tools and good ui around this thing that is kind of opaque and [3474.54 --> 3479.76] always been weird that one or two people on the team understand even remotely like let's just make it so [3479.76 --> 3484.02] it's easier for everyone to understand let's make it so it's a thing that's maintainable is shareable [3484.02 --> 3489.50] is scalable what's the end goal there what's the thing that you're at the end of it like oh if we get [3489.50 --> 3495.14] here we've made it you just made me think of like the difference between vms and the cloud dagger is [3495.14 --> 3506.38] like the make file for like well remember what containers did for applications that is the moment [3506.38 --> 3513.44] that i envisage for all the scripts that we write that moment for all the automation that you write [3513.44 --> 3521.66] the container moment for all those things where we agree to put this these things in containers first [3521.66 --> 3529.30] in a way that's immutable in a way that's content addressable and if we do that if this grows to a [3529.30 --> 3537.00] certain point we are no longer writing automation the automation knows how to consume the resources [3537.00 --> 3543.04] that others have built without us having to go and figure out how to plug them together for example [3543.04 --> 3551.76] how many times have you gone to download the package checked or like a tarball unzipped it check the [3551.76 --> 3560.46] sha 256 make sure it's okay make sure read at the change log read what changed figure out how like what [3560.46 --> 3567.16] things have deprecated all of that times a thousand in your career you will have done this a thousand [3567.16 --> 3576.20] times at least you're giving me ptsd of like flashbacks like is there a better way that i can [3576.20 --> 3581.42] consume these things i can run it i can validate that my software works whatever i'm trying to do [3581.42 --> 3589.50] it combines well with all these things in a way that is is just friendlier you have get like a whole [3589.50 --> 3595.94] new experience of consuming software of building of distributing of doing anything that anything that [3595.94 --> 3602.04] you have to do with your source code before it gets out in front of people encapsulating all that [3602.04 --> 3606.94] knowledge encapsulating all that knowledge in a way that is also documented right because coming back to [3606.94 --> 3613.04] documentation you don't document things but maybe you're willing to document your argument maybe you're [3613.04 --> 3617.36] willing to document your function and explain how this function should be used and how it can be [3617.36 --> 3623.16] interacted and your function is a very small piece the world now gets to use because you're [3623.16 --> 3628.58] passionate about java and you know how to build java apps and to package containers well can you just [3628.58 --> 3634.48] go on a speaking tour about documentation and automation because like i feel like you could fix the world [3634.48 --> 3641.32] one doc at a time one doc at a time yes yes do you feel like there's any mistakes we're going to make [3641.32 --> 3646.30] again i feel like there's uh as much as containers have done well there's a lot of things that [3646.30 --> 3651.20] containers haven't done well for application packaging and sharing do you feel like there's [3651.20 --> 3656.78] something in there that's like this maybe this is in general like not not dagger specifically but maybe [3656.78 --> 3662.02] this like process of like this writing code doing a bunch of stuff to it and then like spitting it out [3662.02 --> 3667.18] packaging it sending it somewhere else maybe that's the problem i think the problem is that the ambition [3667.18 --> 3674.20] is too big and we don't get to capture it in practice well enough so if we disconnect it too much [3674.20 --> 3679.78] from reality if we talk about these hypotheticals without you know having something real to back [3679.78 --> 3687.60] them up everything will just not go where it needs to go because there's too much air in the balloon and it will [3687.60 --> 3693.72] pop and the balloon is just just air that's it that is my fear i think with every technology [3693.72 --> 3701.76] you tend to get that also you start getting um competition you know players that maybe they [3701.76 --> 3707.62] dominate the market but maybe they're not the right solution for the problem but they're the [3707.62 --> 3714.96] they're the you know the big gorilla the 800 pound gorilla in the ring and everyone gravitates towards [3714.96 --> 3719.64] them because they have a monopoly on the market and everyone talks about that and then you get like [3719.64 --> 3728.60] some huge acquisitions and everything goes sideways so i think i think there is a real risk of not [3728.60 --> 3738.54] keeping it rooted in reality going too far out in hypotheticals and maybe not dealing enough with [3738.54 --> 3744.30] like the little paper cuts because there's a lot of paper cuts and we try addressing them so many paper cuts [3744.30 --> 3750.02] yeah and now i'm talking about dagger specifically but you're right even in general like in the in the [3750.02 --> 3755.50] software industry there's like so much stuff which is broken and i think you need some of that because [3755.50 --> 3759.60] it is a form of technical debt right you need to innovate at the same time but if it's too broken [3759.60 --> 3766.84] that's a problem so one of the challenges that we have or that something which i've taken upon myself [3766.84 --> 3772.66] and if you want to read more about it you can go to dagger issue 8184 and guess how i know it off the top [3772.66 --> 3781.08] of my head because i've been on it for months now is how dagger uses dagger at scale what does that mean [3781.08 --> 3791.54] the scale that i'm talking about when our pipeline runs we are spinning up maybe up to 10 15 large [3791.54 --> 3800.76] instances on ec2 that sum up about 500 cpus to run the pipelines all the pipelines for dagger if you have [3800.76 --> 3806.08] five or six pull requests happening at the same time you have thousands of cpus being spun up [3806.08 --> 3813.86] to test various things dagger that costs real money that is a hard problem the less reliable your [3813.86 --> 3820.14] pipeline is and the longer it takes to run the higher the pain the higher the waste and that is a [3820.14 --> 3826.16] hard problem every single team will have this at some point if they're successful we are getting there [3826.16 --> 3832.92] we are looking at our aws bill and we're thinking wow this is expensive really really expensive what [3832.92 --> 3839.46] can we do and there are certain ways in which we use dagger which we need to bust the cache we need to [3839.46 --> 3847.46] do certain things at which point are containers the right way of doing this that comes up what are what [3847.46 --> 3854.28] is the overhead of using overlay fs what is the overhead on the disks what is the overhead on the [3854.28 --> 3860.04] networks of pulling all these bits down what is the overhead of having to recompute the same thing [3860.04 --> 3865.66] when you don't have a distributed cache so there are some hard problems there they're fun to solve [3865.66 --> 3873.26] and this can either be a great success or the biggest lesson of my life and i will accept them both [3873.26 --> 3880.76] with equal joy to my heart i think you make a point there too like i i did a calculation for my [3880.76 --> 3886.80] local developer desktop which is like a amd thread ripper versus something comparable at amazon [3886.80 --> 3894.36] and and when i look at the the machine that i built out of parts basically uh cost around 600 [3894.36 --> 3900.16] something dollars and like to get something to aws it was 101 times more expensive like not not 10 times [3900.16 --> 3904.60] like like for a this is consumer grade not ecc ram it's not all that stuff like i don't need that [3904.60 --> 3908.38] for some of these things right like i'm doing local like development and testing like i don't need [3908.38 --> 3912.08] ecc i don't need all this stuff and when you have something as portable as dagger it makes a lot more [3912.08 --> 3917.44] sense to like hey maybe actually just a couple of these pcs that don't have all of the benefits that [3917.44 --> 3922.30] we don't actually need could save us a lot of money and even if they're not used because if people [3922.30 --> 3925.38] say like oh it's wasted right you're not using it all the time like doesn't actually matter [3925.38 --> 3930.50] when it's a hundred times cheaper i could use it one one hundredth of the time and still save money [3930.50 --> 3936.86] out of it right like that that math works out so just be like oh maybe a pile of build machines just [3936.86 --> 3941.68] sometimes does make sense i think that's fascinating how sometimes it does make sense but sometimes you [3941.68 --> 3946.60] also have to remember that if you're using it and you're not just playing around with it and you need [3946.60 --> 3951.62] it to be reliable you have to replace it you have to maintain it and you have to have people that have [3951.62 --> 3956.48] the knowledge to do it so i mean people build you know see like the i have terraform code and all [3956.48 --> 3960.46] other stuff in the cloud too like all that stuff needs to be maintained and done and so it's not [3960.46 --> 3966.48] free either way like they're not free either way but like i think that we went so hard on the cloud [3966.48 --> 3972.40] we went so all in and it was the answer for everything but now our attitude is automatically that [3972.40 --> 3977.06] we don't need the cloud and everything can be done on-prem and neither of those are correct like [3977.06 --> 3981.82] there's there is a use case and there is a time for both and you have to figure out what works best [3981.82 --> 3986.06] and what is most cost efficient and sometimes one works for a while and then you grow too fast and [3986.06 --> 3992.18] you need to do the other but i do think that it's a disservice to pretend like one size fits all and [3992.18 --> 3997.26] we can do everything on-prem we can't like no yeah absolutely if that was the case we'd still be [3997.26 --> 4003.92] running servers in our grandma's basement you know i mean i still am it's your basement now but but you [4003.92 --> 4009.30] know what i mean what you use for fun and what you use to maintain i don't know a huge enterprise [4009.30 --> 4014.48] are two completely different things and like it's funny because like there's so many companies talking [4014.48 --> 4019.48] about how they run on-prem right now but they're essentially building their own cloud that someone [4019.48 --> 4024.08] else is maintaining it's a little bit cheaper but they're not really truly running on-prem like [4024.08 --> 4030.32] they're sending their hardware to someone else to run so it's like another iteration of the cloud [4030.32 --> 4035.90] almost you know what i mean it's not all aws in all computers in my garage there's like this mix [4035.90 --> 4039.94] of like oh there's colos i can rent some things temporarily i can get bare metal places like [4039.94 --> 4045.56] that's what i'm saying so it's still a it's almost an iteration of like the cloud and hardware and on-prem [4045.56 --> 4051.26] and we're in this weird place that i think because everybody is trying to save money and figure out how [4051.26 --> 4056.52] we do things differently that like people are almost scrambling and it's going to be interesting to see [4056.52 --> 4061.90] what actually saves people money and time and is good for their use case or if they end up down a [4061.90 --> 4067.72] whole nother rabbit hole but i think we just like it's a disservice to pretend like they are all the [4067.72 --> 4071.86] same thing and that you are not essentially using a different type of a cloud you're just using a [4071.86 --> 4077.62] private cloud if we remove the word cloud like in gear it's like it's just vms that's what i'm saying [4077.62 --> 4083.32] they're all vms and so my point was the fact that dagger is a new tool because the downside of having [4083.32 --> 4089.14] on-prem build machines which i've had in plenty of jobs is always the maintenance cost of like oh [4089.14 --> 4093.58] someone tweaked something and now i don't know what broke right but the the dagger encapsulation [4093.58 --> 4099.68] of these jobs of saying hey actually this should run anywhere as long as this engine exists and and [4099.68 --> 4105.72] now we can encapsulate that holistically and not rely on the base os as much or the version of it and [4105.72 --> 4109.72] you even we started the conversation you were saying how like the kernel matters in a lot of these [4109.72 --> 4115.26] cases too and so there is some things that aren't generically possible across the board there is [4115.26 --> 4121.42] still some maintenance cost there but for the most part like dagger should run more portably than my [4121.42 --> 4125.74] make file that is doing bash scripts right like that side of it is like oh now i have that flexibility [4125.74 --> 4131.84] to make the decision of where do i want to run this how should i run this where is cheap compute [4131.84 --> 4137.26] available right like i could do spot instances to lower that price or if i have some extra computers in [4137.26 --> 4141.40] the closet i could just plug them in and do that too like there are there are options there because [4141.40 --> 4147.88] of the portability that you're building into the tool and i don't lose out on the benefits that i [4147.88 --> 4152.92] might get for something that's like hey i'm doing a vm in aws i need logs from it oh cool those are just [4152.92 --> 4157.24] in cloud watch logs and in this case like no we're all talking to an api where you get the tracing [4157.24 --> 4162.98] by default uh by using this dagger cloud or by using you get the logs you get the stuff out of it like [4162.98 --> 4167.06] that's why i think there's so much gravity around get have actions because get have actions is just [4167.06 --> 4170.92] like we give you a bunch of these defaults but it's not portable it's not something that is easy [4170.92 --> 4174.84] to move around and decide i don't want to run this here anymore i think that's going to be the [4174.84 --> 4180.34] startups that really like that become the next big tech companies that are going to really benefit from [4180.34 --> 4185.38] this weird sunken place of tech that we're in where everyone's trying to like make as much money with [4185.38 --> 4191.32] the least amount of what they have is like the startups that allow you to lift and shift and to be [4191.32 --> 4198.32] portable and to be is kind of like into monitor and uh like to monitor all your different clouds [4198.32 --> 4202.86] all your different instances and all what you're running like whatever gives you that portability [4202.86 --> 4207.94] and that observability and monitoring into all the different things because everybody wants to do hybrid [4207.94 --> 4213.48] cloud all the different regulations coming or the fact that now one cloud or one way to do it or [4213.48 --> 4216.98] on-prem is cheaper and they want to lift off of this thing and go to this thing like all the things [4216.98 --> 4222.44] that are going to make software more portable and easier to switch are going to be what really [4222.44 --> 4229.40] expands and like like just hits right now and that's where i think dagger fits perfectly of we could [4229.40 --> 4234.64] say we could run this anywhere the portability of i can i can't write code anywhere or like i need a [4234.64 --> 4240.40] developer machine like i need something that has some tooling locally in in all the web-based [4240.40 --> 4245.52] development tools and whatnot like i've tried all of them i've never stuck with any of them because [4245.52 --> 4249.84] they always had some limitation for me or they ended up being really expensive and i'm like actually [4249.84 --> 4253.94] i already paid for this computer i could use the computer and pay myself the maintenance cost of [4253.94 --> 4258.12] making sure this thing runs and on the other end is like well we have these containers we can package [4258.12 --> 4261.22] them we have this cooper and anything which can run everywhere we're like we have all these things [4261.22 --> 4265.28] that like at the other end we can have the portability but in the middle there between writing [4265.28 --> 4271.24] the code and deploying the code somewhere in that build pipeline the cicd stuff was always super sticky [4271.24 --> 4276.00] and it was super hard to say like i can shift this somewhere i can't lift and shift jenkins i can [4276.00 --> 4280.18] but like it's a lot of work it's not actually just an endpoint like and and you have to just reproduce [4280.18 --> 4285.20] the same thing in a different environments and i think that's really cool about how like you mentioned [4285.20 --> 4289.88] at the beginning right like this is what what containers did for applications dagger could do for that cicd [4289.88 --> 4295.58] middle pipeline of this is portable and we have some options here also when you're scaling like [4295.58 --> 4301.20] release infrastructure what used to work can definitely get to the point where you have too [4301.20 --> 4305.88] many pipelines and you've grown too much and then having to lift and shift that is the most painful [4305.88 --> 4311.88] in the whole world i mean and and that's that's a unique problem to a lot of like not all companies [4311.88 --> 4316.20] have that problem like amazon had that problem that was a that was definitely an amazon there's a lot [4316.20 --> 4321.40] of pipelines here um to make things safe and roll out and all that stuff but not everyone requires [4321.40 --> 4326.76] that amount but also when you're building stuff for different architectures too being able to lift [4326.76 --> 4331.76] and shift them to different things because what might work on mac might not work on windows and what [4331.76 --> 4337.64] you know what i mean so just being able to see if this really does work in that architecture natively [4337.64 --> 4343.66] is really important yeah one thing that we talked about previously i thought that was a really important [4343.66 --> 4350.16] point was around what makes a good developer what would say go further what makes a good [4350.16 --> 4355.92] engineer a good software engineer and it is not the documentation even though that plays a very [4355.92 --> 4361.92] important role it is not the code that they write even though that has to be present it is not how they [4361.92 --> 4367.12] present and how they talk but again that has to be present how they share ideas you know how they [4367.12 --> 4374.14] show into the world and do whatever they have to do those are well-rounded individuals and that [4374.14 --> 4381.28] well-roundedness i can see how it translates to whether it's the cloud whether it's the on-prem whatever [4381.28 --> 4387.00] the next thing is you need to be well-rounded in all these things and you should be using all of them [4387.00 --> 4394.78] because if you don't use it you lose it right very simple very true it will never change so you should be [4394.78 --> 4400.78] using the cloud but you should be using the cloud the way the cloud was meant to be used and you justin [4400.78 --> 4406.72] made a very good point at disney how you were using the cloud possibly the best example of using the [4406.72 --> 4415.94] cloud correctly right span out stand up all these instances and then tear them down that's how it's [4415.94 --> 4420.56] supposed to be used right it's capacity on demand massive capacity you would not want to buy that [4420.56 --> 4426.36] but when you need it it's there and it's a great commodity it is a hundred times more expensive [4426.36 --> 4431.74] because that's what this commodity costs you where else can you turn up and say hey spin up a hundred [4431.74 --> 4436.90] beef instances that wouldn't take months would take maybe years to source all of those things [4436.90 --> 4442.68] and by the way no no i don't want that amd cpu i want that intel cpu oh arm i think i'm going to [4442.68 --> 4448.74] have some arm and by the way i'm going to have i know 10 000 arm cpus where can you get that is the [4448.74 --> 4454.78] convenience of that now if you know what you need right as a business that is successful as a [4454.78 --> 4459.82] startup that's successful well for the things that you know that you definitely need and that's like [4459.82 --> 4466.48] your baseline operating budget in terms of infrastructure have that in a place that makes [4466.48 --> 4473.26] sense have that cheap have that you know cheap i say cost efficient do what makes sense hybrid cloud [4473.26 --> 4480.40] what about wasm what about the wasi runtimes how do they change how we see containers and how we work [4480.40 --> 4487.62] with containers is that coming well i can tell you that react did not work for dagger cloud [4487.62 --> 4492.58] react the dom to generate all the things that we had to generate was just breaking down [4492.58 --> 4500.02] the tech was not up to scratch so what did we do we looked at wasm v3 that's the v3 cloud we went [4500.02 --> 4506.34] through three rewrites to get it to a point where it's as performant as it should be and there's so [4506.34 --> 4511.06] many dragons there like don't think that this is like the holy place where you pick it up and [4511.06 --> 4516.72] everything is awesome oh no the cycle starts again but that's it that's the beauty of it start those [4516.72 --> 4522.98] cycles keep going through them keep learning keep iterating eventually you will be able to consider [4522.98 --> 4528.54] yourself well-rounded and others will look at you and say wow i wish he responded to my pull requests [4528.54 --> 4534.72] i wish he left some comments and did some reviews because i like his reviews i wish he blogged more [4534.72 --> 4542.46] or i wish she went and gave some talks because she's an excellent presenter so look for well-rounded [4542.46 --> 4548.84] not for 10x i think that's a great place to end the uh end the episode and uh gerard again thank you [4548.84 --> 4554.04] so much for starting this podcast for for putting it out there as as a place that people can learn [4554.04 --> 4559.22] and just get access to someone that they may not be able to approach and have a conversation with [4559.22 --> 4564.34] and then learn from their insights on all of the cycles and failures that they've had from the past and [4564.34 --> 4568.16] the things they've learned and then the things they want to share about that um that's what we [4568.16 --> 4572.70] are trying to continue that's the things that we're like not just having cool people on with [4572.70 --> 4576.70] exciting conversations which we do like but really just being able to help people be a little more [4576.70 --> 4581.70] well-rounded about actually that database is important and actually the cicd is important and [4581.70 --> 4588.30] all of the things that go around the code and uh the businesses responsible for matters and how do we [4588.30 --> 4593.78] help them be more responsible and help the people that run them understand them so thank you so much [4593.78 --> 4598.54] and i truly think people don't talk about that part enough like that's the information that people [4598.54 --> 4603.50] need because we don't reward it right but like it's so true though like you can't have any of [4603.50 --> 4607.78] these cool things without the database or without the infrastructure without all of these things so i [4607.78 --> 4613.32] just think that like it brings me like nerdy joy to expose people to things that they may not [4613.32 --> 4619.26] have been exposed to and to have people who do that every day tell them like the caveats or what they [4619.26 --> 4623.62] tried and what they hated and where they failed and what they succeeded at you know where should [4623.62 --> 4629.14] people find you online if they want to reach out continue the conversation gerhardt.io that's a good [4629.14 --> 4639.06] place there's this new space which now it has a domain make it work.fm and make it work.tv [4639.06 --> 4647.30] the way i think of it is movies for nerds. come to blue sky so we can like bug you. i am i have to [4647.30 --> 4653.58] start using it i am there but i have to start using it because i there's too many social places [4653.58 --> 4659.10] and there's so many things to do that's i think the one area in which i'm not as well rounded as i [4659.10 --> 4664.84] could be be more present on the social media. i think blue sky is becoming where everybody seems [4664.84 --> 4669.24] to be going i'm really hoping that it sticks because i can't manage all these social media [4669.24 --> 4674.26] platforms. yeah i mean there's a lot i i'm hoping blue sky succeeds only for the fact that [4674.26 --> 4679.78] it will encourage people to hopefully make their own websites like i think that's the true mission [4679.78 --> 4684.56] and purpose of blue sky is to make the internet be more democratized and people just understand [4684.56 --> 4690.60] i should be able to own and run this data wherever i want and then publish my own content and own a [4690.60 --> 4696.10] domain like that stuff is super important to me and and it's not just another centralization of [4696.10 --> 4700.88] hey we're the new search engine we're the new social media thing it's like everyone come here [4700.88 --> 4706.30] it's like no actually the how everyone interacts together and builds on top of what other people [4706.30 --> 4711.72] are doing in a democratized way i think is very critical for the internet to succeed long term. [4712.56 --> 4717.94] i like that it has the ease of other social medias but it does like you said force you to like [4717.94 --> 4723.54] but not just force you but you get to keep what you're investing in you know the your data your [4723.54 --> 4728.22] content. you can't see it i'm running a pds on my raspberry pi back here behind me and so i have a [4728.22 --> 4732.28] user that's on blue sky that's literally reporting back here i already made a video about it it'll be [4732.28 --> 4738.62] posted soon i'm waiting for blue sky to cool down. come get your nephews and teach them how to run [4738.62 --> 4743.72] stuff on raspberry pi because i've there's too many things to do and i don't have time to figure this [4743.72 --> 4747.20] out and they keep bugging me come get them. we will have someone from blue sky infrastructure [4747.20 --> 4752.18] on the podcast soon they're dealing with all sorts of scaling million people every day for the past [4752.18 --> 4756.90] few days so i'm i'm waiting to post my video because i was talking to them i'm like hey would [4756.90 --> 4761.36] this be a problem for you if everyone started doing this like yeah actually please wait just wait a week [4761.36 --> 4765.48] and we'll be fine but yeah we're gonna have them on the show soon because i'm fascinated to learn [4765.48 --> 4769.96] about their scaling journey for the last couple weeks and months. they've done a great job and also [4769.96 --> 4775.52] they've done a great job on print. yeah question for you justin would it work in kubernetes pds? [4776.24 --> 4780.82] yeah i mean it's just it's a it's a web server with a sqlite database. amazing. that's all it is. [4781.06 --> 4784.92] amazing. it's just you know it's like a git tree in sqlite and that's all it is and it's just like [4784.92 --> 4789.78] yeah if you can publish that somewhere it's that's that's one of the things i think is really cool about [4789.78 --> 4795.52] blue sky's self-owned sort of federation versus mastodon where the mastodon scale-up story was a lot [4795.52 --> 4799.12] more difficult where like you need postgres you have a ruby on rails app you need cash. [4799.12 --> 4804.78] i just can't commit to that type of like it's not that i hate mastodon i just can't commit to that [4804.78 --> 4810.10] type of overhead for social media like i have so many things to do. and the pds is just a it's [4810.10 --> 4813.56] it actually they deploy it in a container it's like if you have a debian thing it's already packaged in [4813.56 --> 4819.14] a container you just want some some way to have some uh you know reliable storage underneath so [4819.14 --> 4823.80] that your sqlite database is always available and backed up but otherwise yeah it's fascinating i've [4823.80 --> 4828.22] already run a few of them um i did a live stream a little while ago but yeah i'm still toying with it [4828.22 --> 4831.94] and figuring out like how does this thing work how would someone want to own this going forward [4831.94 --> 4838.10] because more things are being built on this sort of protocol and personal storage versus app tier [4838.10 --> 4843.20] scraper sort of thing or it's like app tier is the search engine and i have a website and how does that [4843.20 --> 4847.56] interact speaking of that i need to go fix my website why can't you just fix my github pages because you [4847.56 --> 4855.82] obviously like enjoy messing with dns more than i do and like i love dns i hate it so much i hate it [4855.82 --> 4863.02] so much like uh thank you everyone for listening um please reach out uh online to us if you have [4863.02 --> 4869.08] uh suggestions for a show title that we can continue going forward um and and we will hopefully have [4869.08 --> 4873.92] something for you in a couple weeks on where you can find us going forward for next year for 2025 [4873.92 --> 4878.86] and we will talk to you all again soon i feel like gearheart has to come back for like our new [4878.86 --> 4885.26] podcast oh yeah i would be happy to good luck it was so nice meeting you you too i think you've done [4885.26 --> 4891.10] an excellent job with these episodes i think it was a very nice transition and it's great that you're [4891.10 --> 4897.38] able to carry the torch the spirit continue the spirit of ship it i think people appreciated that [4897.38 --> 4902.14] and i'm looking forward to what you'll do next i'm really excited i feel like this can be the [4902.14 --> 4908.30] new iteration of ship it it's a little scary though thank you see you next time [4908.88 --> 4920.02] thanks for listening to ship it with justin garrison and autumn nash if you haven't checked [4920.02 --> 4927.10] out our changelog newsletter do yourself a favor and head to changelog.com slash news there you'll find [4927.10 --> 4934.30] 29 reasons yes 29 reasons why you should subscribe i'll tell you reason number 17 you might actually [4934.30 --> 4940.08] start looking forward to mondays sounds like somebody's got a case of the mondays 28 more [4940.08 --> 4947.34] reasons are waiting for you at changelog.com slash news thanks again to our partners at fly.io over [4947.34 --> 4953.84] 3 million apps have launched on fly and you can too in five minutes or less learn more at fly.io [4953.84 --> 4959.50] and thanks of course to our beat freak in residence we couldn't bump the best beats in the biz without [4959.50 --> 4964.64] breakmaster cylinder that's all for now but come back next week when we continue discussing [4964.64 --> 4967.14] everything that happens after get push [4967.14 --> 4975.86] everything So [4975.86 --> 4978.22] i love you [4978.22 --> 4982.78] you