Latest / Oxide and Friends / Cultural Idiosyncrasies

Transcript

Cultural Idiosyncrasies · Oxide and Friends

• 0:00
  Adam Leventhal: And I'm gonna be sad if the recording does what it did last week.
• 0:06
  Bryan Cantrill: We did okay though.
• 0:07
  Adam Leventhal: It did all work. It was it was your finest hour.
• 0:11
  Bryan Cantrill: Like It was that's my finest hour.
• 0:14
  Adam Leventhal: Bobby, do you think I mean, what's is there a number 2? I don't know. Like, what what you would like just help me out.
• 0:20
  Bryan Cantrill: Name of a fighter.
• 0:21
  Steve Klabnik: You know what
• 0:21
  Bryan Cantrill: I'm saying? So, hey, I, I actually thought, into mere moments ago, what I'm gonna describe to you is gonna sound like a vivid dream, but I assure you this actually happened. There's a there's a car parked, outside of the office here on the litter box side where we record. This car was parked roughly 4 feet from this microphone with its all its windows down blaring in a loop eye of the tiger.
• 0:50
  Adam Leventhal: Were you imagining do you need to walk outside? Say, excuse me, sir? No. I I'm about to start
• 0:57
  Bryan Cantrill: a podcast. Excuse me. We are podcasting in here, and I will thank you. To no. I mean, I have, like, obviously, so many thoughts.
• 1:05
  First of all, well, I'm like, what's gonna be next after Eye of the Tiger? And so it gets through Eye of the Tiger and starts playing Eye of the Tiger again. I'm like, right. No. No.
• 1:14
  That that's not good. Okay. We're gonna die of the tiger. I one Eye of the Tiger front. And then I'm envisioning we're gonna end up with, like, you know, people are criticizing us for having intro music.
• 1:22
  It's like, well, how do you like us now? We've got eye of the tiger and a DMCA takedown on YouTube.
• 1:26
  Adam Leventhal: So I'm kind of envisioning that Not just intro, background music.
• 1:30
  Bryan Cantrill: Back total background music. The mic is definitely picking it up. And then, like, what are we gonna yeah. What am I gonna make? I did actually go outside.
• 1:37
  You get that moment of, like, is my phone playing sound right now? You know, you have that moment where, like, your phone just kinda, like, spontaneously starts playing something. It's, like, my phone's somehow playing I have the tiger? I mean, how can it be that it's loud in here? But no.
• 1:51
  No. It wasn't. So I was like, but no. It's all good now. And then I all of a sudden it just stopped.
• 1:55
  So and he they they drop on. I'm not sure if if they just thought better of it, but I also thought that I was having, you know, a little bit of kind of, like, Rocky 3 vibes where I, you know, I can do the little the the kind of the training video where, you know, you and I are both, like, running in the rain or whatever, getting ready for the podcast. So I felt a good little, hip hop. A little montage. Little montage.
• 2:16
  Little montage. Exactly. Yeah. So here we are.
• 2:20
  Adam Leventhal: Yeah. And and you wrote a pod you wrote a blog over there.
• 2:24
  Bryan Cantrill: I wrote a podcast too. Yeah. I wrote I wrote a little blog entry. Actually, so our colleagues, Matt Keeter and Cliff Biffle wrote blog entries and, talking about this, like, really, pretty interesting bug. I'm not sure you how much you'd follow that particular issue, Adam.
• 2:40
  It was super interesting.
• 2:42
  Adam Leventhal: I was not following it live, but but, but the the blogs are great. And I remember Matt's blog post. When when did he because that was a while ago that he wrote that first one, did that first piece of work. Right? He did that for a
• 2:55
  Bryan Cantrill: while ago being, like, I don't know, a couple weeks ago. Wasn't that long
• 2:57
  Adam Leventhal: ago? No.
• 2:57
  Bryan Cantrill: Yeah. I think ages ago. Yeah. Exactly. Ages ago.
• 3:00
  And the, and then so really, like, innovative and important work, and then it kind of unearth this really interesting regression. And I mean, it did strike me when we hit this. And of course, so the way we hit this is, I'm not sure if you were in the office that day, but I was, the, the switch that Arien was throwing this on was right behind my head as I'm, you know, working, whatever. And the fan starts blaring and I'm kind of giving Arien like the total stink eye of, like, hey. Can you get your, you know, get your dog back on the leash over here?
• 3:30
  And, and then he it's like, clearly, he's kinda struggling with it a little bit, and then the fan dies down. And then he kinda wanders over. He's like, has anyone changed the I squared c code recently? And, of course, I had changed the I squared c code recently. I'm like, it's happening again.
• 3:44
  Like, I should never have given Ariane the stink eye because this is my bug. So I was able to say, like, very engaged to look at it. And fortunately, I was like, oh, no. This is actually weirder. This is not this.
• 3:55
  This is something else. And, but it was really quick to debug. And I think I think it it was definitely interesting to me. I actually was very, extremely grateful for it because that tape was just, like, wall to wall with, like, other conversations. And that was kinda, like, my one moment of technical contribution was being able to help debug this.
• 4:12
  But the it was an interesting object lesson, and I'm like, there's a lot of it's kind of interesting here. And Cliff hit on a bunch of stuff that, like, that I I had occurred to me as well. And he hit on some stuff that actually had not really occurred to me, but of course, was also very apt and in particular, Cliff hit on a bunch of these cultural issues that, that were really interesting and were really true. I, you know, I again, the the it was not the first thought that it kind of popped into my head but, like, Yeah, wait, that actually is a big deal. And the, the, the, these, these kind of cultural things that we really take for granted.
• 4:53
  And as I was kind of, I wanted to write up a blog entry that, that kind of pointed to, to both Matt's blog entry and Cliff's blog entry. And I'm like, you know, actually these kind of cultural issues are are interesting, and so that's what, you know, with with everyone's forgiveness, that's what I really wanted to hit on. So that's what I hit on in this blog entry, on on engineering your culture. So talking about kind of the certain things that we have done in oxide and kind of, you know, what you can and can't do when you're you're entering your culture. And, of course, Adam, I'm like, didn't we do an episode on this?
• 5:23
  Like, why is this all sounding this is all a little too familiar. And we did. We did an episode, like, 2 years ago on this. Yeah. And which I don't know.
• 5:33
  I went back and relistened to that. I don't know if you would get, listened to any of that, but it was it was a good episode. It was interesting. Yeah.
• 5:38
  Adam Leventhal: Yeah. Yeah.
• 5:40
  Bryan Cantrill: And, you know, there's, you know, Tom kind of describing how, Amazon's values came about later than we would have guessed. So I think that would that's a great conversation about values in particular. So I don't think we need to belabor that kind of aspect of culture. I think we and Adam, you had a really good summary there at the end of that podcast about, like, hey, you know, like, pick your companies carefully. These things are not totally immutable, but they can be hard to change, and with respect to values.
• 6:10
  What I wanted to get into here, again, with kind of everyone's forgiveness, but I I one of the things I've begun to appreciate is some things that we do and that any company does for ex any company has things that they do that feel very natural that then to someone else is gonna be like, Oh, that's kind of different and okay, yeah, that's that's surprising and I guess I'm a little bit surprised about some of the ones that are surprising and in particular I was with a a group of ctos over the week and midweek and we were describing demoing, like how people demo. And I I I I'm like, you're I guess it looks like what even occurred to me to be a conversation to have with people. Like, doesn't doesn't everybody demo their work? But as it turns out, people do this really, really differently. Yeah.
• 7:03
  I okay. This is actually this is kind of a relief that you're you're as surprised as I am by this.
• 7:07
  Adam Leventhal: Well, both both so I remember when we started demo day, that it was really important. It's like it felt like a real moment. And it was not obvious, at least to me, that that was what we needed, or that it was gonna become such a cultural lodestone.
• 7:23
  Bryan Cantrill: I think that's right. I think it wasn't obvious at the time. It like feels now it feels so important. I don't know how it wasn't, but it really wasn't. I think you're right.
• 7:32
  I think it's the, Sean Klein had kind of started this with a control plane. Remember, Arian actually at one point is, like, hey, I'm, like, watching these demos, and I'm really jealous, and I wanna go demo the stuff that I'm like, I could I just I I wanna be able to go demo stuff too like where can I go to demo stuff? Like, maybe we should just like throw the door open, which is what we did But here's what was surprising about the way other people do this, or maybe even shocking. Like, yeah. I I was talking to another CTO.
• 7:59
  I was like, yeah. You know, I'm I I was in an organization and demos were really unpopular. Like, that seems strange. I so we they demoed, every 3 weeks. Every group was required to demo.
• 8:16
  They had a 120 people, so big engineering team. And on the order of, like, 6 people in a in a group. And you do the math on that, and that's 3 minutes a demo to fit in an hour. Yeah. And it's like, that sound he's like, that sounds awful.
• 8:37
  Like everybody By construction. Yeah. That sounds awful by construction. Every group has to have a 3 minute demo. It's like, man, I am never gonna take demo day for granted again.
• 8:49
  And it was really surprised that, oh, that we have a totally unstructured, you know, whoever shows up the demo shows up the demo kind of a of a model. But I think it's been a really important feature of the model.
• 9:03
  Adam Leventhal: Yeah. Yeah. Like, I I think, I'm sure that behind the scenes, you're gently encouraging folks when you see great work to to put it in front of people.
• 9:12
  Bryan Cantrill: Oh, okay. That that is true. I actually thought you were gonna say behind the scenes, I'm sure you are, like, organizing this. So this is not actually us, like, like, like, just doing, like, doing a a crowd jumping into the crowd every single Friday, but we are actually jumping into the crowd more or less every single Friday.
• 9:30
  Adam Leventhal: For sure. But, like, I mean, I mean, I I do it too, but probably less frequently. Like, when I see something cool or or I'm like, you know, someone's got some work planned. It's like, that'll make a great demo. Go demo that.
• 9:42
  That is a great way to to share. It's a great great forum to share.
• 9:45
  Bryan Cantrill: Yeah. I think it's good too, because I I actually really like people being able to provide all context for that for for what it is, and what you're about to see and why it's hard. And because I think it's also really important for people to appreciate. It's just like, you know it's just hard for people to appreciate why everything is hard. In part because we we we we have to like create our own layers of abstraction.
• 10:13
  Like, Bob is working on this problem, and I can't, like, I I can't get every single detail of this because then it doesn't I mean, I I need Bob to work on this problem, but it that also allows you to kinda lose sight of, like, what Bob is suffering through. And it's important that we get those details, that we we can see some of those details and, and you kinda do it in a way that feel that is interesting and exciting to people and uplifting and I think, you know, and Alan's raising this point in the chat that the it's a another really it's it's like you can fail in a demo. If the demo doesn't work and I remember early days, I remember how many of those demos turned into, like, group debugging sessions.
• 10:53
  Adam Leventhal: Oh, totally. Especially when we were demoing, the kind of macro demo of the control plane and of the software. Do they all turn into demo sessions? Right? It was all I mean, debug sessions.
• 11:04
  It was all just let's get as far as we can and then figure it out together.
• 11:07
  Bryan Cantrill: I am also amazed at how many of our colleagues can debug their own work on the fly, like, quickly. You know, like, it is, like, absolutely stunning where people are like, god, that, like, that totally blew up. Oh, wait a minute. I see what I did. I'm like, how do you see what you did?
• 11:25
  Like, I definitely don't see what you did.
• 11:28
  Adam Leventhal: Yeah. There's one in particular from this past Friday, when they sort of had a dead fright, and I was kind of still wrapping my head around, you know, what problem we were solving.
• 11:36
  Bryan Cantrill: Yeah. And and so having kind of an environment where people can do that is actually has turned out to be really important. And again, I would just assume that everybody does this, but I guess not. I would think It's weird
• 11:50
  Adam Leventhal: to me that you think that because have you ever been into a company where anyone did that? And, like, we didn't do that. Like, we used to be a company that didn't do that.
• 11:59
  Bryan Cantrill: That is So It's true. So I'm
• 12:01
  Adam Leventhal: not I'm not sure why you're so surprised that the thing that we didn't used to do, other people don't do.
• 12:07
  Bryan Cantrill: Other people don't do it. No. You're right. Right. No.
• 12:09
  You're either absolutely right. It just feels like but isn't it feel like extremely natural? Is that just not it? It just it also to
• 12:16
  Adam Leventhal: us. Yes.
• 12:19
  Bryan Cantrill: Here's the thing. I mean, like, look, We do a lot of weird things at Oxide. In fact, I was with an old friend. I got a couple of weekends back, Adam, and he was like, oh, god. You guys do so many weird things.
• 12:28
  I wanna know about all of the weird things you do. Like, we do not we're not that weird. And I'm like, well, actually, I'm not on this. Nothing you should know. Actually, hold on.
• 12:37
  There's nothing that's occurring to me. It's like we actually are a bit idiosyncratic. And I think that, like, some of that look. Some of the things that we do are gonna be really hard for a company to replicate if I I mean, like, honestly, just even forget uniform salary. Just transparent salary is transparent.
• 12:55
  Compensation is not something that a company can easily adopt. That is really, really challenging. So it's like, I get it. Like, that one's gonna be really hard. But I think there's other things we do that, like no.
• 13:05
  No. No. I think, actually, any company can kind of experiment with this. And this one, demo day feels like that. I have, like, a ad hoc regular unscheduled or unplanned, unstructured demos feels replicable.
• 13:20
  Or is that crazy too? I don't know. Am I just so warped? You know what?
• 13:23
  Adam Leventhal: I think it's I think anyone can try it, and I think that it will not attach in a lot of places, and it will attach some. So I think that we have, like, we work with, like, people who like the theatrics of it. I mean, as we've seen in some demos where We do. Where sometimes people pretend like it's driven off the rails. But in fact, that was all part of the bit, you know, or someone chimes in with and and it turns out to be part of it.
• 13:51
  So I think we work with folks who are where there's a certain amount of showmanship, and and bravado and bravery, candidly. Because it you know, but then as that becomes the cultural norm, it takes just a little bit less bravery to to get up there. Conversely, you know, I think if you, you know, I think I can only remember maybe one demo Friday where maybe there wasn't a demo. Right. Yeah.
• 14:16
  Maybe ever. But imagine if you had 22 of those in a row, you know, the 3rd would feel very high stakes. And if you started having 3, where people didn't really have someone to say something, then, you know, it might kinda dry up. So I think that there's some amount of of luck and just the culture becomes what you make it, becomes what you do. And, but but certainly low stakes, try it everywhere.
• 14:41
  But, you know, I I don't I don't know that I've had perfect confidence that it would touch. Even in some of the companies that I've worked for, some of the organizations I've worked for, or maybe even Brian at Oxide before we did it. You know, I think there's a time early on when I'm not sure it would have really thrived. Right?
• 15:01
  Bryan Cantrill: Yeah. Yeah. No. I think these are all very valid points. I also think that, you know, this is kind of interesting, because, you know, one of the questions that kinda came up when we both when we had this previous conversation 2 years ago, but also when we were talking about doing this conversation tonight, people were like, how do you change a culture?
• 15:16
  I I think, Adam, you make an interesting point about, like, the culture kinda is what you do. And this also might be an interesting way of, like, moving a a a culture shifting a culture just a little bit. People are asking, like, hey. Are these demos of personal projects or are these planned work? So they're definite personal projects.
• 15:36
  Although, we have had people who are like, hey, listen, like, I I've been playing with, jujitsu, not jujitsu, the martial art, although maybe that too. But jujitsu, the JJ, the the, the
• 15:49
  Adam Leventhal: the the rapper? Is that a fair
• 15:50
  Bryan Cantrill: You have rapper? Is it not I I you know, that's feels Someone say JJ? I think I'm contractual.
• 15:55
  Adam Leventhal: You just appear. Steve just appeared when that happened. That's amazing.
• 15:59
  Bryan Cantrill: I know what you're referring to is JJ. I actually refer to as, yeah. Sorry, Steve, go ahead. You've you've conjured it.
• 16:05
  Steve Klabnik: This is rain. Rain did Rain did a good demo on JJ because some people at Oxide are using JJ, and it's cool to share that with other coworkers. It's like, hey. You know, you may also enjoy using this as well. And, like, that's not work exactly, but it's work adjacent enough.
• 16:19
  Or, like, sometimes people's personal projects are close enough to work and or, like, you know, there's, like, some blurriness there that, like, it might be useful.
• 16:26
  Adam Leventhal: So Yeah.
• 16:27
  Bryan Cantrill: So to answer the question, it's all very work ish. It's all very, like, stuff that we're doing, but it's also not always like, hey. This is what I am working on right now. It's sometimes like, hey. This is something that interesting that I tripped on or something I tried to take a a quick kind of sojourn over here.
• 16:43
  So it's it it is all but it all is all work adjacent, I would say.
• 16:47
  Adam Leventhal: But but but even that speaks to a certain freedom that we have in our culture, which is to say, you know, there isn't some master task list that people are plucking things off of. And it's not some sprint where you're supposed to be doing items 1, 2, 3, and 4, where you'd have that fear. Where if you had done items 12, and then took some sojourn, that you'd say, well, if I demo this thing, everyone will wonder what happened to items 345 or whatever. So people have a lot of flexibility and freedom in terms of prioritizing how they spend their time.
• 17:16
  Bryan Cantrill: Yeah. That's right. And I think that that we also I think, also importantly, you know, I I encourage people to demo for its own sake. It's not something that is like, you're not required to give, you know, whatever, and demos per m time or what have you. So it's not or and certainly not having the entire team demo would be, w would be really, I, I, I don't understand.
• 17:42
  I just think you would give the term demo a bad name. And isn't, like, demo an agile thing too? This is, like, epic story demo. I don't know. There's so many nouns now.
• 17:53
  I I I don't
• 17:57
  Adam Leventhal: I think that the My my favorite kind of work, maybe work adjacent fun demo, was and Andrew came on the show to talk about his work, with Wicket, our our kind of c l our 2e, our text based UI for doing initial setup. And he embedded, like, a game with Intuit where you
• 18:19
  Steve Klabnik: He did.
• 18:20
  Adam Leventhal: Where where you drop oxide racks into a truck that is moving across the bottom of the screen. And it's just so quaint and delightful in the midst of, like, this this setup app. It's awesome.
• 18:33
  Bryan Cantrill: And am I remembering this correctly? Wasn't he terrible at the game?
• 18:37
  Adam Leventhal: Oh, yeah. He was garbage at it. Just like he he it's like he had never played before.
• 18:41
  Bryan Cantrill: Like he'd never seen this thing before. I mean, it's just like you I mean, I kinda like, I guess you've been busy, you know, adding a, the Easter egg for the but yes. That's right. That I was, that that was fun.
• 18:52
  Adam Leventhal: But but without fear and, like, I mean, Andrew is not a person who, like, has fear as it is. But I think this is an organization where you can show up and show how you embedded an Easter egg into it. And, you know, it's, like, sort of clearly time boxed, but also really fun and in keeping with the values of the company. It's something delightful. And so, you know, you can present without fear.
• 19:14
  Without anyone saying, hey, you know, you did this, but you didn't review my code, or you did this, but you didn't fix my bug, or whatever.
• 19:20
  Bryan Cantrill: Right. It's something I'm always afraid of. That the, that someone's gonna look at someone's like, you know, you had time to like this tweet on Twitter, but you didn't have time to review my code. It's like, okay. Look.
• 19:31
  Yeah. Sorry. You're right. I know you're right.
• 19:34
  Adam Leventhal: But no.
• 19:34
  Bryan Cantrill: It's true. And and I think that, that and that and I also just I also love that it is not limited to a particular domain has been great, and we've gotten from I mean, I I mean, I've got so many demos that I love, but I loved Augustus demoing the integration that he had built into Salesforce where a someone coming to Salesforce could automatically have a silo provisioned on I just thought it was like and using Drop Shot. I mean, you must love that demo
• 20:00
  Steve Klabnik: because that
• 20:00
  Bryan Cantrill: was Oh, yeah.
• 20:00
  Adam Leventhal: It was outstanding.
• 20:02
  Bryan Cantrill: Building on a bunch of work that you've done, which I also I also actually really love is that another kinda aspect again, I don't know how much of this is, like, our culture versus just, like, something that someone could go build readily in a different culture, but I I love that that all so much of the work that we demo built on the work of other people, and it is really satisfying to see someone else using the thing that you built to do something that's, like, wow. I didn't really intended to be used that way, but that's really neat. Or or maybe I didn't intend to be used that way, but I've never, you know, I getting actually seeing someone build on top of it is really kinda uniquely satisfying.
• 20:37
  Adam Leventhal: Yeah. No. Totally. Just see seeing the things that you thought of as, like, you know, a reach in terms of technical work than other people just building on top of. And this might come out as weird, but kinda taking for granted not to say that that they're like, oh, this is just work that doesn't matter, but rather just like building on top of that as foundational.
• 20:59
  The thing that you thought was, like, kind of exciting.
• 21:02
  Bryan Cantrill: Right. The thing that you're still, like, holding your breath to run, and they're just, like, casually, like, oh, no. It's like I'm just just using this thing. You're like, oh, that thing, like, okay. Yeah.
• 21:09
  Does it work for you? Wow. Okay. Great.
• 21:11
  Adam Leventhal: That's like I mean, on the on that topic, literally every time I demo, like user land DTrace to folks, like there is a part of me that holds its breath, as I'm like, okay, are we instrumenting the instructions properly? Like, I still have that anxiety. Yeah. We are we are folks. Like, we are.
• 21:28
  Like, it works. But every time, I just like, a little bit more
• 21:30
  Bryan Cantrill: But, like,
• 21:30
  Adam Leventhal: why would
• 21:31
  Bryan Cantrill: you say that? Like, why would you volunteer that? You know, you're, like I mean, clearly. Like, what do you of course, it works. Why are you saying wait.
• 21:36
  Wait. Why are you saying that? It's like, because it barely works.
• 21:40
  Adam Leventhal: It works. It used it used to really not work too. I mean, that that's that's true of all technology. Right? That is why we have, like, trepidation around it.
• 21:48
  But before it worked, there was a long time when it didn't.
• 21:51
  Bryan Cantrill: Yeah. And I remember, like, also, that mean the, you know, Lukeman on the demo of the actual full rack coming together was also, like, a big one. It was that that was a huge, milestone. It's like it's been a and it so that's been really great. And is he but as you say, yeah, it shouldn't feel obvious that everyone does this because we ourselves didn't do this.
• 22:10
  So, yeah, dummy. Of course, it's not obvious. Wasn't obvious to us, but that that one has been really great and something that I I think you could do pretty lightly. I also think that, like, if it were a bust, it would be kind of interesting why. Like, it may help understand, like, for example, if it's a bust because demo goes poorly and then someone, like, they there is a lot of fear associated with that.
• 22:33
  It might be kind of interesting to dig into, like, alright, why why is that? Like, why are we afraid to I mean, this is, like, the safest environment we should have for failing, namely, like, kind of in the privacy of our colleagues. It's just, like, just us and our colleagues here. You know, we're not we're, you know, not putting it on the podcast, as I don't know if you had people do this to you. I've been to know an increasing number of people tell me things, but, like, hey, this is, like, in confidence.
• 22:57
  I mean, no, that's fine. Like, like, not for the podcast. I'm like, alright, I get it. Like, am I that am I that loose? It's like everything like, I'm not just, you know,
• 23:06
  Adam Leventhal: I I think it speaks volumes that nobody has said that to me. And I'm I'm sure plenty of people have said that to you.
• 23:11
  Bryan Cantrill: Oh, my god. Daiko man. Joking
• 23:12
  Steve Klabnik: water cooler.
• 23:13
  Bryan Cantrill: This week.
• 23:14
  Adam Leventhal: I'm I'm in a joke
• 23:15
  Steve Klabnik: of water cooler sometime last week that was like, you know, this is like also kind of like a podcast. And everyone was like, absolutely not, Steve. We're not
• 23:23
  Adam Leventhal: we're not a podcast.
• 23:25
  Steve Klabnik: I like talking about just to my coworkers. No one else needs to hear this. I was like, I know. I know. I was just joking.
• 23:29
  I was just saying it kinda gets Okay.
• 23:31
  Bryan Cantrill: Actually, Steve, I'm glad you mentioned this because I did not mention this in the blog entry, but this is another thing that is an idiosyncrasy. That's exactly the kind of thing I'm after in this conversation today. Steve, could you describe water cooler? Because I think that's kind of interesting, especially for a remote company.
• 23:44
  Steve Klabnik: Yeah. So it's funny because, like, I I forget what the original description of this meeting was, but it was one of those things. It's called the daily water cooler, and it's basically, like, right before I eat lunch. It was very convenient. So right before I eat lunch, I get to hang out with my coworkers and, like, talk to them.
• 23:58
  And, like, you know, now we have 2 or 3 3 or 4 Texan oxide folks. But, like, for a while, I was, you know, basically, you know, by by myself or where somebody else is, like, 45 minutes away in Austin. So I rarely get to see everybody, and so it's really nice to have a meeting that's just, like, hang out with your coworkers and just don't talk about work. And about 99% of the time, it, like, is not about work. And so we do have a decent you know, some people never show up.
• 24:23
  Other people are there literally every day and just kinda, like, you know, hang out time to get to know each other better. And I think it's really good in terms of, like, you know, a lot of people talk about, like, oh, you gotta be in the office to have a culture, and it's just not not really true, or at least not in the way that we do it. It's, like, worked out pretty well where I feel like I know my coworkers even though I don't get to see them physically every day.
• 24:44
  Bryan Cantrill: Yeah. And so we and, Emily in the chat is like, hey. We do that kinda weekly, but I've had a I I feel like I'd have a hard time getting folks more often than not. I think part of the key is, like, just gotta make it shrink to fit and the so we do do it daily and it's, like, whoever shows up shows up. And it's okay if, you know, if sometimes there are a couple peep there are only a couple people there, and sometimes there are, like, you know, a lot of people there.
• 25:06
  And the, it ends up being kind of deliberate it's very light. It's people talking about stuff that you'd kinda talk about at the water cooler. It's a lot of I mean, okay, Steve, what percentage of it is us being trolled by whatever's on the Internet? Probably.
• 25:22
  Steve Klabnik: Oh, I mean, like A decent fraction. Yeah.
• 25:24
  Bryan Cantrill: Right. Yeah. Totally.
• 25:25
  Adam Leventhal: I was gonna go with 80%.
• 25:27
  Steve Klabnik: You and me. Yeah. Yeah. Right.
• 25:29
  Bryan Cantrill: Right. But if, like, if you wanna talk about, like, Sam Bankman Fried's sentence too light or not heavy enough, it's like, the sorry. I guess I
• 25:38
  Steve Klabnik: We also have the daily
• 25:39
  Bryan Cantrill: update on the world
• 25:40
  Steve Klabnik: family going for a while recently.
• 25:42
  Bryan Cantrill: That was a
• 25:43
  Steve Klabnik: popular recurring theme. We used to do yes, yes, no, a long time ago. We haven't done that in a
• 25:48
  Bryan Cantrill: little bit. So it it that and actually so, Steve, the origin of that, for whatever it's worth, I think he had we we were doing that by the time you joined. Yes. Maybe?
• 26:00
  Steve Klabnik: Yeah. Because it used to be it used to kinda be like, Steve, did you hear this post on Twitter? And I was like, of course, Brian. I've already heard this post on Twitter. And we'd basically just do that.
• 26:08
  You know? Or but, yeah, like, that was I had been introduced to Yesus Snow, the podcast, because of us talking about it in water cooler. I had not heard of it.
• 26:15
  Bryan Cantrill: But I said, I wanna go. But we we were doing water cooler when you joined the company in in the summer of 2020. And so I think that we'd started that just previously, and this is, another this was because, Adam, my memory is that that is a post pandemic, because Cliff, in particular, reached out to me being, like, hey, I am, like, locked in my house, and I need some human contact that's light. Like, alright. Let's just do this every day, and kind of whoever shows up, shows up.
• 26:43
  And so my memory of that was that was a kind of a pandemic creation that ended up being, I think, really good for a remote company. It's a really good way to get to know people, I feel, and to get to to learn more about them.
• 26:57
  Adam Leventhal: Totally. I mean, we we start I mean, we've we we've been remote, you know, or having some members of our team remote, I think, since I joined. Yeah. Since the beginning. Which was which was pretty early January 2020.
• 27:08
  And I felt like I was going to the office for a long time. Obviously, I wasn't because it was it was only a couple of months of commuting. But I'm not sure we I'm not sure we would have done it absent the pandemic. And I think that remote folks might never have like, we might have never have figured out how to do things properly or in this capacity for remote folks. And we built a ton of momentum as you say.
• 27:31
  I remember even we do a a no meet Wednesday. They were
• 27:35
  Bryan Cantrill: Oh, that's another one. Oh, yeah.
• 27:38
  Adam Leventhal: You're saying, like, oh, let's not you know, but we won't do it on Wednesday. Right? And I think all of us feeling like, no. Like, Wednesday is arguably one of the most important days, because it's like a moment to check-in. It's a moment to go from I'm doing my my or at least for West Coast folks, I I realized, Steve, for folks in other time zones, it's like a kind of different moment of the day.
• 27:57
  But for folks on the West Coast, it's going from, getting ready to being at work. And it's kind of timed at that transition.
• 28:05
  Bryan Cantrill: Yeah. That's okay. So I'm glad you because that's another thing that we do. And again, I know it's not unique to us, but but not having meetings on Wednesdays. Oh, yeah.
• 28:16
  Adam Leventhal: That feels like the easiest one to bring somewhere else where if they have not figured it out, you will be a hero. Just like having protected time is is just
• 28:27
  Bryan Cantrill: yeah. That one and that one was survival for me personally. Because I I and I know I wasn't alone in this, but, I was having and I Steve, I can't remember when we started that, but that was
• 28:42
  Steve Klabnik: I remember you decreeing it, but I it was really early on in my tenure. Like, I remember you saying, like, alright. Let's do this now. Or at least I remember that. I remember becoming pot becoming policy.
• 28:53
  Let's put it that way.
• 28:54
  Bryan Cantrill: I mean, I did decree it. I actually did decree it. It's a bit out of character for me. I did. I, I had to, because I was the only focused time that I had was on weekends.
• 29:08
  And I'm like, this is not good. This isn't it is really not good when you're like, oh, I will wait until the weekend to have focused work time. I'm like, that is not healthy. And I could see that for myself. I'm like, I, that is not going to be sustainable for me.
• 29:22
  I'm going to break, I'm going to physically break if we do this. And so, yeah, I, it is definitely not the rule by decree is definitely not the way, but the the I I feel on that one. I've I've I don't know.
• 29:34
  Steve Klabnik: I This was a popular the people rejoice at the benevolence. No. I'm
• 29:39
  Bryan Cantrill: just kidding.
• 29:40
  Adam Leventhal: No. For sure. I mean
• 29:41
  Steve Klabnik: Probably because I probably said that even just because, like, so recently on Hacker News, everyone has decided that you personally make every decision to the company, and I have to keep Which
• 29:49
  Bryan Cantrill: is either combo.
• 29:50
  Steve Klabnik: Brian did not name that repo. Cliff did act and, like, all this stuff. Because, yeah, I I also don't wanna be like, Brian doesn't do shit. Why are you saying that he does all this stuff at the at the job either? You know, there's a balance to be there, Hatter.
• 30:01
  Like, the like, I think, like, we're talking about culture. Right? A lot of people because, like, you know, you're well known, assume that, like, you decree things all the time, and that's, like, not actually true.
• 30:11
  Bryan Cantrill: Yeah. But I I this one, I I may have tried to get another finger on the revolver on this one. So but I I I I I really needed that one. And I think that's been important, I think not just for me. I think that's been important for lots of people to and, actually, what's been was really interesting for me is the degree to which that was also really important for for for Steve.
• 30:33
  Steve Tuck, our boss. Right? The for the that was important for, like, the for Travis and sales. Like, the it was important for everybody to be like, I know I actually need, like some, some time that I know I can get some extended time to work on the things that I need to work on. And, like, yeah, not necessarily an engineering artifact.
• 30:52
  It's an email, or it's just, like, it's, you know, dealing with this, you know, dealing with this payroll provider or whatever it is. Right? It's been, that's been really and I also think I I know some people do this on Friday. I personally love doing this on Wednesday. I don't know how people think about the the day choice, But I like Wednesday because it gives you that, like, midweek island that you can grab on on on on Tuesday.
• 31:16
  Adam Leventhal: I think that, sometimes it can feel too like, if it's on Monday or Friday, it can feel too tempting to be like, well, you know. Yeah. Or or at least that can be the perception. And I think having it on Wednesday certainly gets rid of that perception.
• 31:32
  Bryan Cantrill: Yeah. I think that that's right. I think that that by getting it on Wednesday, you and I so to me, it's been great. I think that it's also I think it it has also at least for me personally, it has some kind of, you know, we do demos on Friday. There's kind of this flow of the week where you've got demos on Friday, you've got this kind of focused work on Wednesday, And if you I mean, plenty of us have been like, okay, I really wanna work this Wednesday because I've got something I wanna I'm trying to get ready for Friday.
• 31:56
  So, that's been another, another big, like, idiosyncrasy. I think also just, like, in general, the pandemic has been really important for us. And because we were, Adam, as you said, like, we were remote from the beginning with Patrick joining the team, you know, the earliest, in that kind of earliest, tranche of folks, and, being remote out out of Minnesota, but it was only with the pandemic that we were all forced to be remote. I remember being, like, early on to Patrick being, like, this kinda sucks. He's like, yes.
• 32:25
  Welcome. Yes. Yeah. Finally. Yeah.
• 32:28
  Finally. What do you think I'm dealing with for 10 years, pal? But I feel like we got a lot of stuff working a lot better, including recording every meeting, which I know we've talked about here occasionally as well, and my my my prediction of of of amnocracy. Do we always talk about in here? We talk about it a lot?
• 32:45
  Adam Leventhal: Some of us do.
• 32:47
  Bryan Cantrill: Okay. I know. I I get it. This is an intervention. But the we do not record that daily water cooler.
• 32:52
  We don't record 1 on ones. We don't we don't record every single conversation, but we do record every meeting, and that has been I think that's been really huge. That's another one. I'd be really curious how many I'd like, our other companies doing that, because I feel that's a really good thing to go do. It it's, that's been really valuable for for for us anyway.
• 33:17
  The, you know, we got some very early advice from, Jeff Rothschild, who, we, Jeff Rothschild was the, we had interviewed on for on the metal and was dispensing some free advice, which was great, from the information of the company. And one of the things he said is make sure everyone's invited to every meeting, and recording every meeting is a way of basically assuring that. One of the questions was do we record the demos, and we absolutely do. And that's recording the demo, it's been great to, to have people be able to to, kinda go back to past demos and watch them. The other thing that's been super valuable to do is recorded debugging sessions.
• 33:55
  And Adam, I'm not sure how much you have done this or Yeah. I mean, I know, like, Alan has done this a bunch and holy mackerel is that valuable to go back to a, you know, you're all working together to debug a problem. And it has been I I I mean, I've gone back to those recordings for a bunch of different reasons. I sometimes you go back to a recording of a problem that, like, you worked on a while ago, and then kinda had to drop and needed to come back to. It's really nice to go, like, watch a recording of yourself, to, debug say in particular, if Daniel and I were debugging a power sequencer problem, I had to go work on something else.
• 34:32
  I came back to it, like, only like 3 or 4 weeks later, and I went back to watch the recording to kind of ramp up on it, and I did not recognize a single word I was saying.
• 34:40
  Adam Leventhal: I was like,
• 34:42
  Bryan Cantrill: who is
• 34:42
  Adam Leventhal: this guy? This guy is really insightful.
• 34:44
  Bryan Cantrill: Right? Like, man, this guy's got a lot of state on the problem. Like, let this guy solve the problem. I'm like, I do not I literally have to like watch myself. I'm like, I need to rewatch myself to understand, like, past Brian, what did you know?
• 34:54
  Tell me your secrets. Yeah.
• 34:57
  Adam Leventhal: My favorite one of those actually is is one that, I wanna do a I've wanted to do an oxide and frets on, but that, what turned out to be the, the chip bug or the speculation bug. I'm not I'm not sure how much I'm supposed to reveal here. But, Can't reveal as much as
• 35:12
  Bryan Cantrill: you want. And, you know, I know we do we do need to do an episode of this because they no. And you're right. This is a this is a, this is a banger of a bug.
• 35:17
  Adam Leventhal: Yeah. But but we had, tons of people involved in the bug debugging sessions, throwing out lots of different ideas. And there's debugging, I feel like around the clock. So it it gave you an opportunity, like, you didn't have the FOMO of, like, okay. If I drop off that this is all just written on the water or whatever.
• 35:34
  But I can check back in and find out.
• 35:37
  Bryan Cantrill: Yeah. It was like c span for debugging a problem. It was just like, oh, like, they're okay. They're not like they've drifted up. There's only 1 person on the floor right now debugging.
• 35:44
  You know, I
• 35:45
  Adam Leventhal: it was Frank. Frank, Frank. Brian's still on there just, like, reading the phone book, I guess.
• 35:50
  Bryan Cantrill: The phone Filmmastering. The Correct. And so so these we were recording these in Google Meet as because someone is asking, well, do you just get the audio or to get the terminals? And then we have basically, when someone's driving, they're sharing their screen. So you are seeing what they're kind of doing and it is that that one is is is super valuable to Adam because you debugging is one of those things that scales really, really well with people.
• 36:17
  Adam Leventhal: Yeah. Yeah. You know what
• 36:18
  Bryan Cantrill: I mean? Like, because you if you have, you know, if you can get over that you've got, you know, 20 people looking over your shoulder, but if you are, if you are if you have 20 people or more easily, like, you know, 5 or 6 people, like, watching someone debug, it is so valuable because someone could be, like, I don't know. I get, like, what about this idea? I mean, just kinda even throw out ideas. And, Adam, it was your idea, actually, is if I recall that you were we were looking at just wild data corruption that made no sense.
• 36:48
  And if I recall correctly, it was your observation that, like, hey, isn't this that virtual address looks a like, isn't it, like, just kinda close to the virtual address we saw earlier?
• 36:57
  Adam Leventhal: I mean, we well, I saw this like clustering of addresses. And again, we should we should, we should do a whole episode of this one and bring the whole cast of dozens who worked on it. But part of the parallelism you're describing is, like, I went out and pursued some kind of arguably stupid hypothesis of, like, let me just look at these numbers and see where they are. Which if if I was single threaded, if it was, like, one person working on it, you're like, well, that's, like, the 10th idea. So, like, we'll do that, but I guess but only when we've exhausted literally every other avenue.
• 37:30
  Whereas, you know, when when everyone else has hands on keyboard, it gives the opportunity to go pursue lots of things in parallel.
• 37:37
  Bryan Cantrill: Yeah. And I I mean, that also feels like something that that pretty much anyone can maybe how much people are doing this. I think it'd be really interesting to watch, and this is an example of of being remote is a huge, huge feature, where if you are all together in an office, everyone's gonna have to go into this. I mean, it's, like, very, very helpful to be remote. You know another one, as long as we're on the subject, just the presence of the chat in a meeting.
• 38:05
  And again, I got to assume that, like, a lot of companies kind of culturally have this, maybe not, but I assume they do at this point, where when you are a remote company, you have that chat that I have found to be really important as this, like, kind of parallel conversation that and we have a kind of we definitely have a, like, a cultural acceptance of kind of witticisms and banter happening in the chat. But it's also really, really helpful when someone can ask a question, and someone else can answer it in the chat without the whoever is speaking having to, like, pause to answer it. I think that's a really important kind of thing.
• 38:46
  Adam Leventhal: Yeah. I agree. My favorite
• 38:47
  Steve Klabnik: all do adopt the the 4th person pronoun, though. That hasn't happened in Oxide yet, but it's only gonna be a matter of time, I think.
• 38:54
  Bryan Cantrill: A 4th person pronoun? What do you mean?
• 38:57
  Steve Klabnik: We don't address chat as chat. Like, we tend to, like, talk to individuals or, like, respond to people's oh, yeah. Hey. Thanks, Brian, for linking that thing. Yeah.
• 39:06
  Totally. We don't say, like, oh, chat. Like, thanks, chat.
• 39:09
  Adam Leventhal: The collective chat pronoun. Interesting. Yeah.
• 39:12
  Steve Klabnik: Yeah. This
• 39:12
  Bryan Cantrill: is, like, a thing that's happening actively.
• 39:14
  Steve Klabnik: Yeah. Yeah. Actively. Yeah. Yeah.
• 39:15
  Now, amongst the kids, the you, this from what I'm told,
• 39:18
  Bryan Cantrill: is that, chat
• 39:19
  Steve Klabnik: as, like, a because you're so used to this being, like, a thing that, in many cases, you know, you're in a friend's group and you'll say, like, hey, chat. What do you think about x or whatever? Yeah.
• 39:30
  Bryan Cantrill: And it and is chat, like, a synonym for, like is it, like, peeps, or is this a synonym for, like, boy, boy?
• 39:36
  Steve Klabnik: Person pronoun. Right? Is this kinda like you're referring to the group of you're referring to the group of the people who are not, like who are present but are not, like, currently involved in the conversation directly is kind of the, like, the vibe. Yeah.
• 39:49
  Bryan Cantrill: The okay. Interesting. I you know, this is gonna be the I I'm I'm now, some of the things my daughter says around the house are now making a little bit more sense. So when my when my daughter is addressing the chat at the dinner table, I think I know what's going on now.
• 40:02
  Adam Leventhal: Finally.
• 40:03
  Bryan Cantrill: Yeah. Finally.
• 40:04
  Adam Leventhal: I mean, there is a quality of chat where, like, sometimes Steve, again, our our c other Steve, CEO and boss, will give a long presentation where there's been a very busy chat. And he'll say, okay. Jeez. I I missed a lot of messages in chat. And and Brian, often your summary is either it was all answered, or it was all Blather, or kind of 5050.
• 40:25
  Bryan Cantrill: Little both. It was all Blather that was answered. It was all banter. Yeah. No.
• 40:28
  Totally. No. I think it's actually and I think, again, in all hands okay. So another thing that we do I mean, I I know, like, lots of companies have in all hands. I think that, like, the remote world has made it easier to do in all hands, because it's like, you don't have to physically you can just get people together pretty quickly.
• 40:45
  I feel another thing that we do, and maybe this is just like a pure outgrowth of our transparency, but I think it's really useful, the entire company, to see what the board sees. Are we going through our board deck? I don't know, Adam, what you feel about that one.
• 40:58
  Adam Leventhal: Yeah. I mean, I I did that at my company as well. I agree that I think it's weird. Right? I think that is idiosyncratic in that I think there is a mystique around the board where folks aren't necessarily, feeling comfortable sharing that.
• 41:12
  Although, I, you know, I learned that from the company I was at previously, not a very transparent company, but one where, the CEO, I think, did a great job of, like, sharing, you know, explicitly not all of the content of the board deck. Right. Because there's some some parts of the board deck that he couldn't share. It's fine. But I thought he did a good job in terms of, like, making sure that the company at large knows what's being represented to the board and that there are no surprises.
• 41:39
  And, and, you know, it also keeps everyone honest in terms of the material being presented to the you know, if I were on a board, I would love it to hear that the content was being presented to the team. Because I know I was not getting, you know, BS.
• 41:55
  Bryan Cantrill: I think so too. I think you'd almost like a you'd almost like a a board member should be asking that question. Like, hey, is are the employ are the rank and file seeing what you're presenting to us? Because, yeah, it's really important that we see what we're telling the board, and and I I think it's just, you know, what we're committing to and and what, you know, what kind of we're expressing as as top priorities. So, yeah, I I I think, I think this is really important, to to be kind of present that, get to give that transparency, and and I can't imagine kind of doing it any other way, but I I can appreciate that not every company, would do it that way.
• 42:34
  Adam Leventhal: Yeah.
• 42:36
  Bryan Cantrill: Other idiosyncrasies, the, is we mentioned, obviously, recording every meeting. The, one thing that that just to get it out there, because I know this is something that can be a hot button, and I know it's perf season out there. So if you're going through perf wherever you are, we have no formalized performance review process. And, that one, I I think, is something that is and I give presentation on this, you know, years ago, but I just found that, formalized review didn't actually solve any problem. It introduces a lot of problems, but doesn't solve any problems.
• 43:16
  If I give if I I just yeah. Go ahead, Adam.
• 43:19
  Adam Leventhal: Can I ask you oh, just a question? When you say we have no formalized review process, what what do we have a review process?
• 43:26
  Bryan Cantrill: Okay. Look. No.
• 43:28
  Adam Leventhal: No. I'm I'm asking what formalize is doing. I'd like
• 43:32
  Bryan Cantrill: Right. Right. Right.
• 43:34
  Adam Leventhal: So maybe I've not done anything worthy of praise is what I'm saying.
• 43:37
  Bryan Cantrill: No. No. Okay. No. So I I got you.
• 43:39
  I know. It's a we have Have
• 43:40
  Steve Klabnik: you been keeping up with your performance numbers for the last year. I don't know. I don't know what
• 43:44
  Bryan Cantrill: Who knows? This one's wrong.
• 43:46
  Adam Leventhal: I don't read those emails.
• 43:48
  Bryan Cantrill: That explains a lot. Okay. Alright. This is all coming. Okay.
• 43:51
  So you weren't blowing me off. And, yet you, I was amazed that you were, no. So okay. Look. When you say I'm saying we have no formalized form to reprocess and you're like, do hey, pal, admit needless words.
• 44:06
  We could actually get rid of a bunch of things. So here's what I do I do believe in feedback, especially and this is, like, this is hard. I don't know. I'm not I I, you know, I it's something that I
• 44:17
  Adam Leventhal: You wouldn't have to do it in this forum, but okay. I'm ready. I'm ready. Go ahead.
• 44:20
  Bryan Cantrill: I'm ready. Okay. Yeah. Are you ready? So okay.
• 44:21
  So, Adam, I'd like you to sit down, please. So,
• 44:24
  Adam Leventhal: feeling in the organization.
• 44:26
  Bryan Cantrill: There's a feeling in the organization. Let me actually hold on. So, actually, I'm gonna hand back the paragraph that I had you write. I'm gonna hand back to you with introduced grammatical errors as your formalized review for the year. No.
• 44:38
  So I I think the thing that I do try really hard to do, and actually is increasingly important, especially as we get bigger, but just in general, I I think that the the most grievous sin of any kind of leadership is to not pass on positive feedback, which is actually, an easier sin than it might sound. So in other words, like, I I really try when someone is praising our colleague to me, I'm like, let's go pass that on to that person. Let's make sure that that person knows. And the because I think it's it's positive feedback is extremely important. And it's extremely important that people see, like, hey.
• 45:20
  Like, the work that you did was important and is is appreciated. I think that's the most important kind of feedback. And I think, like, feedback of, like, hey. The the the I think feedback where you're trying to adjust behavior or trying to, like, curtail negative behavior, I think, is also important. But, in general, I've always found that, like, there's a there's a better bank shot.
• 45:44
  In other words, like, the negative behavior is coming out of something else, and you can kinda fix the other thing. The, giving people, like, a direct feedback of, like, hey. You need to be do less of this. It's often that's gonna backfire is what I found. Yeah.
• 45:59
  But maybe I just
• 45:59
  Adam Leventhal: I think that's right. I let me try this out on YouTube, Brian, because I also think that so first on the on the praise, on, like, positive feedback. 1st, like, I think it's really easy to see something that someone else has done on your team and say, well, that person is an adult. Like, I don't need to pat him on the head or whatever. Like Yes.
• 46:21
  But but nobody I remember, actually, this is years ago years ago at Oxide. I was having a low moment. I can't remember why. I mean, probably because it was, like, the depth of the pandemic and whatever. But, you know, someone just said, hey.
• 46:33
  Nice work on this thing. And it just filled me up so much. And I think that nobody's ever taken praise and turned around be like, what an asshole. Like, how dare you Yes. Say that I did a good job on a thing.
• 46:48
  Especially when you mean it earnestly. Like, when when I mean, obviously, if if people have not done a good job and they're in a tallow praise. But when people see real work that you really appreciate, like, it is it it can feel not worth your time. It can feel redundant, but it is almost never unappreciated. I mean, it's almost always the right thing to do.
• 47:06
  Bryan Cantrill: Yeah. I think it I and I think and one further, I think that people feel that if I give praise for this well, first of all, this thing is obviously praiseworthy. So this person clearly knows that what they've done is great. It's like, well, they may or may not know. So, like, you think they know because you think it's obviously great.
• 47:23
  It giving voice to that is, is probably a good idea. And, but I think also importantly, like you don't cheapen yourself by praising someone else.
• 47:33
  Adam Leventhal: You know, especially when it's earnest, like you do, if it's if it's not earnest. Right. But but if, yeah, it is. It doesn't take anything away from you. Right?
• 47:42
  It's not zero
• 47:44
  Bryan Cantrill: sum. That's right. And I and I think it's really important to then, you know, if you feel it, say it. Like, if you if you're feeling positive feedback, like, definitely say it because I think it's just important that we know that, like, oh, god. I didn't really okay.
• 47:57
  Yeah. I didn't realize that someone appreciated that or, oh, thanks. So That was, and then even if you know, it was great, I think it's helpful to know like, yeah, I like that was a really hard push and I it was great to get all the positive feedback. I mean, I you know, our our colleague, Ben Nacker recent recently demoed a, a new query language at at demo Friday. You might think, like, what is wrong with you people?
• 48:20
  Like, how many bus do you invent reinvent everything? And the answer is, like, no, but also yes. It but the there's a very good reason for a new query language, and I think but I think there was a it was it was such an outpouring of enthusiasm for that, which I think was great, very big, because Ben had been kind of slogging on that stuff. It's been like, it's hard, you know? And it was, I think it would felt really great.
• 48:42
  So even, you know, I even though I was the nth person to say it, I wanted to make sure that I also added my own praise to the pile because it's great work. And, it's important that people know it's like, no, this is this is important.
• 48:54
  Adam Leventhal: Yeah. And and I'm gonna go a little further out into the limb, which is, when it does come time to, like, when you when someone has stepped out of line or when someone someone has done something that you find objectionable or whatever, it it also makes it easier for you to say and for them to hear when there's that foundation of trust
• 49:12
  Bryan Cantrill: on the foundation.
• 49:12
  Adam Leventhal: This person's told me other things that have been that that have resonated and been positive. So when they're saying, you know, this felt wrong or, like, wasn't right or or whatever, I I think it's on this foundation of trust.
• 49:26
  Bryan Cantrill: Oh, okay. I I I yeah. Yeah. Yeah. I see where this is going.
• 49:28
  Just give it to me.
• 49:29
  Adam Leventhal: Just give it to me. Just building no. We're building to it. No. When when I I took some, I think the best management training I got was actually this, positive coaching allow alliance for Yes.
• 49:41
  For coaching little league. And one of the things that stuck with me there were actually, it was great training. San Francisco little league was terrific. But one of the pieces was that you need to have this ratio of, of 5 to 1 praise to critique. And 5 to 1 was the number that they gave for, you know, the 9 year olds I was coaching.
• 50:00
  But I actually think, like, it's not a bad number. And you whatever the number it is, it it it might vary, but it's it's not far off. And it's not like and they made a great point of, like, it can't just be ephemeral praise. But, like, nice work tying your shoes. Nice work showing up on time.
• 50:15
  Like, also, like, you're a terrible hitter.
• 50:18
  Bryan Cantrill: But, you know, we should Right. Right. I need to rattle off. Let's see. 1 I gotta dig up one more thing.
• 50:25
  So let's see. You're wearing you're you're wearing your pants. So congrats. Right. Also, it's like, god, you look lost at the plate.
• 50:31
  Can you
• 50:32
  Adam Leventhal: It's just just I don't even know how to coach you. You're so terrible. But I but I I think, again, pick whatever ratio you wanted. Every human has, like, a different ratio, but, people need it. You know, like, if if all they hear is negatives, then they just stop listening.
• 50:49
  Bryan Cantrill: Yeah. And I actually think that every, just as an aside, if you've got a kid who is an athlete at all, who's just engaged in athletics at all, you yourself should at least take this positive coaching alliance training, and there's a bunch of stuff that I love out of that because I don't wanna take the same training when I when I coach baseball. One of them is just like and I I'm biased towards baseball. That's what my my oldest big baseball player. I love next pitch.
• 51:15
  Next pitch is such a great life lessons. Namely that, like, you just made a mistake, we need to focus on the next pitch. Just like it's, you know, you gotta brush it off and go to the next pitch, which I think is also great. And then I think that they the other Adam the thing that really stuck with me, and again, if you're a parent of a of a kid who is engaged in any athletic activity, be they terrific or be they really struggling, in terms, like, how to engage with your own kid. And I just love to, like, just be sure to, like, if you wanna talk to your kid after a game, just be sure you say, hey.
• 51:46
  I loved watching you play. Like, that's it. You don't need to and I I think that's great. You know, I think that, like, that kind of, that's the best feedback to give your kid is, like, I love watching you play. I I thought it was fun watching you play, and, I think that it, just anyway.
• 52:04
  But but just another huge plug for the positive coaching alliance. I think it's really, really, really good stuff.
• 52:08
  Adam Leventhal: Since we're since we're riffing on that, there was one other piece that that I love that actually is applicable, although it's initially not going to sound applicable. In the league that I was coaching in initially, it was like the players were, I think, 9 or 10, something like that. And and the umpires were all these, like, 14 year olds, 15 year olds. And they're like, well, if you disagree with the call and you, like a 6 foot 1, 190 pound man, walk up to this 15 year old, it doesn't matter how polite you're being. It doesn't matter that you're asking if you saw it properly or or talking about the weather.
• 52:43
  Like, the optics from 30 feet out are atrocious. Right. And so the the advice they gave was, you know, stand next to them. You know, imagine you're kind of looking at something off in the distance, you know, like side by side. And and but the thing that I took away from that kind of more more broadly is there's a, you know, regarding the artifact.
• 53:05
  Not like your bug or your code or, you know, kind of looking at it where we're putting our our arms around each other's shoulders and looking at the problem together in way because sometimes people can feel that responsibility, you know, when there is a problem or when there's a screw up or whatever, kind of move like, even if I don't feel like a person is culpable, you know, that person might feel like they're blaming themselves. I mean, I I certainly blame myself. So doing everything you can to, like, really bring that kind of neutral disposition.
• 53:40
  Bryan Cantrill: Yeah. That's really interesting. That that's really interesting. And because it it can, like, it can be hard to step out of yourself on that, either way. I feel.
• 53:52
  Thinking for myself, I I, Matthew's asking in the chat if there's one change to oxide culture that people could make, what would it be? And everyone is, now chiming in. The the chat wars have now begun.
• 54:09
  Adam Leventhal: Oh, I I I know. I know. I know.
• 54:11
  Steve Klabnik: Own it. I'll own it. I said, I don't know if this is culture, but moving away from Matrix. And my beef is not actually with Matrix. It's just that there are no good clients, and so, therefore, I kinda blame I don't know anything about Matrix, the protocol.
• 54:23
  Technically, I don't really care about that. But, like, all of the clients are just bad, and it's it's the least favorite part about my job. I don't know that I would necessarily advocate that we move to something else in Discord. Discord is not necessarily appropriate, actually, but, like, it is the only chat program I actually like using. So, I will I will about suggesting that.
• 54:42
  Adam Leventhal: I will take your millennial and raise you an excerpt, which is I mis emailed desperately. Like, I Oh. Under I understand the chat has taken over, but I feel like chat FOMO is really challenging. Like, every once in a while, you know, I'll be in the middle of some focused work or whatever, and some discussion will happen on chat. And someone will hit me up like, hey.
• 55:03
  You're gonna really care about this thing. And it just kills me that, like, it needed to happen right then. Like, at Wednesday at 2:13 in the afternoon is the time when we decided on chat. This sort of, like, critical thing. And I don't think that this is unique to Oxide.
• 55:18
  I think a lots of folks have moved to, like, a chat first culture. In fact, my brother joined a company a while ago that, like, where email was forbidden. Like, everything happened in chat.
• 55:28
  Bryan Cantrill: Email was forbidden? Wow. Basically.
• 55:32
  Steve Klabnik: Do you want do you want the real cynical response to that? Go. Yeah. Yeah. Email stays around forever, but places like Slack have a retention policy that goes away.
• 55:40
  Adam Leventhal: So if someone's trying to
• 55:41
  Steve Klabnik: do Instagram something that's old, then being chat first is
• 55:45
  Adam Leventhal: A lot of companies actually email doesn't stick around forever. Like my wife was in an organization where they actually, they're a bunch of companies that have like, we destroy email as a policy. Good luck scrolling it away kind of stuff. But fair enough.
• 55:59
  Bryan Cantrill: Yeah. I that's interesting that the, well, first of all, that email, I mean, what I'll see your exer and, like, raise you a lost generation. Let's get back to, like, written letters. Let's just send letters back. If it's so important
• 56:13
  Steve Klabnik: I'm gonna go out for carrier pigeons here. I'm just saying that, like, you know, if we use carrier pigeon, then, like, maybe it's much better. Like, would
• 56:20
  Adam Leventhal: it kill you to send me a singing telegram once in a while?
• 56:23
  Bryan Cantrill: I, you know, I I I I thought, you know, one one was enough. But the, was that what you told me? The the the thing is amazing, Adam, is you go back through our past emails and, like, past lives. Like, from, like, the, you know, like, the 2000. It's like, yeah, we used to write really thoughtful emails.
• 56:39
  Adam Leventhal: Thoughtful, like, long emails Yeah. Where people then could consider it and, like, not have to respond instantaneously. I mean, it was also I mean, whatever. I'm giving a pitch for email. But it's nice also when you get back from vacation, like, there there there are emails you can go through and sorta do your own prioritization to figure out what needs to be attended to.
• 57:00
  As opposed to chat where I'm I'm not reading all that. Like, just write that down to 0 and start from start from scratch and hope that nothing too important happened.
• 57:10
  Bryan Cantrill: So I gotta say, I'm with Emily in the chat, and I I really miss Jabber. We had Jabber, which was exactly that. It was IRC with searchable logs. And we had to, I can't even remember, and if if Josh or Patrick are here, can remember why we had the decommissioned Jabber. But we had the decommissioned Jabber, and, we were using to to matter most, which actually I I actually did like matter most, but, we we had to decommission our Jabber server, so we gave it a, we gave it a Viking funeral, where the the Jabber server was, we had a bot that would rewrite things that the bot received, with using, like, the chicken scheme font, and then there was an error in the bot where the bot began responding to itself, and the whole thing exploded in this like glorious Viking funeral for Jabber.
• 58:04
  I, and then we turned it off. I really miss Jabber. I thought Jabber's Jabber was kinda was kinda great, but, I yeah. Okay. So chat is gonna be the thing that, rips us apart.
• 58:17
  So thank you very much for asking us how we would change our culture. This is how we destroyed Oxide's culture in the great chat civil war of 2024. So thank you very much, Matthew, for, for firing the first shot there. I guess, Cloudnick. I guess that's Steve.
• 58:28
  I guess that's on you. That's not on on Matthew. I just I just immediately. You're
• 58:32
  Steve Klabnik: yeah. You're the one. Yeah. My fault encouraging it.
• 58:35
  Bryan Cantrill: Yeah. Pretty much. Pretty much. But that's it's definitely a good question. I do think, like, one of the important things is that, like, they there is a mutability to culture.
• 58:43
  I mean, this is one of the things I think Cliff is always really good about expressing. Remember when, you know, the much earlier in the lifetime of the company, someone was talking about the hiring process to Cliff, and Cliff's like, yeah, you know, we think it's, like, it's pretty good so far. And I just really liked that, like, so far. Like, we are because we you constantly have to look for places where we can reasonably improve things. And so I think we're we're, except for Chad, apparently, which is
• 59:13
  Adam Leventhal: actually, there's some kernel of wisdom there, which is constantly improving things, except for things like chat actually is divisive. And It's no matter what no matter what we chose, it would not clearly be better to everyone. So there is actually, I think, some wisdom to ignoring some of these problems. We're like, look. People are gonna be grumbled no matter what.
• 59:38
  I'm sorry. That sucks. And there's no clear better thing that we're gonna do, like, clear to everyone. So I don't know. Just toughen it up.
• 59:46
  Like, we we just gotta not poke that bear.
• 59:49
  Bryan Cantrill: Yeah. I do think that there is and I think it's a good challenge because I think that one of the things you wanna avoid is is where you do have things that are device because other things are device. I mean, there are a lot of things that people at Oxide agree on, and there's some things people at Oxide vehemently disagree on, and you kinda wanna just avoid those things, honestly, at especially when they're not, like, exactly germane to what we're building, which means chat kind of isn't. You know, it's important, but it's also I don't know. It's in it's it's in that it's very important to kind of our daily quality of life, but, I I do think it's actually super important to be able to search Chad.
• 1:00:21
  So that's actually one of the things that one of my own concerns, one of the things that drives me bonkers about Slack is that the is that you don't own your own data, and it's like I that data is super important. And, I I mean, many times I have searched chat. I mean, I think, Adam, we really try to avoid making decisions in chat, but we, but the there's load bearing discussions that happen in chat for sure, and being able to search back, is, is really valuable for me.
• 1:00:54
  Adam Leventhal: Yeah, for sure. I mean, it wouldn't be oxen friends 2024 if we didn't anticipate an LLM future. Because it does seem like Yes. Both chat and transcripts of all these recorded meetings. Do there it feels like there is some opportunity for distillation here for helping people, you know, sort and and and figure out what what important things happened.
• 1:01:18
  Bryan Cantrill: Totally. Totally. So, yeah, hopefully, I know again, that would be wouldn't that be nice? That would be feels really nice. If you could have, like actually, I don't need to pay attention to the the the the bot.
• 1:01:31
  My my alter ego is gonna be watching chat for me.
• 1:01:34
  Adam Leventhal: And maybe you can respond for me too. Be like, nah.
• 1:01:36
  Bryan Cantrill: I don't like that. And you gotta keep
• 1:01:38
  Steve Klabnik: it interesting related feature here. They've started rolling out in beta to some places. I've not seen it on this server yet, but, they will send you a push notification that is like, hey. These folks are talking about this topic in this channel right now. And it's very funny how accurate and also not accurate it is basically, like, sort of at the same time.
• 1:02:00
  Like, sometimes it gets the technical, like, semantics correct, but, like, with a description that's, like, a little funny or, like, not how I would describe it or whatever. But it's definitely kind of an interesting, weird like, the first time I remember seeing it, I was like, what is going on right now? But, yeah. Like, there that there's something happening there.
• 1:02:21
  Bryan Cantrill: Yeah. That will be that talk about like, Hey, I think these people are talking about you over here. It just feels like it's like, it feels like middle school gossip. That's just like you yeah. You should know.
• 1:02:33
  They're,
• 1:02:34
  Adam Leventhal: I want the one that says, hey, those people are talking about you, so I'll send an email with the germane details and read it at your leisure.
• 1:02:41
  Bryan Cantrill: Send you an email. There you go. You could this is Yeah.
• 1:02:44
  Steve Klabnik: This could be Maybe
• 1:02:44
  Adam Leventhal: where I am. That's what I'm saying. Yeah.
• 1:02:46
  Bryan Cantrill: This is like the Jitterbug phone that's in, like, a giant phone that only has, like, 4 buttons or whatever. It is constantly advertising itself on the nightly news. This could be there we go. Yeah. I I I just I can I can already see the commercial?
• 1:03:00
  It's It's like another chat message. Why doesn't anyone send me an email anymore? Now they do. What would you call that friendly email bot? You would need to have something that would, like, give, like, a little tip of the hat to the nineties.
• 1:03:13
  It's gonna be something that's gonna need to it's gonna need to be, like, email companion or something like this. I don't know what it's gonna be. Yeah. Resurrect Microsoft Bob. Microsoft Bob, in hindsight, like, maybe ahead of its time.
• 1:03:26
  No? I we if we revisited Bob, is it
• 1:03:28
  Adam Leventhal: time to revisit Bob?
• 1:03:32
  Bryan Cantrill: I I know I know that that that high pitched I know what that means. Yeah. I know I know I know what that means.
• 1:03:38
  Adam Leventhal: Moving right along. Right. Okay.
• 1:03:40
  Bryan Cantrill: Let's I tell you what, I will forget that you ever said that, Brian, if you forget that. I so the, another okay. Another one that is, that is important to me that I think is also, a little unusual, sadly, is that we don't use engineering metrics, which is to say metrics by which we evaluate engineers. I say that with, like, one I've kinda used metrics once in my life, Adam, and I think you know what I'm gonna talk about. I have no idea.
• 1:04:06
  When we, when, when, when we quantitatively show that you as a very junior engineer, we're doing we're doing more work than an entire remote office that we knew was not was just not very engaged. And I feel like the, and but I feel like even that because we made you kind of aware of that metric that we had done this.
• 1:04:27
  Adam Leventhal: Oh, yeah. You know who else was aware of it? The remote office that I then visited. Oh, really? Please continue.
• 1:04:33
  Bryan Cantrill: Yeah. Anyway How did that awkward? Did they hold you No. They all they loved me.
• 1:04:37
  Adam Leventhal: They were so happy to see me.
• 1:04:40
  Bryan Cantrill: I bet. Sorry.
• 1:04:41
  Adam Leventhal: Yeah. No. No. No. Whatever.
• 1:04:43
  Bryan Cantrill: Did I I so I because I mean, part of my I think that, like, you just have to be super careful because people will change their behavior for those metrics. And I've just seen behavior not get changed in the way you want. It's really, and you end up I'd be I I'd not say that it's impossible to do, but, boy, you gotta be super careful about how you I mean, certainly, you look at, like, lines of code or, you know, get pushes or, you know, blah blah blah. It's like every one of those is gonna be really I mean, you can kinda, I guess, view it as maybe telling you something that you already thought you knew, but but really danger dangerous to index more heavily than that. I don't know.
• 1:05:24
  Adam, what do you think?
• 1:05:26
  Adam Leventhal: Totally agreed. I've never been in a culture though, that really embraced any kind of quantitative basis for engineering productivity. But agreed, it feels like, the things that are easy to measure would all produce exactly the wrong behaviors and everything else is really hard to measure and quantify. I've I have not seen it done well.
• 1:05:51
  Bryan Cantrill: I've not seen it done well. And I think it and that it is often done very poorly and, people then are often it it's it's used to be because you have this idea of, like I mean, there is this perception, especially among people who are outside of software. Like, let's just talk about the elephant in the room. There's a perception that software engineers are, like, are lazy. Like, what are these folks doing?
• 1:06:16
  Like, I just don't know. Like, why is our software late? It just feels like these guys are, like, arguing over which chat to use or whatever. It's like, what are they what are they actually doing? You know, they're already and and I think that it's it feels like when metrics are put in place, they are are they're replacing trust or undermining it.
• 1:06:39
  Like, look, it's I'm not gonna trust you. I'm actually gonna measure you. And and goes, there's a real there's just a big danger to that. I think that, like, it it just does under my trust, and, it it doesn't I just I don't know. I I I think it's it's the wrong it takes it takes a culture in the wrong direction.
• 1:06:57
  Adam Leventhal: I'm with you. I I also think that, you know, organizations can sort of lose track of what's important. I think usually that is a failure of leadership, either engineering leadership or company leadership. But if you lose track of a mission, then it's easy for folks kinda seeing the black box of engineering. They're like, well, the stuff I want isn't coming out of there.
• 1:07:20
  And the stuff that is coming out of there isn't really stuff that I want.
• 1:07:23
  Bryan Cantrill: It's just stuff that So
• 1:07:24
  Adam Leventhal: so how like, what are the levers that I have? And and I'm sure it's a very especially as organizations grow, it's it's frustrating. Like, it's frustrating. And, you know, and then the engineers are always talking about technical debt, whatever that is. Sounds like some made up thing or whatever.
• 1:07:41
  Totally. So I get where it comes from, and it does seem like the the wrong response to what I'm sure is a very frustrating situation. And there are lots of folks, lots of snake oil that people wanna sell along these lines. Oh, for sure.
• 1:07:57
  Bryan Cantrill: Yeah. In terms of there there's there are a lot of yeah. I mean okay. So had you heard of Dora Metrics? I mean, and this is stupid.
• 1:08:04
  I'm sure you have.
• 1:08:04
  Adam Leventhal: Dora's like Dora the Explorer Metrics?
• 1:08:07
  Bryan Cantrill: Or the Explorer Metrics, but but all caps Dora. This is like okay, so, e, so DORA metrics are, I would, I think pretty common. This actually makes you feel better that you had heard of it, honestly. Because I was in a conversation where people were like, how are you the only person who's not heard of Dora? Like, I don't know.
• 1:08:28
  Adam Leventhal: Were you like, I know about the OODA loop.
• 1:08:30
  Bryan Cantrill: I know about the OODA loop now, and they're like, OODA loop. Like, no one's talked about the OODA loop for 5 years. Like, did you okay. Don't tell me you just learned about the OODA loop.
• 1:08:39
  Adam Leventhal: No, no, no, no, of course. No, no, no, no, no. Please, Please. No. I've been on
• 1:08:43
  Steve Klabnik: the OODA loop for years. I'm gonna integrate it into all my future decisions now that I know about it.
• 1:08:50
  Bryan Cantrill: No. This is the the dev, DevOps research and assessment is a and this is a
• 1:08:55
  Adam Leventhal: Okay. Thanks. I hate it.
• 1:08:59
  Bryan Cantrill: And so this is, it it it is like trying to use so it uses metrics like frequency of deployments, amount of time between an acceptance and an deployment, how frequently deployments fail, and how long it takes to restore a service or recover from a failure. Those are Dora metrics.
• 1:09:18
  Adam Leventhal: And I mean, those seem those seem like things worth measuring. I don't know.
• 1:09:22
  Bryan Cantrill: They seem like things sort of yeah. Totally. Well, this is like where and this is kinda where it this is like the path to ruin with these metrics where it's like, we do feel plausible and the the mistake is an over indexing them. Like, I mean, yes. Like, I mean, I don't know about frequency frequency of deployments.
• 1:09:40
  I'm like, oh, gotta gotta be careful with that one because
• 1:09:43
  Adam Leventhal: That's some real kinda cargo cult stuff going on. I mean, where where is that really what you need? Is that really what you're talking about? Like, there it can be. It might be.
• 1:09:52
  But it can also be that you are parading around with coconuts on your head to make, like, air
• 1:09:59
  Bryan Cantrill: drops happen. That's right. And, and then, I mean, like, looking at how frequently deployments fail, I think that that again, that's valuable. What are you gonna wire that up to? I mean, it's so it's like, okay.
• 1:10:09
  You're gonna what does it mean if that metric see, what you don't want is to be so enshrined in these metrics, and, god forbid, you start connecting these metrics to things like compensation or things that, like, where people feel like, oh, god. Like, I don't want I I I'm kinda changing the way I think about the problem, not because of of what the metric is trying to measure, but because the metric itself. It's like, that's not that's not good. I mean, for example, like, the how long it takes to recover service or cover from a failure. Like, okay.
• 1:10:39
  That's important. Right? It but that also highlights a and then like the how frequently deployments fail. That also is very important. But the there's a big natural tension there where so if you've got a service and you know that restarting the service will likely ameliorate the immediate problem, but it will also lose state that you need to debug it, what do you do?
• 1:11:05
  And Right.
• 1:11:06
  Adam Leventhal: And you've just told people. You just you just gave people the answer when they when the answer really is, it depends.
• 1:11:13
  Bryan Cantrill: Right. Answer is, it depends. And it's like you if you wanna not see this problem again. So it's like I by by shortening the time it took to restore service, you actually could have increased the frequency with which you actually lose service because you have you are losing the opportunity to debug it. And so, oh, boy, you gotta be super, super, super careful.
• 1:11:38
  The it's it just you can not stop and I'd be curious to know if other people have got, like, no. I've got, like, great experiences with this, and this has been really, really important. And, I I mean, I I I get it. I just think you gotta be super careful. Yeah.
• 1:11:58
  So, what's about the other idiosyncrasies? I'd be curious if other folks have idiosyncrasies in their own organization or their own experience. Steve, I don't know if you've been to me from kind of RFTs
• 1:12:08
  Adam Leventhal: our RFTs are hugely idiosyncratic.
• 1:12:12
  Bryan Cantrill: Yes, they are. Yes, they
• 1:12:14
  Adam Leventhal: are. The the fact that we are what number are we up to now? Like, are are we like at 470 or something like that?
• 1:12:22
  Bryan Cantrill: You gotta be very close to 500. We gotta be knocking on the door. I'm gonna say, like, 494. I'm not looking.
• 1:12:28
  Adam Leventhal: You're always you always nailed us. Like, when you No.
• 1:12:31
  Bryan Cantrill: No. I don't.
• 1:12:32
  Adam Leventhal: Oh, but it was a
• 1:12:33
  Bryan Cantrill: where are we at?
• 1:12:34
  Steve Klabnik: But It the most recent one according to the website, not ones that have been, like, you know, on GitHub or whatever is, 496, 65. 465. 4
• 1:12:43
  Adam Leventhal: yeah.
• 1:12:43
  Bryan Cantrill: Which is 5. Okay. I'm okay. Then you know what? I that that graph has not moved us into the light enough.
• 1:12:49
  Speaking of metrics, our RFP production is down this month. Nice.
• 1:12:54
  Adam Leventhal: Quarter. You never get a bigger bonus, boss.
• 1:12:57
  Bryan Cantrill: Right. There's there's an RFD gap here, folks. I need to gather the team. I have something to tell them.
• 1:13:04
  Adam Leventhal: So these are requests for discussion. So the these, this is kind of our our best and only artifact of, like, here's something I plan to do. Here's a gap. Here's something I wanna change. Here's a technical area where I wanna go deep.
• 1:13:20
  And sometimes it it really is a discussion more than you know, we had one recently that that, Karen put together surveying all the different query languages for different metrics, from different cloud providers. And actually, Brian, that led to the work that you referenced earlier, Ben's work on our own query language. So you're gonna have RFPs on a variety of different topics, some of which really come to some conclusion, some of which are a little more open ended and informational. But, man, we have a lot of them. I think that when, when folks join, there isn't a fairly insurmount, like you can't read them all.
• 1:14:01
  We are both have a volume of words. I don't know what it is, but it's gotta be in the several millions, maybe tens of millions, volume of words and rate of production that you can't ever catch up to.
• 1:14:13
  Bryan Cantrill: Yeah. And I think that the all true, the, the RFD site, which is Steve Glabick's putting up the chat, has been, open sourced relatively recently. So Ben Leonard has done terrific work on the RFD site and along with, Augustus and Crespo, a bunch of folks who've done really great work on it. The, the the RFT site has allowed us to I mean, god, just being able to search RFTs. I don't remember when we couldn't search RFTs.
• 1:14:36
  Steve Klabnik: It was really bad.
• 1:14:38
  Bryan Cantrill: Yeah. And, actually, the well, this is, like, where I want I I mean, talk about LOM help. You know, I actually tried relatively early on in the company to make a glossary of terms in RFPs, and I just collapsed under the weight of it all, because there's just so much that you need a glossary for. So I think that, but that has been really, having that's been really great. And what we wanted feels like there should be another tool that does this, namely want a Google Docs kind of front end on a Git back end using, like, an Ask Me doc as the medium.
• 1:15:13
  Yep. Right? I mean, that's that's kinda what we want, and then we use we we do use GitHub for the discussions, which has all the issues of GitHub. The discussion. Medium, which is good.
• 1:15:24
  I mean, there's some there's some things about it that are really good, but it's been, but, that's been really, mea. RFTs that's true. That is that is an idiosyncrasy, and that actually we brought over from Joyant, but it when I adopted what's that?
• 1:15:42
  Steve Klabnik: You gotta say what it is for the recording. Nobody's gonna know what Sean said in chat if you just answer.
• 1:15:47
  Bryan Cantrill: What did Sean say in in, chat?
• 1:15:50
  Steve Klabnik: I thought you responded to thing. I made a dumb assumption. Sorry. Sean said something really good about RFTs, and it sounded like that's what you're about to talk about. And so I was like,
• 1:15:57
  Bryan Cantrill: oh, no.
• 1:15:57
  Steve Klabnik: You're not saying what it is.
• 1:15:59
  Bryan Cantrill: No. I'm just saying no. No. No. Go ahead.
• 1:16:00
  Yeah. Yeah. Go ahead, Steve.
• 1:16:01
  Steve Klabnik: So I was just saying that, like, it's nice coming from a big company that, like, RFPs. Like, you can write an RFP and someone else can implement it, or you can implement an RFP someone else wrote. Like, there's not the level of assumption that, like, this is purely for you yourself and that, like, there's a sort of decoupled ownership model of the work.
• 1:16:18
  Bryan Cantrill: Yes. Yeah. It's all it is very cooperative. Yeah. The the like, you're not writing an RFP as Sean was saying in the chat.
• 1:16:26
  You're not writing an RFP to, like, look good in a perf process or what have you. Yeah. That's right. Well, no. Because I think it's actually really important because I think that that aspect of it oh, gosh.
• 1:16:34
  Sean, I'm glad that you're identifying that, and and Steve, thank you for for reading Sean aloud. The, because that aspect of it is really important because it allows someone to weigh in with their someone has got the ability to go understand those thoughts, and, of course, like, understanding, like, why we've done certain things or what have you. And we're getting more and more of our own RFTs out there publicly. I'd like to get a lot of them out there publicly, but, we've got, some of them are I think we'll we'll at least some of them have remain private because they have they represent material that's under NDA with various vendors, but I think a lot of them could be public. Certainly, our hiring process is one that's out there.
• 1:17:17
  Actually, I loved the I don't know if you saw, Matthew. You got a great, post. Matthew, I'm not sure if that was to if that was Mastodon or a blog entry I saw. I guess it was a blog entry on I had a great post on observability. I don't know if you saw this No.
• 1:17:30
  Adam. But he Matthew ran across rfd 68. He's like, oh, wow. That is actually really helpful, which rfd 68 is an RFP that describes kind of how we think about, partnership and how we kind of select a a partner and differentiating partners from vendors. And that's been really important for us.
• 1:17:50
  That was an early one that we wanted to get out there because we send that RFD to partners, and be like, this is what we're looking for. And what we found is that, like, that just the presence of that RFP, and I don't know if you experienced this, but certainly early on, just the presence of that RFD and that document and sending it to people was an interesting kind of litmus test because the people that we really wanted to partner with love that document. They're like, oh, this is great. This is exactly the way this is putting words to things that we we have thought internally at, you know, Samtech, or we have thought internally at, you know, kind of pick your your your partner. And you know someone is like, I have sent it to others, perhaps multiple times, and perhaps not always under the best of circumstances.
• 1:18:42
  And, knows how much of the docs getting read. Yeah. You know, that's helping to inform. It's like, you know, maybe this is not maybe that this marriage is not gonna last or this particular part or what have you because, we don't and Matthew dropped the post into the chat about, like and so Matthew was using that as kind of a lens for observant observability companies that he was looking at, which I thought was really interesting. And, I, you know, to me, it's great to be able to put that stuff, 1, to be able to write it for ourselves, but then to be able to share that thinking with others.
• 1:19:17
  And, Matthew, for you to pick that up and use that as your own kind of rubric, I thought it was great. And so I think about Adam, I think it oh, it's a bunch of aspects of the of the RFP process that have been really terrific.
• 1:19:30
  Adam Leventhal: Now, of course, there's also the elephant in the room and you always ask me which elephant. In this case, like, you know what's weird that we do as a company? This.
• 1:19:39
  Bryan Cantrill: This is weird.
• 1:19:40
  Adam Leventhal: We that we do this podcast and that we get people
• 1:19:43
  Bryan Cantrill: You're doing you're doing a weird thing right now. Right. You're do yeah.
• 1:19:47
  Adam Leventhal: I agree. That we that we get folks from the team often on the back of a demo Friday. Could folks I mean, I'm I'm kind of pulling back the curtains. Often, we see a great demo. And then, you know, Brian are ding DM ing each other saying on chat, which instead of email, I don't know why.
• 1:20:02
  And, saying we gotta get them on the show. And we, you know, folks, you know, from the storage team we're on re recently, Helios was on earlier this year. Getting folks not just to talk about their work internally, but talk about it publicly and talk about it in this forum, talking about all all of you. And I think for a lot of folks, it's both terrify terrifying and motivating.
• 1:20:26
  Bryan Cantrill: Yes. And I think that, like and obviously, like, we don't wanna push everyone into being an overshare and it's totally fine if people aren't comfortable with it. That's great. Like if not, but for, I think it's been great to also give a vector for those folks that are interested in it. And then I think what's been great is some of those folks have then been picked up.
• 1:20:46
  I don't know if you saw did you just to Nathaniel's interview on Microarch Club. Yeah.
• 1:20:51
  Adam Leventhal: Yeah. Yeah.
• 1:20:52
  Bryan Cantrill: I thought it was really neat that that that, you know, Daniel had kind of listened to Nathaniel here on Oxide and Friends and then asked him to be on his own podcast, which is really horrific. And we and by the way, if other companies do this, like, again, I I don't think we've got, we've we other companies can take whatever of our weirdnesses they want, but if any of the company has a podcast like this, please point us to it because I I would love to listen to them. I think they're really like, it's interesting to get, like, an inside view into, you know, kind of a way a company thinks. So, yeah, you added you're right. That is the offer of the room.
• 1:21:24
  Adam Leventhal: Totally. But and also to anyone who's thinking about this and they're like, well, you know, we could do it, but we'd be talking to nobody. We sort of were talking to nobody. Like, you go back and look at the podcast stats. Like, we, you know, we get numbers in a day that, like, used to take a month and a half for us to get.
• 1:21:41
  So I I would just say for a long time, we were just talking to ourselves. And I think that there was there's still tremendous amount of value in that. Oh, for sure. Especially if you're recording it.
• 1:21:51
  Bryan Cantrill: Yeah. Yeah.
• 1:21:51
  Adam Leventhal: Yeah. That's right. So like, I I think not getting not getting wrapped around the axle about, like, how big the audience is. And certainly, we don't track any metrics about we don't have any goals for the show. We're not advertising on the thing.
• 1:22:04
  You know, we're we're doing it because it's a part of our culture. Because it's the thing that we do and because it's fun and because it's interesting. Because we learn stuff. And and actually, every once in a while, some some prospect does stumble in, and that's great. But, but, yeah, it's just it's just a thing we do and become part of our culture.
• 1:22:21
  Bryan Cantrill: It is. And I think it I mean, it's also true that people are saying this in the chat. Like, hey, this also reflects your culture too. That you it reflects the transparency that a a lot of companies, people will be nervous about what to say on the podcast, what not to say, and the fact that we're, like, no. I mean, clearly, you can say whatever you want because obviously we are.
• 1:22:38
  We're just, like, I think our this is where our lack of intro music is actually helpful, Adam. And just like, you know, you you you jokers clearly are just just backing on, so, keeps it casual. But, yeah, no. It's been it's been really terrific, and and it's been, whether it has been reflects our culture, drives it, or what have you, it's been it's been a lot of fun. As folks say into the chat, like, you even if it's not getting customers, it's getting folks that are, that are that are defending us on various forums.
• 1:23:07
  Thank you very much, the the for you, those of you on on any of the good fight on hacker is a red or wherever
• 1:23:14
  Steve Klabnik: in the comment minds toiling away,
• 1:23:17
  Bryan Cantrill: toiling away in the comment bites. Awesome. Well, I so are there any other we we hit the elephant in the room. I'm glad that we did. Are there other, any other elephants, Adam?
• 1:23:30
  There were we this is a pretty good tour of idiosyncrasies though. I think we've we've had on a lot of them. A lot of the
• 1:23:36
  Adam Leventhal: Yeah. I'll I'll I'll just maybe just a very small one. But, you know, compensation is is even we we talked about that, but, benefits are great. Like, health care benefits are great. And, it's it's, I don't know.
• 1:23:52
  It's something that, like, especially folks with families really appreciate that, that oxide, like, it it doesn't just say, hey. We're we're friendly to folks of diverse backgrounds or folks with families or whatever. Money where your mouth is. When people get apprehensive about joining a start up and we say, well, look, we're gonna health care for all your kids. So we must we must be serious about this aspect of it.
• 1:24:17
  You know, health health care benefits for folks with different kinds of needs. You must be serious about it. I've got a Yeah.
• 1:24:23
  Steve Klabnik: Absolutely. Doubt on that actually for recently. I I have gotten a therapist this year. It's been great. I selected her based purely on selecting.
• 1:24:30
  I was like, I'm not gonna worry. It's gonna pay out for out of pocket, and I'm gonna figure it out later. And, even though I'm in Texas, like, turns out my therapist, 100% covered by insurance. So, like, that was not even, like, you know, and that's often one thing that's, like, really hard to get people to, like, have coverage of and, like, all that kind of stuff. And so just just a random, like, oh, yeah.
• 1:24:49
  Benefits are really, really good, actually.
• 1:24:51
  Bryan Cantrill: Yeah. I okay. So I'm embarrassed that people are saying rock and stone in the chat, and I'm having to ask chat g p t what it means. What is rock and stone? What is Steve, what is rock and stone?
• 1:25:03
  The yes, yes, no. So there is a rock and stone.
• 1:25:05
  Steve Klabnik: Rock galactic that I have played the tutorial of and basically no further. Although, there is a new little alternate universe game that I've been enjoying. But my understanding is that that's, like so it's about mining your dwarves in space mining stuff, and I believe that's, like, a slogan that the dwarves say. That's the only context I know. I don't know anything deeper about it than that.
• 1:25:23
  Bryan Cantrill: Apparently, like, everyone in the planet knows this thing because everyone is I that there was no don't don't you dare abandon me, Adam. Don't you dare. No. No.
• 1:25:33
  Adam Leventhal: I was just as lost. I was just as lost.
• 1:25:36
  Bryan Cantrill: Thank you. Thank you. It's like, that's what I was trying to tell him. It's about the dwarf mining game. Brian, everybody knows that.
• 1:25:44
  Adam Leventhal: Come on, loser. Get in. We're going dwarf mining.
• 1:25:46
  Bryan Cantrill: Go in dwarf mining. Well, indeed, rock and stone, indeed. I is that am I using that correctly? Rock and stone half? Is this rock and stone?
• 1:26:01
  This is where my kids usually beg for me to stop, so I'll stop. Well, I this has been a fun discussion. And thank you for indulging us as always. Thank you to Matt and to Cliff for kicking this off, with, again, 2 terrific blog entries, that were really inspiring to read. And we've got some, don't know.
• 1:26:21
  I'm I'm not gonna announce anything now, but we've got some exciting shows in the works. So that's all I'm gonna say. I'm just gonna tease it a little bit.
• 1:26:27
  Adam Leventhal: We have someone light up for, next week who is thinking about backing out, but we're gonna really lean on him and have a great show next week.
• 1:26:36
  Bryan Cantrill: Yes. It's gonna be fun. So Yeah. With that and, how life works again, have you Adam, how are you how are you are we gonna do our our daily checkup our our weekly checkups too much? Oh, yeah.
• 1:26:47
  Adam Leventhal: I started. It's it is it is delightful already. We gotta we gotta set a date. We gotta set a date. So
• 1:26:53
  Bryan Cantrill: Gotta set a date. It's gonna be in May.
• 1:26:54
  Adam Leventhal: We'll set a date.
• 1:26:55
  Bryan Cantrill: We'll set a date this week. I promise. So
• 1:26:57
  Adam Leventhal: But it it's been great. It it's been a great recommendation.
• 1:27:00
  Bryan Cantrill: Awesome. So How Life Works by Philip Ball. So go ahead and take an excuse to read that, and join us for a discussion in May. It's gonna be fun. Alrighty.
• 1:27:08
  I guess book clubs and podcasts also in idiosyncrasies, so let's not forget that one. So alright. Thanks, everyone. Talk to you next time.