Whitney Lee: So Coté, you were an analyst, right? Coté: Yes. Twice. I loved it so much. Whitney Lee: Yeah. I can't tell if you're being sarcastic with that tone. Um, there's so much I don't understand about that world and I thought I could research it or I could ask you and learn out loud. So here's some. There's so much I don't know, but here's what I can infer. Like if you, there must be some sort of data element in some sort of opinion elements, because if it was pure data, they wouldn't need you. Right. So, yeah. So then, um, I guess I'm interested in that balance, but I'm also interested in like, once you have an opinion, you are, um, you're. For or against a technology, or you're saying it's performing well or not, but like, there is a company with money at stake behind your opinion. And that company is probably going to be pretty disagreeable and, and, and challenge you a lot as an analyst. Coté: Yes. Yeah. Whitney Lee: Do I have the right impression? Coté: Uh, yes. And I, and I, and I think you've identified the, um, how would I characterize it? You've identified the most interesting. Invisible part of the analyst world for people who are not in it, Whitney Lee: because, okay, you know, like, Coté: like, because there's always lots of commentary about like, Oh, unless I was placed well on this chart, uh, I believe that they were paid to say this, so I don't believe it unless I win, in which case it's totally unbiased. And so like, I think, I think the whole, like to use the old term, the whole idea of, you know, pay for play or payola, like, um, I don't know. Well, once you're in the analyst world, you're like, boy, I wish that would be awesome. Right? Like if only it were that easy, then I would be doing like a quarterly chart of ranking vendors with some slight variation between them. And then I could, uh, but I think, I think, I think, I think as, as our, uh, as our, as our guests this week, we'll attest to there's no retiring early. If you're an analyst and, uh, why don't you introduce yourself guest? Rachel Stephens: Hello. Hi, I'm Rachel Stevens. I'm with RedMonk and delighted to be here. Thank you for having me. Coté: Yes, of course. Whitney Lee: What are your thoughts about qualitative versus quantitative and also whether you're in the war zone, like I think you are. Rachel Stephens: Yeah, so RedMonk is a developer focused analyst firm. So really spending a lot of time on just that practitioner and people with the hands on the keyboard view of the world. And there are some senses in which that view of the world does not lend itself well to the developer. Pay for play because nobody pays for anything in the developer tools world. I would say for us, the way we think about it is like the data driven research, if we can find it is great. Otherwise, kind of augmenting it with qualitative conversations, triangulating what we're hearing and reading and putting just kind of trying to put puzzle pieces together for people. And the business model is, I guess there's a few ways to describe it, but it's, we research so that people are willing to consult with us. And the reason people are willing to consult with us is because we've researched and it just kind of feeds each other. So I would, I'd say my job is kind of equal parts research and consulting. Whitney Lee: So, so who pays you and what do they want? in return for their money. This is fair. Rachel Stephens: Um, so the other pippy way you can try to describe it is you can talk to me for free, but you have to pay to hear me talk. So I want all of the inputs. I want to hear everybody in the industry tell me what are you working on? What are you seeing? What has the landscape? Like I want all of the inputs. And then the consulting part is me kind of doing that synthesis back out. For other people, it's just like, here, here's what I'm seeing. Here's how it's all coming together. Here's areas that we are seeing traction in things like that. Yeah. When Coté: I was trying to be a snarky little shit back when I was an analyst, I would tell Whitney Lee: people, at Coté: least in this instance, my, my, uh, snarky little shit then, but I'm not so much now is, is, as you say, I would tell people like, Oh, I get paid to talk. Like, you know, like I, so, um, I don't want to talk right now. Whitney Lee: So what's a, can you give me a concrete example of something you've worked on? Rachel Stephens: Yeah. So I feel like I've spent a lot of time recently kind of in the open source license world. Not so like one of my colleagues is Steve O'Grady and like, not in the Steve O'Grady way where I'm. Like a crusadian for open source in social media. Going to leave that to others, but for me, it's, how are we seeing impacts of license changes, um, in the wild? Um, which is really tricky because a lot of the license changes that happen, happen in proprietary, like, um, companies that are not yet public companies and don't have to release their financials. So it's a really small view of the world. Like, what are we seeing? And so it's like one of those things, sorry, it's like, can I take a little chunk of data? Combine it with some commentary about what we're seeing overall. And then letting people, like, I think one of the things about analysts, and I'm sure Coté can attest to this, is like that piece of research in particular was the biggest Rorschach test I've ever, ever published in terms of like, either, either you imprinted your own confirmation bias on it, or you like imprinted your own outrage and like same, same data, but like interpreted wildly different ways, because all the data is incomplete. And so like, you can see, You can take it either way you want, but it's interesting. Whitney Lee: What's, what, how do you avoid your own confirmation bias? Do you have a certain mindset or anything that helps you do that? Rachel Stephens: I think it's being willing to be wrong, I think is a really important part of the analyst job. Or like being willing to learn. Both of those things go hand in hand. But is it something where I have, I think it's also just like how you survive in social media in 24, 24. It's like, if something is evoking a strong feeling, you should probably check that. So those, those kinds of things are just like, what, what is, what does my gut say? Like, is my gut kind of, am I, am I leaning too much into that? Or do I need more data? Things like that. It's an art, not a science. Whitney Lee: That's, that sounds like Enjoyable, honestly. And, and can I say something? I love being wrong. Like it's a delight to be like, Oh, something I didn't think of. It's like a whole new way of thinking is now available to me. That's exciting. There's so much to explore there as opposed to anyway, I enjoy it. I've discovered I'd enjoy it. Just like I used to be a picky eater and now I love eating food. I don't like exactly because it's a new experience, which is more valuable than the discomfort of eating something you don't like. I should Coté: have you talk to my kids. Whitney Lee: My tip. I don't know if I can help with how young, how young are your kids, Rachel? Or old? I guess three, seven and three. Rachel Stephens: Yeah. My little one is going on a school bus for the first time today. And he is. Just over the moon. Like he's going on his first field trip and he has to go on a bus. And like, we talked about school buses all weekend. So I'll report back next time I come on the school bus situation. Whitney Lee: So at the age where you can be delighted by a school bus, you're not going to go with the logic of like a new experience is more valuable than an uncomfortable experience. This is, Rachel Stephens: I mean, like, he should be, like, he's delighted by the newness of a school bus. Like, ideally, we could be delighted by, like, chicken nuggets that are not shaped like dinosaurs. Like, maybe we could eat all variety of chicken Whitney Lee: nuggets. Like, I don't know. That's so funny. So what have you learned about open source? Rachel Stephens: Oh gosh. The license changes. I want to hear Coté's take on this, but I feel like one of the things that has come up for me a lot lately is like the people on the sidelines, which includes me, I feel are like the people shouting from the sidelines. It's like, open source is not a business model. But then I think we stopped the sentence. I think we don't necessarily spend enough time examining like how hard it is to make software business and how hard it is to make these open core businesses work. And so, like, I think the one that was really interesting that came out recently was the, um, article from buoyant. Came out Rachel Stephens: last week and they're talking about how they kind of, he didn't do like a rug pull on the license, but they changed the way that they were distributing their software. So they didn't have free supported versions of the software earlier this year and they're now profitable. And it really upset a lot of people for a very long time, but it's also one of those things like, well, how else do you want me to make money? And so I think we're grappling with that in some really interesting ways in this industry and people are taking. All kinds of takes in terms of, I mean, it truly is a full spectrum of people who are religiously opposed to any kind of not open source license. Like the way over here is and the people who are, and there's this whole spectrum in between. And I don't think we have good nuance to talk about how to make. The business side work at different points on that spectrum. Coté: Yeah. Yeah. I guess, what, what, what do I think of that? Uh, speaking of having a, uh, go getting, going out there and having an opinion. Right. Like, I mean, I think let's see, so from the bottoms up, like I think that. Like Whitney and I talked about this a lot recently, like I think it's really hard, not impossible, but very difficult to do like infrastructure and programming software if you don't have open source as part of the way that you're doing it. And I think primarily it's because You're going to not release your software for several years and like get no feedback. Right. Whereas like, if you're doing it in the open from the beginning, you like are getting all of that feedback while you're working on it. And, and then go ahead. And Whitney Lee: it's a matter of everything, integrating with everything else. If it's out in the open, people will build their own integrations of the software. Coté: That's a great point. And like, like even, well, you, you too, and I know Whitney, you probably know this better than I would, but like, I feel like in the kubernetes world. There was this weird couple of years where service mesh meant nothing and everything. And like, I think it wasn't like integrated into the open source world enough. I mean, even if it was open source, but the point being that like the various things running around didn't like latch on to officialdom and then once it did. Like it was fine. Things got figured out or not, but maybe that's a bad example. I don't know, historically, but like, so that's the first problem. And then, and then, I don't know, the second thing is like, I think, I think we, as a community, I mean, you were, you were, you were kind of doing it, Rachel, like we need to have a stronger stand on like, there's no open source business model, like, like just like, like in, in the sense of like, these two things exist, but there's no virtuous cycle between them. Right. Like there's no, like, there's no connection between the two of them that really like matters. And so therefore, like, you can get upset about rug pull stuff because like, you change the norms of a relationship that you had, but it's not like gravity where it's like wrong. It's just like, Oh, sorry about that. I don't know. Like those, those are my, my jumbled up thoughts, but yeah, I think there used to be open source business models, but now like, I like the way you put it. It's just like doing anything in software is really difficult. Nevermind the way you do it. Whitney Lee: From my, you know, new to tech perspective, it seems like that the open source business models were successful to get VC funding, but we're, we're not able to turn a profit if they were based purely on open source. Is that true? Is there an exception to that? Do you know of a business that turns a profit off of a pure open source product? Rachel Stephens: Off of a pure open source product? I mean, everything is, like, if you're in that open source world, it's It's going to usually be open core. So you're upselling something or you're doing a services based model. Something like that. Who's doing that well? Who's doing that well? Oh, that's a really good question. I mean, like, who is doing that well? What is Coté: GitLab's deal? Do they have closed source that they sell? Rachel Stephens: GitLab has, so they're doing like end to end, like entire SDLC plays, and then they have an enterprise version. That's, um, like all, all the fun things enterprise want, like the Rbacks and Naudits and all those fun things. But I think for me. Who's doing that? Well, like everyone's going to say Red Hat, that's the classic. Um, but I think that it's going to get me in trouble. Like they've definitely, um, done, they have made some choices in the open source ecosystems in recent years that have, have made, made some concerns. But like, if you look at IBM's financials, like Red Hat is definitely the bright star of where we are. In terms of finances, finances, they're doing well in terms of like this endos community, maybe they're doing less well in terms of how that community is feeling. So it's like, it's just a balancing act in terms of how people are thinking about what is their value? What is it that you have to prioritize? And Coté: I think they're, they're a good example. I, I should, I should know better how they combine their stuff together. Although now that I'm not an analyst, maybe I shouldn't, uh, but especially to comment on them as someone who works at like a competitor. But I think, I think there's another distinction in open source that's always existed, but maybe is where the business side comes in now of like. You could call it like for free, you can get a pile of stuff. And if you want to pay money, you can get something that's ready to work. Right. And so like, I, that's probably the distinction between with GitLab and, and Red Hat, where like, you can get the stuff, but if you just want to run it, like, then you need to pay for it. I don't know, something like that. I don't know. I think open source. It's like, if Rachel Stephens: you want, if you want integrated experience. If you want all of the enterprise stuff, like, like I want it on prem to run really well, I want, I want kind of SLAs. I want all of the audit ability stuff. Like all of those things are usually where the security things like up charges come in all of those places, but also just like the, that's where managed services come in as well. Like, I just don't want to deal with any of this at all. I want someone else to run it for me. Whitney Lee: And, and does the business model of when there's just one open source technology, like solo that that's built on Istio or buoyant on Linkerd, of course decided that it wasn't enough. And, or what's another example? Nermada on Kyverno or Acuity on Argo CD. When there's just one open source project under it, do you, have you seen success with that model? Rachel Stephens: Are we talking like single vendor open? So it's like one vendor is backing one project and that like, it's not like that whole, it's like, I wouldn't necessarily put Linkerd in that category because it is CNCF backed and they kind of have, they, they, they have met the CNCF standards of having like multiple vendor support for the projects, but it's definitely like, I think Hashi would be a good example there of being there as kind of the one project. Community is really behind it, but like it's really centralized in terms of how the project is used, who has control, and those, and it is de facto like Terraform is Hashis. In as much as it is communities, it is Hashis. And so, I think that you can, you can have rightful skepticism. I think right now there are definitely companies that we talk to who are treating single vendor backed projects, um, in at least a similar category as proprietary. Because we've had enough red polls now that people are increasingly skeptical of whether or not there's actually a sustainable community behind some of these projects. Whitney Lee: So speaking of HashiCorp, like their decision then to to close source Terraform, have you seen that work for them? Like in the, now that we're a few months past, how's it, how's it going? Rachel Stephens: So we, that, that was the financial piece that I mentioned earlier. Um, and they were one of the companies that actually had public financials that I could look at. And so, so the, the net of that research is. It's really mixed bag in terms of where people's at. Everyone's revenue continued to climb, but usually climbed about a pace from where they were pre license change. Like there wasn't a significant like uptick in the rate of revenue um, absorption. Um, but the net income all over the place, or I mean net income's all over. Like I just, I was really, really shocked that like Mongo you see, like Mongo was one of the companies and they had huge growth. You look at Hashi and they kind of have just been in a steady decline since IPO. So there really was like absolutely no correlation between companies that made this change and what we actually saw in the numbers. So it's, I guess, the verdict is still out. And I definitely had a lot of people reach out and were grumpy with me because like my sample size was really small because I had to go with publicly available financial data. But when I, when I did this, like, this is what saved my company. It's like, I don't doubt that, but also I don't have that metric. I can't include your data if your data is not available. And so like, it's definitely an incomplete story, but it's also not, um, it's not a clean story. And. Whitney Lee: Something you said earlier, too, is that there are different degrees to which people can close their open source projects. So we talked about the buoyant level where they just stopped releasing supported artifacts versus the HashiCorp level where they straight up closed it. What else is there? What other gray? Rachel Stephens: What else can you do in between those two? If you think of Redis, like first they closed their modules and then they closed the core. You can definitely do things like that. I think you can move to more restrictive licenses. So like, not necessarily a rug pull, but, Grafana might be one of the examples there. They're still open source, but it's a more restrictive license than it was. Um, So yeah, I would say there's definitely ways that people can do it. Whitney Lee: Do you think the software community is getting more tolerant of companies doing this? Are they growing more understanding that companies do need to make money? Or are people mad every time? Rachel Stephens: I wouldn't say tolerant, but I would say Like, jaded and numb. Yeah, Coté: yeah. Like I, I, I think, I think that's accurate. Like it's, it's almost normal now. Like, like the, that it happens. It's not like so shocking, like, you know, uh, other drama comes up in other strange, sad, bizarre ways nowadays. Uh, but, um, I think, yeah, I think, I think maybe in like three or so years or maybe even less, like it'll not be that big of a deal, like, I'm sure there'll always be drama of like, this was supposed to be like, Purely open source and something crazy happened. Like, you know, like when, uh, I don't know if some CNCF project got taken over by a vendor and completely steered some way that everyone objected to, it'd be like, Oh, there's a hole in the governance, right? Like, or something like that. Um, but, um, you know, that you are making me think of something else is like, what's, what's, what's your sense for pricing in all of this, where there probably is some price that becomes too expensive. And then some like, I don't know, forking or closing happens. Like, like there must be some equilibrium where like, we can do stuff in open source. If you'll pay the prices that make it sustainable. But then on the buy side, they're like, I don't want to pay your prices. And then, so on the sell side, you have to be like, how can I make you pay my prices? Right. And like, that's, that's where this, I don't know what kind of game that is. Uh, but like, there must be some equilibrium or something where like you, the two prisoners rat each other out instead of playing along and then. And then the, and then it goes haywire. And it's probably based on price. I would imagine. Rachel Stephens: There's a couple of things that, that calls to mind for me. One, I'm going to reference that buoyant piece that Lilia Morgan released recently again. And he talks like my company, like buoyant is now profitable. And I have been able to employ more linker deep maintainers because of this. It's like, there is part of that where. By making a different decision in and around open, he felt more financially able to support open, which I think is maybe not intuitive for a lot of people when you're thinking about open source values. So I thought that was really interesting. And then the other one I thought was interesting is, I'm not remembering the name right now, but the Sentry campaign about, like, the CFO, like, chief, or But it's basically like that graphic with like the money is like, we all need to pay the maintainers. We need to make like this open source, you know, and I don't believe Sentry is open source, but I think that it's one of those ones where I, um, We start to see some pushback in the, there, there needs to be, even if it is not necessarily me offering all of my products as open source products, I need to be able to give back to that ecosystem in a way that is sustainable. Coté: And I think Rachel Stephens: hopefully we can come up with some kind of answers there. Coté: Yeah, that, that is like the, um. The, the, the, the snarky word for this sort of thing is the freeloaders, right? Which I think, I think was, was that, was that what Red Hat invoked when they did all their CentOS stuff? I forget. Or like, that's when I first remember that word. And like, Rachel Stephens: yeah, Coté: and, and so that was like, But that is the other, the other side of the moral boat in all of this is like, yeah, freeloading does suck, right? Like, so there is like, there is, uh, you know, if, if, if you take this, uh, an open source community software and just totally profit off of it without giving something back, well, well, for one, one side, like if you are a, let's, let's say you're not a vendor, you're an enterprise and you're I think you're free to freeload, basically, like, like no one really gets upset if you take the open source stuff and don't pay back into the community, more or less. It's really only if you're trying to, uh, to sell something in that area that people get upset, which is another interesting moral dimension that like it's only people only get upset. If there's a competitor using the code, but if you're not a competitor, no problem, like not, not, not a big deal. Rachel Stephens: Cause if you're not a competitor, you're just the total addressable market. Coté: Yeah. I mean, I think, I think, I think you kind of set it up well at the beginning of just like, we used to have these other norms. And now we don't, and like, it's, I think it's taking everyone a long time to like readjust to it. Like the, the, the 2000 ways of thinking about this are definitely not the 2020 ways of thinking about it. It's just totally Whitney Lee: different. Oh, one thing I'm wondering is so far, all the examples I know about are, are like. HashiCorp and Terraform. Redis and Valki. Or not Valki, you know, Redis. Red. Red. But they're all Redis and Redis. . But they're all backed by, or, or Buoyance and Linky, like they're all backed by one company. Like do you think it's possible to do the, these license changing changes when, um, there's a open source project that's being maintained across a lot of companies? Rachel Stephens: Maybe it's feasible, but it would be really hard because the, um, the governance structure in theory would be there to prevent it. So like, maybe, but I, I haven't seen that. I mean, I guess like my SQL, maybe it would be like the example I could think of. Um, but Whitney Lee: I do think Buoyant's solution of. No longer open sourcing stable releases, but keeping all of the code still open source. I think that's a, an elegant solution. That's my opinion. What do you think? Rachel Stephens: I don't have a problem with it. And for me, it's, they have, I think one of the things that we watched in kind of the CNCF land was. What does it mean to have a graduated project whose primary backer is doing something different? And so like watching the CNCF kind of coalesce around the project, I think was also really interesting to watch, watching linkerd kind of with their recent announcement coming out and like how they're thinking about maintainership. Like, I think it's been a successful and interesting thing to watch, like both the foundations that are in place to try to help and also the company behind it and the project overall, like it feels like everybody coalesced in some interesting and healthy ways. Whitney Lee: What did you learn from the CNCF through the process about the CNCF? Rachel Stephens: I think it's, I think it was just interesting to see other maintainers come into the fold to audibly tell customers, like, we support this project. This is not going to be an abandoned project. This is still going to be available as a community and supported as a community project. And we saw that come together in a way that, um, I think provided reassurance, um, in a time where like any kind of change to open source could feel. Um, destabilizing. Whitney Lee: Is your job of an analyst, you just read news articles all day long? How do you keep it in your brain? What's your Whitney Lee: strategy? Rachel Stephens: What's your input strategy? I, so I read a lot and I take notes while I read and I have, and it's like not an elegant note solution. I use like simple note. And it's like text only. And I just like, all of my stuff just goes into one spot and I search it. And like, eventually I'm like, I should put all of my notes into like an LLM and like, have like my own version of like Rachel Chachibichi, but I don't do that. It's just, it's me, it's me into like simple text search. Whitney Lee: So, so I'm relatively new to tech. And, and AI has been such a disrupting force. Like everyone was chugging along, talking about infrastructure and platform engineering stuff I loved and still do love. And then all of a sudden everything's about AI and I don't like it one bit, but I can feel, but that's the point. The point is I'm sure that that disrupting technologies have happened before and will happen again. And so, um, I guess I'm interested in. When has it happened before? Because I wasn't around for it. What does it feel like? And will this be over soon? Rachel Stephens: My personal opinion, like, I don't think this is web three. Like, I don't think it's going to be like hot for 18 months and then just die. Um, I do think that there are legs on AI, um, primarily just because once you personally have that experience. This is increased productivity for me personally, like, even if it's questionable whether or not it has increased team productivity, which I think is very much incent. If you look at like, how are we going to maintain all of this code? Like, I think there's still really big SDLC questions in and around AI. But I think that. As an individual user, like, it's really kind of magical. Like I don't write my own Python scripts anymore and it's delightful. I love it. Like, I love, I love not like writing my own API queries. It's delightful, but I, um, I'm just me. And I'm like, I'm not pushing my scripts to production. Like, I don't have to do anything to maintain it, but like, I would be so sad if I went away just because it makes, makes my life easier. And usually. Things that can make a broad swath of the population's lives easier. Like it's hard to put that genie back in the bottle. Coté: So, so what do you, what do you use it for? Go, go over a, I don't know, what, what are a few things that, that, uh, whether it's obvious or interesting, what, what are some things you do? Rachel Stephens: Yeah. So like previous, so like I code, I'm going to put that in like really, really strong air quotes, um, in that I use R to make all of my charts. I'm publishing things and Rachel Stephens: I. I can do it, but like, there are things where I'll get in there and be like, oh, like, how in the fuck do I like label the legend alphabetically instead of in the order it is? Like all those, like those, just like those little things that, and before I just had like a giant, like Google doc file where I would put like my own code snippets so I could like search things. Coté: Oh, right, right. Rachel Stephens: Now I don't have to do that. Now I just like go in and I ask, it's like, hey, like, Here's my code, I want it to do this, tell me what I change. And it's amazingly good at that. Like it's so convenient at just those, those little tweaks. But again, want to stress that this is not production code. This is a script to make a chart. Right, right. Rachel Stephens: If, if, if I fuck it up or a chat PT fucks it up, like it doesn't have any, like, Outage, impacts, like, so like I have the luxury of getting to use it in a really confined and safe space, um, which is good. Coté: Well, how about, how about outside of coding? Do you use it for, uh, for other things? Rachel Stephens: Um, sometimes, like, I feel slightly guilty about this. Sometimes I pipe in, um, press releases and then have it summarize them for me so I don't have to read the press release. Well, they probably wrote it with AI, Whitney Lee: so Rachel Stephens: you're good. Um, I'll do, I like the summarization feature, I use the summarization feature a lot in terms of just like, give me a summary here, and then I decide whether or not I want to dive all the way in and on. I Coté: could, I could, thinking about a lot of, I mean, I, I read, well, I don't know if I read the same amount, but I still read a lot of stuff, and I think, I don't know the exact right prompt, but I. I would like a prompt that's like, is there anything in this? Like, which definitely when you read a lot of press releases, like you just kind of like, yes, we're going to accelerate the adoption of macroeconomic Edwins, like, and, and just like, it's, I, sometimes it's good to use for just like, tell me the three sentences in here that actually say anything. Um, Yes. Rachel Stephens: I have a gist that I have of just like trying out different prompts to see which ones give me the best summaries. Coté: Yeah. And I'm, Rachel Stephens: I'm, I'm still working on that magic. Coté: Is there any like, uh, like, like Redmonk team thing you use AI for? Like, have y'all cracked, like how to, how to apply it at a team level? Rachel Stephens: From a team level. So like, we are really, really careful that none of our clients stuff. So like, I can put a press release in, Coté: but Rachel Stephens: like, I, like any kind of client, like slide deck or something like none of, none of our client NDA stuff touches anything related to AI because we don't want to fuck up somebody's NDA. Um, so, so I, I think we've all played with local models a little, but like in terms of like cross team, anything, like we haven't done anything cause we've just really like the, the entire analyst model is built on trust and we don't necessarily want Coté: to hand that Rachel Stephens: trust over to a third party software. And, Coté: and O'Grady would probably be like, Well, why pay for it? Like, I think, I think that would also be a, uh, the CFO is into, uh, affordable tool. Rachel Stephens: He, he, he, he is chief build it himself. Um, AI model of, Coté: yeah, that's for sure. I mean, I mean, you bring up like, like I, I think, I think there's, There's a, there's a wall, at least that I've reached with it in my professional life, which is like, well, I can't use it for this. So the utility doesn't, this theoretic utility doesn't exist. Like, I'm pretty sure if I could use it at work, it would be great. I can't. So like, all Rachel Stephens: of like all of our collective notes into one spot, like that would be amazing. Like in query, like it would be so delightful. I can see all of the things that I would love. But we are, we're proceeding very cautiously with anything related to client stuff. Coté: Yeah. We just, we just need, maybe next year we can just, uh, I'll go buck wild and, uh, see what happens. We'll see. We'll see what comes out of it. Well, so I'm first. I'll, I'll, I'll be watching the other Buckwild people over here, secure in my policy. So, so the, the, uh, uh, the new Dora report came out recently. Oh my gosh, Rachel Stephens: I have so many thoughts on this. Coté: Oh, well, great. I was only going to ask one question, but maybe, maybe you can help me understand 130 pages. I think it's only 115. I Rachel Stephens: haven't read the whole report yet, but I got briefed on it beforehand and I felt like there was a lot of really counterintuitive findings in there. Coté: So one of them, one of them, I don't know if it's counterintuitive, but, but what with your, your, uh, as, as you and I used to joke, your numbers background, uh, like I went, I went in some sense of like the, the platform engineering improvements findings that they had and like, I, maybe I'm going to mess it up, but it was basically something like, uh, if you do platform engineering, you're going to get like, uh, 6 percent better. Right. And there's some nuance to it. And that's actually 6 percent better. And at the individual level, you can get like 8 percent better versus 10 percent better and there's slowing, there's a bunch of stuff, but the broader point is like, in reading that my first reaction was like, well, based on the marketing material that I've participated in over the past 10 years, this is terrible, but on the other hand, I was, I was thinking like, you know, if I got 8 percent returns annually compounding on my investments, I would be retired by now. I mean, not maybe, but so, so like when you read, like when I read that there, that some initiative that you're, you've done gives you 8 percent productivity improvements. Is that good or bad? Like what, how do I think about that? Rachel Stephens: I think this is a fair question. I think most companies just make up whatever interest rate they're comparing. Just like, I would Rachel Stephens: say, I would say that most like comparison IRRs are like. Finger in the wind, like five sounds good. I do not believe that most teams are, um, like maybe at one point somebody did the analysis and then that just gets like propagated through all the spreadsheets for all time. And it never, it's like 5 percent it's the company benchmark. Coté: Yeah. They're probably like, if we wanted to change that, we'd go talk to that S they would have to talk to that asshole. Who's the SVP. And like, I got other shit to do with my life. Like, let's just go with it. Rachel Stephens: So. It just makes me think back. I was talking to someone who I worked with a long, long time ago, and there was this huge analysis. I spent like six months of this analysis trying to figure out how to allocate costs across this whole swath of company infrastructure. And it was really involved, really multifaceted. And I have found that that number is still just in use like 15 years later. It's just, it's not the cost of what it is. Uh oh. That is not how that analysis was supposed to be used. Coté: Yeah. Yeah. Yeah. So, so definitely like if you're looking at the, um, the internal rate of return. Right. Which is like the, um, uh, make my definition correct here if I get it wrong. But the, it's the sort of like, if we have a pile of cash, if we do nothing else with it, here's the return we're going to get on the pile of cash. Right. As, as a company. And so therefore, whenever you propose spending this pile of cash, you need to exceed the IRR requirement. Probably once you load in all the risk and it might work or not work, I don't know, by 5%. I mean, depending on the industry, I'm sure in groceries, it's just like, if you can get 0. 25 percent better, do it, right? But like, whatever. Uh, but like, So you've got to kind of like do better than nothing, Rachel Stephens: I guess. I think that's a great way to put it. Like, I think that's a very understandable definition. So I can, I think no notes. Coté: And, and, and so, so, so going back to, to the thing is like, so if we're going to go through, I think introducing platform engineering, as we understand it now is pretty risky based on the anecdotes, how do you say it? Anecdata that I have. Uh, and like, it just. It's difficult. Like, I mean, is, uh, 8 percent more productive teams is, is that good? Like, what, what would you hope for? Rachel Stephens: I think this is one of the things that I have, like, so when you talked about individual versus team productivity, and this is kind of more in the AI space, but something I've been thinking about a lot is we keep seeing these metrics come across We improve, this is like the one that GitHub was throwing out all the time when they introduced CodePilot, it's like 55 percent improvement in product, but like what is, so like one that's based off of a study, not real life. Two, we're talking about individuals, not teams, so it's like you get to code through all these different things faster. Three, Like, how are we maintaining this long term? And then four, like, is a 55 percent that's like, can't, do we have the workflows enabled that like, if somebody has all of a sudden moving through their priorities, 55 percent faster that we're actually getting 55 percent value because like, what are they working on next? So like thinking about all those different things. And then like, if you go back to the DORA report, where like, it's really questionable whether or not these AI tools are creating. This level of like, they were saying that there's differences in stability and throughput in ways that have not happened before. And there are like, whether AI tools are contributing to that seems unclear, but it seems like there's possibly a correlation there. And so I think that there is interesting thing to be kind of figuring out there in terms of how all of that comes together, but in terms of platform engineering. If we want to think about the, the growth rate. So, so this, I'm sorry, I'm all over the place. There's a Charity Majors quote that I loved from earlier this summer when she was on Oxide and Friends with Brian Cantrell and Adam, and she said, infrastructure is everything below. what you care about. And so I want, I want to put all of my like innovation investments in the part that like differentiates me. Everything below that, that's, that's my infrastructure. And so what, what is infrastructure is going to really vary from team to team and person or like company to company. Like what is actually considered infrastructure and where are you differentiating is going to matter based off of what it is that you're working on. But like, in theory, like making those infrastructure investments like a platform so that people can do better, like It's so appealing because like, oh yes, like, like, let's get that standardized. Let's make it so that everybody can do their work and we can innovate up here. And Coté: yeah, Rachel Stephens: you, you, you've lived to the James Water value. I'm like, you know this, but I think that for me, it's, um. It's, it gets really scary for the, um, for the people who are in charge. When all of a sudden you say that those investments, like these are supposed to be the stability investments. And then up here is where we're innovating. And like, usually innovation and risk are the things that are like, go together. And so like, when you start to say like, Oh, like the down here, investments are also the ones that are really risky too. And then everybody was like, it's a whole house of cards that tumbles really fast and everybody gets all anxious. And so we all just like to pretend that the infrastructure is stable. Coté: Yeah, on that topic, I was just talking with someone this morning at some, some big place and, uh, we, we were, we were trying to figure out how you put metrics around. Infrastructure that are associated with the business's success and performance. Right. And one kind of, as you were saying, I think you were going over an example of like, you got to come up with a model. If you have centralized infrastructure, you have to come up with a model to attribute the success of 5, 000 applications to it, you know, whatever. Like that would be a pretty phenomenal platform if you had 5, 000 applications running on it. But you know, the more applications you have running on centralized infrastructure. Well, I mean, I think the more difficult it becomes to attribute business success to the platform, it's kind of like in aggregate, you can be like, the company hasn't burned down. So we're doing something right with the computers, but like in any individual, like dollar or euro, whatever you're spending. It's hard to say that's helped, like, you know, however you're doing that. Whitney Lee: I, I've, oh, sorry. I recently heard this idea of treating a platform as a startup within a company, which has been very appealing to me because it's like, find one proof of concept, one thing you can, one team you can help, one capability you can make, build it, prove the value, use that to get more money. Build and iterate and improve. And so you can see those improvements. You have kind of a baseline to measure from, and you're not doing anything without, um, like it doesn't have a ton of risk associated with it. Like you're only, you're increasing the risk as you're increasing proof of success. So, um, I like that mental model, uh, around platform engineering and as a way to introduce platform engineering. Coté: Yeah. And Whitney Lee: yeah. Coté: And maybe if you deliver that way from the start, it gets easy to track Success. Because I'm sure, well, I'm not sure. I w I would hope like the big public cloud providers don't have this issue. They probably know like this chunk of our centralized infrastructure makes this amount of money. Right? Like, so maybe it's just cause you have lack of, uh, lack, lack of tooling around it or whatever my brain is trailing off on that, that exciting rabbit hole. But the, the, the other, the other thing that, that, that this person and I were saying is, but you also have to be defensive. That if you did your job perfectly and the business was stupid, you don't want to get punished. Right? Like it's almost like you have to show that you provided flawless failure. Right. That, and so like, it even compounds the problem that like, if you do the infrastructure job well, and the business just like bones it up, then like you still want to get promoted for having done a good job, right. And instead of punished. Whitney Lee: That sounds like a humanity thing, right? Another thing that's interesting about infrastructure and developer productivity and how they relate. So we talked about it in the context of platform engineering, but then also, I think we can discuss it in the context of AI. Like Rachel was saying, AI, uh, can make 55 percent faster developers. Well, I would like that to not be attributed to me. Fair, fair, fair that she read somewhere. Somebody said, yes. But the point is like, okay, we have that speed, but then at what infrastructure costs, like a pure dollar amount cost, or in terms of like the cost of, um, having to run super huge workloads, the fact that it might break things that we have to reinvent our processes around infrastructure, now that we're dealing with AI work sized workloads. I think it's a similar problem in another space. Rachel Stephens: I think you're absolutely correct. That makes a lot of sense to me. The other thing, so when you're tying back to like the momentum of AI, I had a conversation with someone who's top down mandate on tooling is 10 percent of their budget goes towards AI related products because there's just so much FUD in and around this space in terms of like, if we are not on top of this, we will like, we will lose out forever. And so, And then you also have the fear on the other side, where it's like, we don't want to leak all of our data accidentally. It's just like, there's the fear if we do, and there's the fear if we don't. And everyone is trying to figure out how all these come together. But like, I really do believe, Whitney, that like, what you're saying, if you are making infrastructure investments that enable, yes, those AI workloads, but also all of the other workloads, like, that's the future. That is a bonus for all of the things. And I, I wish that more people maybe were approaching it that way, rather than just like hair on fire, we, we've got to throw money at AI. Whitney Lee: And I think with infrastructure, it's hard too, because it, it doesn't relate directly to the value stream. It's, it's indirect. And so it makes it a bit of a harder sell too. Yeah. Coté: Yes. That's, that's been, that's been my marketing quagmire for the decade, coming up with various ways of, uh, of, of doing that. You say, so you brought up, uh. You brought up, uh, an example, like you need to spend 10, 10 percent of your budget on this AI stuff and, you know, to elaborate on the joke, it's like, you know, you would ask, what do you want me to spend it on? And, and they would be like, got to go to lunch, right? Like, I don't know, but it's better, whatever you pick better be awesome. And, and like, so, so that brings up like, you know, the whole, this is especially big in the developer world is, is like the whole idea of like, this is very James Governor take on things like software development is fashion. Right. Like, it's just like things that you go through here and there. I think the, uh, I think, I think, uh, my former coworker, Bridget, she used to like to phrase resume driven development, all that kind of stuff. And like, what's, uh, what's the state of fashion driving selection nowadays? You think, you think it's bigger, smaller, like, is it a thing? How does it play out? Rachel Stephens: This is fair. So one of the things, so if we're going to go back to like the term resume driven development, like in my career in enterprise, the things that made me more valuable as a colleague, most of the time, We're like those internal knowing where the bodies were buried kind of things. Like we made this decision for this reason, because it pissed this person off. And therefore these are like, these are all of the things I do. It's like, so it's like, yes, knowing the data, like, yes. Knowing the, like the architecture, how to use the tools, like all of those things. But it's like, it was understanding that internal landscape about why things had happened over a period of time. So it's like the more deeply I knew the company itself. The more useful I was to the company, whereas like when I'm wanting a job, like none of that is like, I mean, like in broad, like social skills, it's applicable, but like from wanting to get a new job, it's like, I know this tool, I know this technology, I can pick this up this quickly. And. One of the things that I think could be interesting if we can ever get it right, it's like can we get an AI system in place where like some of that social knowledge is shared internally in a way where like we don't have gatekeepers, we don't have somebody who's like hoarding something just because like that is their job security, like we can document how and why things work, like this is totally me being, um, Naive, probably, to the point of Pollyanna ish, but can we make it to the point where, like, some of that hoarding of internal knowledge becomes more shared and distributed and, like, something besides just a SharePoint? And we can, um, start to, like, Can we, can we let people focus on some of those things? And like, maybe we can make better tech decisions as an industry, because we like, instead of like having to chase the resume driven development, like maybe people like get a chance to look at the tools. Like that was probably Anna. But, um, I do, I do think that there's a way where it's like not an either or. Whitney Lee: Isn't there like, uh, like people say their zoom meetings are being recorded or somehow the company has access to transcripts of all the zoom meetings. Like maybe, maybe that's where AI comes in and preserves a history of how decisions are made. If it's even true, that might be an urban legend. That's, that's a great way to Coté: make everyone paranoid. Rachel Stephens: I think there's going to be, I feel like we're entering like a log era where it's like, there's going to be a whole lot of unstructured data and there's going to be a really hard signal to noise problem. It's like, yeah, you can, you can start to solve some of those things. But right now we're. Right now we're just making more noise. Do you Whitney Lee: think AI is going to fundamentally change humanity? Do you think your children's adult lives are going to be drastically different than ours because of AI? Or do you think it's more like a trend? Rachel Stephens: I'm kind of glad the week before the election I'm just hoping my children have adult lives. I'm sorry, we can cut Rachel Stephens: that part Rachel Stephens: out. I do think it will matter. Like, I think just watching, um, people try to figure out how to adapt to AI in education spaces, it seems unlikely that they will be untouched by AI. Like, whether or not it fundamentally changes things, like, I kind of hope not. But I do think it will be part of at least the landscape that they have to navigate. Whitney Lee: Uh huh. Sometimes I wonder if all the easy tasks automated away by AI. How's anyone ever going to become an expert? Rachel Stephens: I think this is a really fair question. And I had this conversation with, um, it was a customer at an Adobe breakfast forever ago. And it was when they introduced like content generated autofill or something. It's like that used to be. A junior designer's job is to like go in and like trace all of those little things, like make, make what I don't know anything about art clearly. And so, but anyways, um, it used to be a junior designer's job and now you can do it with this feature that Adobe has introduced. Like, what is the career path to get into this field? And I think that we're just going to continue having that conversation over and over again. And we started having this conversation well before it was called AI. Um, but I think that's going to be accelerated and more heightened now. Coté: Yeah. You know, to, to invoke the old Jurassic Park thing, right? Like there's, there's not a lot of people asking if we, uh, if, if we want this thing to, to speed things up, it's kind of an example of like, uh, I don't know, the ridiculous example I always think is like, when you look up how many different types of screwdrivers there are, like, at least my conclusion is just like capitalism needs to calm down. Right. Like, I don't think we need this many types of screwdrivers. We probably could have five, but I'm sure someone was like, here's a way to make money with a screwdriver. And I don't know, maybe that's the engineering screwdriver. Rachel Stephens: I feel like if you're going to go to war with all the screwdrivers, but you have to also go to war with all the people making all those different screws and then it's just going to spiral out of control. Exactly. It's a classic screw Coté: screwdriver conundrum, which, which came first. Well, as, I still have some, uh, so some analyst world questions having, having worked in the analyst world for a while. I haven't actually worked inside it. You know, I talk with people like yourselves and the gardeners and forestry people every now and then, but I think you've been doing this like 10 years now, is that right? Yeah, close. I'm at eight. Yeah, yeah, yeah. Yeah. That might as well be true, depending on how you round. Yeah. And, and I think so that, you know, let's, let's, let's say, uh, I'm not, I don't like to do math in public. But that's, uh, 2016, 2018 or something. Right. Yeah. And S and so like, as an analyst, I Whitney Lee: think Coté: you've lived through the like dominance of video, As a format on the web, like that seems to be the premiere. That seems to be the top medium for content on the web. Now, I don't know if that's, that's true, but that's my anecdata to use that dumb phrase that I love so much again. And so like, what. How, how, how is, how does that affect the way that, that not only you and RedMonk work, but when you look at the other analysts, what good and bad are they doing about video as, as that medium shift? Rachel Stephens: Yeah, I think for me, it's two ways, both as a producer of content and a consumer of content. Oh, that's a good framing. Coté: Yeah. Rachel Stephens: As a consumer of content, I will take written every single day. Um, because you can skim it. Like I just, I can consume it faster than I can video. I do love, like, I think podcasts and videos are great for kind of, um, videos are great if I can listen, so it's like the, like the, Around the house, like I'm washing dishes and I can listen to a podcast or I'm going for a walk and I can like catch up on a conference talk that I wanted. It's like, those are good. It's like if I'm having to like sit in front of YouTube and like watch a tutorial, like I have to be in a really specific place with a really specific need for that to be the correct medium for me. Yeah. Rachel Stephens: Um, but as a producer, it's like, we do, we focus a lot on written work, but like, we've also tried to branch out into these other mediums as like, we're doing more video, video interviews, we're doing more podcasts and it's just really this hope of reaching people where they are and trying to figure out what sticks, but like. If I were to say what, at least for me, gets the most responses from people, it's the written stuff. Coté: Yeah. Yeah. I mean, I mean, I feel like quantity wise, y'all have done more videos than written stuff over the past couple of years and, and to, to, to be fair, it's videos are smaller, easier, right? So like writing Rachel Stephens: easier. And I think also there was in pandemic land, a very, um, variable chunk of money that had previously gone to in person events that suddenly needed a home. Some videos were one of those things. And so part of it is that, that, um, that hangover from just where, where the money was in 2021 to 2022. Whitney Lee: To bring this full circle, I'm confused again about what an analyst Okay, I'm Rachel Stephens: so glad I'm a clarifying Whitney Lee: analyst. Rachel Stephens: Well, I thought someone was The more Whitney Lee: I talk, Whitney Lee: the less you understand what it is. We're analysts. I thought you were, like, you said you were paid to, uh, you listen to anybody, but you're paid to talk. But now, if you're trying to make content, then you're, you're trying to get people to, to listen to you talk. So I'm confused. This is Rachel Stephens: fair. Um, sometimes I talk for free and sometimes I And what do you hope to get? Turns out it's not a perfect metaphor. No. Um, for things like this, like, delighted to come chat with you all. Like, a lot of podcasts I will guest on, um, and that's not, like, sometimes clients will be like, we want to do a webinar, and we want, like, And I'm analyst to come give a landscape view of how they see the ecosystem. And then we're going to talk as a company about how we fit into that ecosystem. And those situations like we would be doing the content, but it would be, um, a paid content. Yeah. Whitney Lee: Okay. So when you're making RedBot content, you're, you're doing so. Public content. You're still doing that in service to your clients. Rachel Stephens: Sometimes it's a mixed bag. Rachel Stephens: Some as a client wants to have, they want to put some of their company forward or one of their products forward and we'll do that. And sometimes it's like someone in the community that we're really excited to talk to and we just invite them on. So it's, it's a mix. Whitney Lee: That's cool. And it's about, I'm sure public perception too. Rachel Stephens: Yeah. Coté: So when, when you're, uh, you know, just as far as the, uh, Rachel's tools and, and, and tech tips here. Like, so, so when, when you're consuming this video content, like how, how, how do you do it so that you have the same, I don't know the words, learning effect that you would from text content. Right. Cause like, and, and I'll, I'll, I'll, I'll tell you a little bit about like how I think about it. Why I'm asking this question is like, I, I, I do like, like probably all of us here do like I'll, I'll save up. A YouTube video to listen to while I go on a walk and Coté: is it a good excuse to take a long walk? But I'm not sure if it's as good as if I sat down at my desk and like took notes while I was watching it, let alone as good as if like, this was a write up that I read and I can go and highlight things, but I don't feel like I have a choice now, like I need to be able to consume videos. As, as content. So like, how, how are you making that work, uh, as well? Rachel Stephens: It's both ways. I, so we were talking before I started about Whitney and her flashcards, which is amazing, like the way that I maintain things is like, I am like, I take stenographer level notes. Like I, I take really extensive notes, um, which I don't always like, obviously you don't do when you're on a walk or something like that, but like the other thing is. The being in a different place can trigger a different, like, I still remember exactly where I was on a walking path. When James Waters said something that like inspired a blog post, I'm like, Oh yeah, that's a really smart thing. It's just like, there's Rachel Stephens: like, there's, there's something about, um, being in different places and experiencing like experiential learning, which is different than note taking learning. So it just matters. Um, I think it's what the content is, what you're trying to get out of it. And you just kind of have to roll the dice. And there are definitely times when I like have to like, take my phone out and like, type things down. And before I'm walking, I'm just like, I really like that quote. Coté: But that's, that's, that's a good reminder to like, in the, I mean, there's two sides of a medium. I'm, I'm not up on my McLuhan or whatever, but like, you know, the, the medium will force the way you talk about something, but then it also forces the way you consume something and kind of like what you're saying, what you should do is think about what advantage do I have while I'm walking, To consume media. And how does that change the way I'm doing it? And I would imagine, I don't know, you got more blood going to your brain. I, I don't even know what it is, but like there is. You're right. There is something, at least in my own experience, I think there's a higher rate of epiphanies while I'm moving around than while I'm sitting at my desk. Rachel Stephens: Yes. And I mean, like rough rule of thumb is panels and like fireside chat esque things are, I think for me, better suited for walking. Okay. Well, Whitney Lee: and then there's, It's different stuff you're hoping to get from it. Like, I'm reading this article and I need to do my own write up. I need notes because it needs to have a final format. Or I'm just exposing myself to material and it doesn't really matter if it sticks very much or not. Yes. Yeah, yeah. Coté: Huh. Yeah. All right. So, so the other, the other analyst thing I had is, uh, There's, there's more of them out there, but, but there's been a lot of like consolidation and new analyst firms. Like I think, I think, and I'm not looking, unless you want to, I'm not looking for you to talk smack about any things, but just reflect on being an industry analyst of the industry analyst business and, and like, you know, there's like the Futurum group out there that seems to have like, they gobbled up a few, like that CTO advisor guy. And I think it's like, was it, it's like the Patrick Moore just renamed his firm, I think to Futurum. I don't know the funding of it, but like, there's a bunch of consolidation and they seem to have like gone batshit for video in a good way. Like they do a video all the time. That's kind of who I was thinking of when I was thinking of like, how do you consume video? Uh, but like, like what else? I don't know. Is there some consolidation stuff going on or some different analyst startups? Or is it just that one group? Like, is there interesting business models happening? Rachel Stephens: I would say definitely a lot of the consolidation I see is in and around the Futurum group. So I, I don't know all of the business models behind it, but definitely like from inside the industry, that is also what I'm seeing. Coté: Okay. Yeah. Because it seems like, you know, well, not, you know, but I, but I, I did M& A for a tiny bit and the analyst business having been in it is a terrible business. If like, like if you want to do a rollup, right? Like the margins are small, like brand can be a thing, but the biggest problem is like Like, you know, uh, like you can imagine if Redmonk got acquired, like whatever the lockup period was, a lot of people would leave because they would just be like, finally, right? Like I've got a big chunk of money and, uh, now I don't have to do this. Right. And, and then the way that you would, if you buy an analyst, if you buy any company, you're looking to get, speaking of IRR, I don't know what the standard rate is, but probably at least you want a 20 percent return. Otherwise, why did you take on all this risk? And like, I don't know how you would like spend a lot of money on an analyst firm and then grow it 20 percent like with all these other risks of people just leaving. So it seems like, you know, it seems like a difficult thing to like do a rollup of, of analysts, uh, versus not like, I don't know, it Rachel Stephens: would definitely, I mean, in theory you could be. Acquiring clientele, things like that. But like in a big sense, it would be an aqua hire and there's a lot of risk to aqua hire. Coté: Right. Right. In which case, why not just hire the person, right? Like, like, I mean, I mean, I guess, I guess it may, maybe, maybe there's some sort of like rolling thing to it where like, if you pay, I mean, how much multiple could you pay in an analyst firm, like three or four X? Like, it just seems like you don't have enough, you know, Revenue to play with. Maybe I haven't thought about it enough, but, um, anyhow. Rachel Stephens: Or it's possible you've thought about it entirely too much. Coté: Like, like, like, like maybe to round it off. Right. Like, I mean, the ultimate like lone wolf analyst now is like Ben Thompson. Right. And like, if you do, if you do the calculations, he's got to be clearing, like, Some kind of low seven figures. And you know, he's got expenses as far as like, I think he actually has a staff of maybe like four people or so, and this, that, and the other blah, blah, blah. Right. And actually now he's got more since he has all these podcasts. So who knows what his payouts are? And I think he flies business class everywhere. So you got that going on too, but doing plenty fine. Right. But you can imagine that if you bought like the Ben Thompson empire, unless you gave him a lot of money. He wouldn't stick around for a while. And the next thing you know, he would start like Stratechery where he finally spelled it correctly. Right? Like he would just go off and start his own thing. So it's like, I don't know. I know, I know I'm beating the dead horse here, but it seems like that's a, that's a tough, tough thing to do. Uh, Rachel Stephens: If you wanted to buy Stratechery, like what you would want is Ben Thompson. And like, I don't know if he, he wants to be, he's obviously Coté: not interested in that. Rachel Stephens: Yeah. Coté: He won't even write a book for Christ's sake. Whitney Lee: I have a question that I think will help me understand. The, the, your job, your job in specific, but like also analysts in general. And that is, um, well, how would the software industry be different if analysts didn't exist? Rachel Stephens: This Whitney Lee: is Rachel Stephens: a good question. I think in some senses it might not be different at all. For what we kind of talk about how it is, it is. This outside in view of how companies are speaking to the market. So it's like, you tell us what it is that you are working on, how you're presenting the things, how you're trying to reach different audiences. And then we help you fine tune that messaging. So it's like the thing that a CTO wants to hear, it's not the same thing a developer wants to hear. So how do you take the same set of information and present it? in a way that is compelling to both audiences. Like, who needs what information first? What level of detail do you need? Who needs to, like, kind of, where am I diving in versus where am I zooming out? So a lot of it is just kind of helping remove that friction in terms of, because it's hard to see that when you're in the product day in day out. And I think the other thing that we do a lot of times, at least for some of the bigger clients, is some of that internal politics. So it's like, one group thinks this, and one group thinks this. And let's bring in a third party, and let's like, what do you all think? And we can kind of help, um, it's not necessarily that we, you know, Um, so, um, We can lay fingers on the scale, but we can present what it is that we view of the landscape, and they can do with it what they will. Um, so. Whitney Lee: Interesting. I hadn't considered that use case. Rachel Stephens: And then there's also some where people will come and like help us see around the corner. Like, especially in and around license changes and stuff we'll have, like we might not agree with what our client does in terms of a license change. But they'll come talk to us first, and we'll help them kind of, we'll say what it is we think, and then they will make their decision, and it's like, okay, given that that's your decision, here's what you should be prepared for. Things like that. Whitney Lee: Are there a thousand, like, liability forms you have people sign before you say what you think? Um, our Rachel Stephens: RNDA and contracting processes, um, Prolonged. Very often. Whitney Lee: It seems like a dangerous business. I Coté: think it helps that one of the founders is married to a lawyer. So you've got, you've got literal in house counsel at Coté: any point. This has been great. I think this is the, uh, let's we'll call it third time you you've been on software to find interviews. The first time I think was six minutes. But we'll round up to, to, uh, to a full thing. Uh, but it's, it's always, it's always nice talking to you and, and, you know, to, to add onto a little bit of what you're saying as a, and Whitney was asking you as an endorsement. I think, I think a lot of what industry analysts do, um, isn't really seen and therefore not appreciated very much. And I, I, you know, I, I would say, um, you know, you know, to, to, to use the old fun phrase, Uh, what's valuable about an individual in a company who's been there for a while is they know where all the dead bodies are, or they, they know the history of decisions and why they were made. And then also they know the APIs of individuals and groups. And, you know, like, you know, you can't use green on your slides with that person. They're going to freak out. Like, so you've got that. And I think at an industry level, that's a lot of what industry analysts, uh, bring to bear. Like it doesn't really get published, but it's valuable to talk with them because they have this experience. Industry memory, uh, if you will. And in particular, I think unless this has changed, like RedMonk is very good with, uh, on the vendor side with, with a lot of that as well, just knowing like these things have been tried. And like the research you've been over, here's the way open source stuff works if you're doing that. And, um, I think now that I've worked with, with vendor side people a long time. I don't know a charitable way to put it, but they don't always have a lot of time to be experts. Right. And so like, it's good to have, instead they get budget to hire people who are experts. Rachel Stephens: It's kind of like what we were saying before about resume driven development. What is valuable to most companies is that internal company focus. And so having someone who can come in and inject that outside company view and whose job it is to kind of consume on, read all this, like most people don't get paid to read all day. Coté: Yeah. Rachel Stephens: And so like, can somebody synthesize data for me and bring it outside? Yeah. And Coté: you know, putting all that into, putting all that into action. Operationalizing it, if you'll pardon the phrase, that's like the difficult part of things. Right. So like, you know, you can pay anyone to tell you stuff, but like figuring out how to make it work in your organization is that difficult. I think, uh, what do they call that in economics? Your separation of concerns. What's, uh, I, I forget what it is. Do you, the, uh, the, the Portuguese make the wine and the British buy it, I think, uh, is, is how it works. Is that Cardinal Richelieu or something? Uh, anyhow, thanks for being on. If people wanted to like, uh, look you up, follow your stuff, which they should, your, your blog, uh, where would they go? Rachel Stephens: Just come on over to RedMonk. com. R E D M O N K. com. Coté: Yes. And you can see the, uh, there's five of you now, is that five analysts? I should say. Yes, Rachel Stephens: five analysts now. And we are, um, we would be happy to chat with any of you on like the world's fractured social media landscape. So rather than trying to list all of those, just come find us at redmonk. com and branch out from there. Coté: Oh, that's a whole other topic. I forgot, wanted to ask about. We're wrapping up, so, uh, so, so we'll do that. Well, you've listened to another fantastic, I think, episode of Software Defined Interviews. This has been episode 86. See, I didn't have to ask you this. I was ready. So if you want to get the, uh, things that we talked about and a link to Rachel and, uh, Redmonk, you can go to softwaredefinedinterview. com. Sorry, Software Defined Interviews. There's more than one. Now you should go look for the domain name. SoftwareDefinedInterviews. com slash 86 and find that. And if you've listened this far, you should of course subscribe and I don't know, do whatever that social media bullshit you do is to help us out. You know what you should be doing. We'll see everyone next time. Goodbye. Whitney Lee: Thank you. Rachel Stephens: Goodbye.