Open source threaded team chat?! (Interview)

28:16but I'm curious about Zulip when it comes to the software and the way it all works and does it fit into all the different places that you communicate because more often I'm using Slack on my phone even even though I stand at my desk for hours every day you communicate all day long and all night long and so Zulip on the phones, Android iOS Zulip on the web, Zulip apps Do you have all that necessary surface area accounted for, and how do you all manage that? Yeah, absolutely. So Zulupe, you can use it just in a browser tab. There's also a desktop app for all the major platforms. And then, yeah, Android and iOS apps.

28:59And we're actually currently in the process of re-rating our mobile apps from the ground up using a different framework. We're switching to Flutter-based apps. So our current apps are definitely functional, but not as smooth and beautiful as we would like them to be. And so that next generation app is really going to get us all the way there. So we're very excited for it. And for old school folks out there, there's also a terminal client for Zillow, if anybody wants to use it that way. How about API? Is it programmable? Yeah, there's an open API and actually our mobile and terminal apps use the API to communicate with the server.

29:40So we're constantly kind of testing it ourselves and using it ourselves and relying on that documentation ourselves. So absolutely. Is the desktop app an Electron app? It is, yes. Have you considered Atari app? My understanding is that the engineering team was thinking about it and was kind of waiting for... Because if you wanted to get the nerds excited, I think if you came out and said Zulip desktop app is now no longer using Electron, then it'd be like we just throw slack right out the window. Wouldn't we, Adam? All of us nerds would be like, oh, finally, something we can use here. Yeah, pretty much.

30:27okay friends i'm here in the breaks with annie sexton over at fly and you know we use fly here at changelow we love fly it is such an awesome platform and we love building on it but for those who don't know much about Fly, what's special about building on Fly? Fly gives you a lot of flexibility, like a lot of flexibility on multiple fronts. And on top of that, you get, so I've talked a lot about the networking, and that's obviously one thing, but there's various data stores that we partner with that are really easy to use. Actually, one of my favorite partners is Tigress. I can't say enough good things about them when it comes to object storage.

31:07I've never in my life I thought I would have so many opinions about object storage, but I do now. Tigris is a partner of Fly, and it's S3 compatible object storage that basically seems like it's a CDN, but is not. It's basically object storage that's globally distributed without needing to actually set up a CDN at all. It's like automatically distributed around the world. And it's also incredibly easy to use and set up. Like creating a bucket is literally one command. So it's partners like that that I think are this sort of extra icing on top of Fly that really makes it sort of the platform that has everything that you need.

31:43So we use Tigris here at Changelog. Are they built on top of Fly? Is this one of those examples of being able to build on Fly? Yeah, so Tigris is built on top of Fly's infrastructure, and that's what allows it to be globally distributed. you did. I do have a video on this, but basically the way it works is whenever, like, let's say a user uploads an asset to a particular bucket. Well, that gets uploaded directly to the region closest to the user. Whereas with a CDN, there's sort of like a centralized place where assets need to get copied to. And then eventually they get sort of trickled out to all of the different global locations.

32:17Whereas with Tigris, the moment you upload something, it's available in that region instantly. And then it's eventually cached in all the other regions as well as it's requested. In fact, with Tigris, you don't even have to select which regions things are stored in. You just get these regions for free. And then on top of that, it is so much easier to work with. I feel like the way they manage permissions, the way they handle bucket creation, making things public or private is just so much simpler than other solutions. And the good news is that you don't actually need to change your code if you're already using S3.

32:50It's S3 compatible. So like whatever SDK you're using is probably just fine. and all you got to do is update the credentials. So it's super easy. Very cool. Thanks, Annie. So Fly has everything you need. Over 3 million applications, including ours, here at Changelog Multiple Applications, have launched on Fly. Boosted by global Anycast load balancing, zero configuration private networking, hardware isolation, instant wire guard VPN connections, push-button deployments that scale to thousands of instances. It's all there for you right now. To pull your app in five minutes, go to fly.io. Again, fly.io.

33:29And by our friends over at Paragon, use paragon.com. Check them out. Ship every SaaS integration your users need. With more than 100 plus pre-built connectors, you can add dozens of integrations to your app quickly and reliably with their embedded iPass for developers. And I'm here with co-founder and CEO, Brandon Fu. So Brandon, talk to me about the friction developers feel with integrations, SSO, dealing with rate limits, retries, auth, all the things. Yeah, so there's a lot here and I think there's a lot of aspects to the different problems that you have to solve in the integration story in building these integrations and also providing them in a user friendly way for your customers to self serve and onboard and consume those integrations.

34:14So part of what the Paragon SDK provides is that embedded user experience, again, what we call our connect portal. That's going to provide the authentication for your users to connect their accounts. That's going to be the initial onboarding. But in addition to that, your users may also want to configure different options or settings for their integrations. A common example that we see for Salesforce or for CRM integrations in general is that your users may want to select some type of custom object mapping. Every CRM can be configured differently, so your users might want to map objects to some different type of record in their Salesforce or different fields in their Salesforce.

34:49And typically, that's what developers would have to build on their own, is this UI for your users to configure these different settings for every single integration. That's also going to be what's provided by the Paragon SDK is not just that initial onboarding and authentication experience, but also the configuration end user UX for different settings like custom field mapping, selecting which types of features on your integration that your user might want to configure. And that's also going to be provided fully out of the box by Paragon SDK. With integrations, different APIs might have different rate limits.

35:25They might have different policies that you have to conform with. And your developers typically have to learn these different nuances for every API and write code individually to conform to those different nuances. With Paragon, because we build and maintain the connector with each of the integrations that we support in our catalog, we're automatically going to handle for things like retries, things like rate limits. And so we look at this as sort of the backend or infrastructure layer of the integration problem that we have spent the last five years essentially building and optimizing the Paragon infrastructure to act as the integration infrastructure for your application.

36:03Okay. Paragon is built for product management. It's built for engineering. It's built for everybody. Ship hundreds of native integrations into your SaaS application in days or build your own custom connector with any API. Learn more at useparagon.com slash changelog. Again, useparagon.com slash changelog. That's U-S-E-P-A-R-A-G-O-N dot com slash changelog.

36:47What are some of the biggest challenges you all are facing? Good question. I guess, I mean, one sort of thing that is complex for us is the competitive landscape, like Slack and Microsoft Teams being the sort of big gorillas in the room and Teams effectively gives away their chat for free and oftentimes kind of as part of their Microsoft suite and it's really hard to get folks kind of, And, you know, at the same time that it's free, it's not free, right? In the sense that people are spending their time and their energy and their attention in ways that aren't making them productive, you know, like they're wasting your employees time as your most valuable resource.

37:33And so, you know, wasting that time and energy on an app that's frustrating or hard to use or is not organized in ways that you'd want it to be is a major cause. but it's hard for companies to budget it that way and to really evaluate it that way. So I think one thing we're really trying to do is get better at telling that story and really communicating with folks and trying to explain this, make people really sort of feel in their guts this sort of, okay, this app might be free or it might be kind of an easy choice like Slack for most, a lot of folks are familiar with it. It's sort of like nobody got fired for buying IBM, probably nobody got fired for picking Slack for their chat.

38:16And there's lots of things that are great about it compared to products that had come previously. But choosing a chat app is just so important to how folks are going to collaborate in your organization. And so that's really the message we're trying to get across. And that's, I think, a big challenge for us is to really get people off of their default mode or the easy decision there and really, really get folks to consider and evaluate our product and to take that time and attention away from so many other things that they need to be doing to really think about this choice in a very intentional way.

38:55It is hard to compete against free, especially when the Goliath is giving it away for free. Yeah, I mean, there's, you know, Microsoft is facing anti-competitive lawsuits in Europe because of how they've set things up. Yeah, it's unfortunate, especially you would think as a user, like you said, nobody got fired for buying IBM. I didn't make that up, but I don't disagree with it to some degree. Except for what if you're missing out on what is free and open source, but you can also pay for it when the Zulip name isn't as polished as maybe Microsoft, obviously. You know, that's the hard part is that you kind of have to win them with showing up, you know, with the open source-ness of what you're doing.

39:46the way you've been in the trenches with the communities, the way you've sponsored things, not just simply the larger brand name and the literal freeness that you can get with Teams. Now, I know that at certain points, organizations have to pay for Teams, but it's pretty much free for the entrance. And then you pay once you're literally locked in. Yeah. And I think in the past, a couple of things that have held us back have been one, the design of the app. that's really something that we've been focused on improving. That's been a major, major investment for us over the past year or two and continues to be.

40:23For the longest time, our users would tell us that the user experience in Zulip is second to none, but the design could use some work. And that's not such a big problem necessarily for folks who have got kind of like, once you've gotten used to an app, you might kind of stop noticing some of these things. But in the initial evaluation, it makes a huge difference. If you open an app, you're like, oh, this doesn't look modern. This doesn't look beautiful. And so we're really trying to get away from that and have folks have an immediate kind of like positive response to the app as well as enjoying the UI over the long term.

41:00And then another thing we've been really focusing on is that the onboarding experience, because there is a little bit of a different mental model for Zulip compared to other apps folks might have seen. And we do want to have that be easy to understand and easy to onboard people, easy to get everybody in your organization, kind of have folks get started with that. And also just, I think almost any app when you first encounter it might feel a little overwhelming. Like, you know, if you've never seen Discord before and you open it up and there's a lot going on. Yeah. But there's some of these apps that we're competing with.

41:34Most folks have seen them before. And so now they kind of have forgotten that first initial feeling of like, oh, there's so much happening. It's different. So we really want to help folks through that experience with Zulip because we do have a lot of users who are coming in who haven't interacted with it before to really get them across this threshold of like, oh, I get it. This is comfortable. Some things about it are different, but a lot of patterns I'm familiar with from other applications. work here as well and it really is pretty intuitive once I kind of like have a handle on it. Well I asked in our Slack community just moments before hopping on if anybody's used Zulip and what they think about it and like one person said used it at a different company liked it a lot it's like it's kind of like Slack the higher-ups replaced it with Teams as Zulip wasn't quote auditable so that was it wasn't the free part it was the auditable which to me makes not 100 % sense, but there you go.

42:26Yeah, I'm not so sure because we do provide different ways to export your data, including compliance exports, or you can just export it. I don't know. Okay, okay. Yeah, he says it was infinitely better than Teams. So there you go. All right, well. So that's cool. But that's an example of that kind of what might be a little bit, I don't know, it's like, I guess folks have their own priorities and I don't want to second guess the management, but that's just where that perspective of your team's efficiency and how happy they are with the software they have to use every day. I don't know anybody who likes Teams.

43:00I know lots of people that use it. I'm not a Microsoft hater anymore. I used to be when I was a younger man. But I will say that I don't know anybody who says, like, Microsoft Teams, that's good software right there. We love it. No one says that. Has anyone ever said that to you, Adam? Not directly. Indirectly? Like you were listening to them talk to their spouse or something? I don't know. I love Teams. Through the tea leaves or something. Now, Discord, people seem to love, and I'm not really sure why, personally. I've signed in. I've joined some Discords. It seems like a hot mess to me, but it's very big in gaming communities, musicians, and crypto scam artists I know use it, other communities.

43:41And I'm not sure what it is about Discord. I know they have some cool audio features built in. They kind of have a lot of different stuff because it came out of, I think, gamers would hang out and talk to each other initially. Do you have a lot of, do you ever have to compete with Discord or do you have to explain Zulip in light of Discord and how you all differentiate from them? So it depends. Discord is not so much designed for business use or use within organizations that needs to be closed and have sort of, because it's a single account across all your organizations, it's sort of a different structure there.

44:16We do have folks who are Discord users who have definitely requested some features that Discord has. I would say that, yeah, their kind of video and calling and the way they do that is quite nice. And that's something we've heard folks interested in. Something that we're actually working towards is Discord has, maybe if you haven't administered organizations, you haven't explored that side of it. But they have really nice and flexible ways to manage permissions and groups within an organization. And so that's actually a big project that we have going on right now, like really, really flexible permissions management where you can create an arbitrary group and then give that group kind of an arbitrary set of permissions within your organization.

44:59And I think that's going to be really, really nice for anyone administering a larger organization. That's one thing I really wish we had in our Slack, Jared, is that we have people come and they share things they should not, a.k.a. spam. Yes. And I would just like to be able to eventually boot them because I delete the message. And I look at them and I'm like, well, you're clearly not here for the reasons everyone else is here for. You violated the code of conduct intended for this place. There's no way for us in our current state to enforce these kind of things aside from just deleting messages.

45:40Sure, we could probably log into Slack and delete their user, but that doesn't stop them from coming back. I'm not sure if any platform can really do that to prevent somebody from recreating a new account or whatever. But I do wish we had some moderation tools where I'm sure even the community inside of our Slack would step up and say, you know what? I'll help you guys because it's 2 a.m. and you're sleeping and I'm not because I'm in a different country. And if I see a spam message, it doesn't have to sit there for eight hours until the morning or whatever time it is. when we look at Slack again, it's like, well, hey, this thing has been sitting there with, you know, people piling on or looking at it or clicking it.

46:17And we can't do that stuff. So I do, I wouldn't mind having some moderation tools. Yeah, we have some tools like, I guess if you deactivate a user, they won't be able to rejoin with the same email. And you can also disallow throwaway email domains if you want to prevent, definitely is helpful for preventing spam. But you can also, personally, you can mute a user. So if you, as an individual, don't want to see somebody's content, we do let folks have the option of muting that person. And that just hides all their stuff for you. So you never have to interact with it. It would be cool if you could auto-block new users if they start a message with dear sir slash madam.

46:59Auto-block, sorry. Write a bot, I guess. yeah some sort of pattern match oh yeah yeah I mean it doesn't happen often we get some spam here and there and mostly I get it go join a slack or find a place to belong and share your messages and you do that with enough numbers you'll get people I get it it's a numbers game but it doesn't make any sense to me because you're not really getting the long term benefit you actually want for a brand And so it's such a nasty thing, really. And like I said, it doesn't happen too often, but often enough where I'm like, yeah, I wouldn't mind some tooling. What would a migration look like?

47:43So for something like moving from Slack into Zulip? Just for instance. Yeah, sure. Just for instance. As a random example. Hypothetically speaking. Yeah, yeah, yeah. Apropos of nothing. Yeah, so we have instructions on our health center for how to go about it. So basically what you would want to do is, assuming you want to keep your message history, you can export that through Slack. It might be limited, I guess, now, depending on your situation. And then if you're moving, say, to Zulip Cloud, so that's our managed SaaS offering, you would just send over that data to us and we would import that into a new organization for you.

48:23And so you could preserve all your, not just the messages, but also the user data. So you'd have a running start on that. And then we also, I don't know if you guys have integrations, but also to make it easier to move over your integrations, if you have any, we have Slack-compatible webhooks. So basically you could just kind of remap where your webhooks are sending in their data to be Zulip. And then on your own time later on, if you want to move over to more like Zulip native integrations, then you can do that but things would be working for you right away so yeah and you can tell folks where to log in or we can automatically send emails to all the users that you imported with their login information however you want to manage that and they would just get an email and they would maybe have a password reset on their first sign up or obviously you're not going to import their passwords yeah and we have all the social auth as well so if folks want to log in with their you know google account and good hub or anything like that that's also on offer that's the it's 90 of my anxiety if we're hypothetically speaking about things yes we are is i feel like i'm just i've been like in this waiting pattern in my own brain you know i haven't taken any action i've been like just thinking that maybe slack would someday get it and somehow just recognize that there's so communities that have you know built up their thing around them and that many of us in even developer land or just let's just say tech land have numerous uh logos slash icons in our slack sidebar so we bounce from one workspace to another and i like that i don't want to be in a world where i can't where i have to like still i guess keep slack or i just like the unification of it and And as a user, I don't want to have to go to the Slack app and then the Zulip app and then the whatever app.

50:22I would just like a unification if it was possible. I'm sure it is. I think there are some out there, but there's diminishing returns. My point is that I've been just anxious about what it would take to literally migrate if ever we actually had to. Because we got 7 ,000-ish people in our main channel. Not all of them are obviously present and active every day. I'm sure some of them come and go. Maybe some of them lurk. I have no idea because I don't really have any analytics to our usage in terms of just beyond messages I'm paying attention to. So I just wonder if we ever moved to something else, how much would we lose?

51:02How hard would it be to get even our active people to stay involved? Right. Like would they come with us and would they continue to hang out? Or they'd be like, Zulip, what? Why? Yeah, I mean, I can't promise anything about your specific experience, but we have had folks tell us that when they moved to Zulip, they actually started getting much better community engagement because it works quite nicely for folks who are not around all the time. So, I mean, one kind of category of folks, as you were saying, is there may be people who are lurking or who are just kind of coming by once in a while to check in on something.

51:38and if you're coming to something like Slack, you might see the latest messages. It's going to be not really possible for you to catch up on what you missed or if you're checking out every couple of weeks or every month in an active organization. Whereas for Zulip, if you just want to check in on things occasionally, folks will come in and they'll look at that recent conversations view you maybe saw when you were exploring the app And instead of having to look at sort of individual messages and try to figure out what's going on, they'll just see that list of topics and they can be like, oh, this topic sounds interesting.

52:12Let me jump into that. And so you don't even have to feel obliged to kind of make everything be marked as red or kind of manage your own reds necessarily if it's just something where you're not following every little detail. you really can kind of just skim that list of what's been going on and jump in to the ones that are of interest. And so, yeah, so we've had folks say that, you know, something like an open source project, that it can actually really be great for community engagement because people can select the parts that are interesting to them and just follow those and jump in on those.

52:45You can even configure notification. There's a concept of following topics. So once you've seen something that's interesting, if it's a community you're not engaged with very regularly, you can follow that particular topic and, say, get email notifications when there's more messages just to that topic. And so there's really ways to follow specific conversations and find things to engage with for occasional users in a community. I do like how you can set your Zulip to public as well. Can you do that on a per-channel basis? Yeah, exactly. So this is something that there's an overall organization setting for whether you want to have public channels as an option.

53:23So, for example, some businesses might not want to share anything and they just want to turn that all the way off. And then, yeah, for any given channel, you can configure it to be kind of public for logged in folks, private or public even without logging in. So, yeah, what you guys were seeing in the development community is a bunch of channels that we've marked as completely public. And then, yeah, you can just kind of come by and not have to log in and just view messages there. And then, of course, if you want to participate, then you would create an account and log in and post. Now, are those public channels, do they get indexed by search engines?

54:00They don't. We do have a way to, a tool for exporting your Azure data. And then you can get that indexed by search engines. and kind of an archive of all the messages. But it's actually kind of a major technical, the reason is it's a major technical project to make those indexable by search engines. And we just haven't had a chance to prioritize that project yet. But that's definitely on the radar. But it just requires quite a bit of technical work to make that work. Yeah. I mean, that'd be pretty cool for public ones because then it would double as an indexable forum. For sure. Because a lot of those conversations become kind of canonical.

54:39resources, or they could be, but they are lost to the ether. But if they were actually indexed... Yeah, one thing we do a lot is linking to conversations. So you can link either to a conversation or even to a particular message within that conversation. And so, for example, when we file an issue for Zulub feature, we'll generally link to a conversation where we had some initial brainstorming discussion of that feature. And so when folks are working on it, they can get that extra content and context, and then also So if they have a follow-up question, they can just pose that question in the same conversation and continue from where it left off.

55:15So that linking does make some things easier to find. So one thing you might not know, Alia, about Adam is that he is an avid home labber. And so what would a migration look like to the self-hosted? If Adam were to become our system administrator and run our Zulip community out of his home lab, what would that look like? Yeah, so it's pretty similar, except for you would skip the part where you email us and then you would do that for yourself. One last step, even easier, Adam. Exactly. Yeah, we have an installation guide that is pretty straightforward. We really do work hard to make it easy to self-hose Zulup and also make the installation process as easy as possible.

55:59Really smooth upgrade process when the new version comes out. So it's definitely a priority for us. and there's detailed documentation on how you need to do everything. So it should be very doable for you if that's something that you enjoy. You just have a Docker image? Yes. Sorry, this is not the part that I personally work on nearly as much as some other things. All good. But yes. You hear that, Adam? They got a Docker image. Okay. And what aspects of Zulip Cloud, the hosted version, are completely inaccessible to you as a self-hoster? Are there specific features that you will never be able to use in self-hosted, or is it all there, but you have to worry about backing it up and making sure it's up and all that kind of stuff?

56:46It is all there. So Zulupe is 100 % open source. There's nothing that we're locking away from self-hosters. If you self-host, this is the one thing that... So we do offer paid plans for self-hosters. You don't have to sign up for one, but they're an offer. And the kind of two major things that we're providing with those paid plans. So one is mobile push notifications. So the way that App Store policies work, both on Android and iOS, is that if you have a mobile app, which apps are also 100 % open source, but you probably want to use the app that we put in the Play Store or the App Store, rather than kind of rolling your own, which is a whole thing.

57:30And so the way those app store policies work is that a single app can only get push notifications from a single server. It's kind of like an anti-spam security measure on their end. And so for your self-hosted server to send notifications to the Zulip mobile apps, what you do is basically bounce that traffic through our server. And so that's a service that a lot of folks who are self-hosting choose to pay for as part of our plans. And then the other piece is just support. So if you want any kind of support with running your Zulip server, so there is community-based support in our development chat.

58:13So folks do come by and get some help there. But if you need SLAs or if you need something more than just asking a question on chat and seeing if folks are around to reply, then we do have support offerings as well. So those are the types of plans for self-hosted organizations. I did find a repo, and I know that you may not be able to go deep on this if you can. It's okay. It's on your Zulip org on GitHub. It's docker-zulip. Zulip. So I assume this is official. It's container configurations, images, etc. for all of it. There's a Docker Compose file there. Yeah, so I guess the way it's described in our docs is it's an officially supported experimental Docker image.

58:59Official yet experimental. So, you know, tread softly, but officially. 102 lines in this Compose file. I mean, that's a lot of lines. So you've got SSL certificates set up for folks. You can set up a custom MCA certificate if you want to. You can point to a different Git repo. So you can point to the official where you can have your own fork, which I think is pretty cool. And you're just a doc compose up away from running Zulip locally. Sounds pretty awesome. There is also an architecture document on your docs, which I found to be pretty good at describing the way the whole thing works and the various parts, Postgres backend, they're using Redis and Memcached in certain areas.

59:40it's a Django web app for the backend. And then there's a single page app, which is written in TypeScript, probably React, I'm not sure, for the web in-browser experience. Obviously the mobile clients you mentioned are getting rewritten into, did you say Flutter? That's right. Yeah, and so they're all using that same backend API. Now if you're self-hosted and you want to connect your phone app to that, are you just basically saying like zulip.changelog.com? Like would we just create a... Yeah, just when you sign in, you put in that URL for your server and you're good. Wham, bam. What do you think, Adam?

1:00:18You want to dockerize us? Well, see now, that's a great question, obviously. But now you have to be your own uptime for your own chat apps. That's the high price of self-hosted. That is the high price of self-hosted. I would want to compare Zulip Cloud and other ways first But I'm not against the idea of self-hosting. I just think it takes a lot of responsibility to do so. Yeah. I assume, how then, maybe you answered this already and I was reading docs or the doc composed file while you said it. And if I missed it, I'm sorry. No worries. But how does like your iOS, Android app work with a self-hosted scenario?

1:01:03Do you point it at like a URL kind of thing? Like if I was... Yeah, yeah, I asked that one. Yeah, exactly. Yeah, so when users... Asked and answered. No worries, yeah. So when users log in... Repeat yourself, please. When users log in, they'll just put in the URL for your server, and then they're good to go. You just CNAME a subdomain, and you're good. So self-hosting, yeah, I mean, you would have to have, even if it was literally self-hosting in a closet, or self-hosting on DigitalOcean, render those two that are mentioned in your docs. We obviously prefer Fly, fly.io. Not paid to say that, but just definitely very passionate.

1:01:39So I guess we can self-host on Fly, right, Jared? We wouldn't have to. 100%. You can self-host anywhere. I just thought it'd be cool to run out of your closet. It would be cool, except for, I think, I don't know if the uptime would be as good. I mean, the ping, the latency, Garehub may have opinions about it, that's for sure. It's just chat, you know? It is just chat. Worst case scenario is we can't send each other memes for a few hours. I mean, we've had folks self-host Zulip air-gapped like on a ship. Oh, really? Where they weren't going to have connectivity with the wider internet, just as their chat within that ship community.

1:02:11So if we ever decided to travel the world, maybe on a sailboat. Yeah. Like our friend Alex McCaw did. We could have Zulip on that sailboat with us. That would be cool. Self-contained Zulip and I guess local area network only, right? Yeah. Might not be required if you have five people on your boat. You could even go local machine only. You could unplug that machine from the whole internet and have Zulip just on that machine if you wanted to. Truth. Truth. Not very useful that way, but you can do it. Via the terminal. The terminal app even. Yeah.

1:02:57What's up, friends? I'm here with Kyle Carberry, CTO at Coder.com. So, Kyle, I've known Coder as the IDE in the cloud and over time, you've iterated to become a fully open source cloud development environment, a CDE. How do you explain what Coder is and what it does? Coder is a platform to provision you a development environment on any cloud infrastructure that might be in a VM, that might be inside of a container. But Coder is kind of a developer's route to provision infrastructure for them to write software inside of. We started with the IDE, which is kind of like putting VS code in the browser, which is what most people are certainly familiar with us for.

1:03:34And we kind of funneled that into more of a platform where people provision the infrastructure. And a lot of people do use a web ID with Coder. A lot of people use a local IDE and just connect in. Okay, so what are teams coming to you for? Who's coming to you? What people really come to us for, particularly this problem is really exacerbated if you're a large enterprise, is when you have like 500 engineers that are trying to update like a version of Python, and instead we allow one engineer to go through that tedious work of updating some scripts or some Docker container, and then you can actually just deploy that in one click to say like 500 engineers and make it really, really simple.

1:04:08Let's laser focus in on the platform engineer. It is that team's job to provide the best infrastructure, the best platform for their given applications, for their teams. What are some signs or signals for platform engineers to think about when it might be time to consider a cloud development environment like Coder.com? So as a platform engineer, developers might constantly be opening IT tickets that their computer isn't working properly. They might constantly want to update dependencies, but that's a big mess. You constantly have to email people across your team to say, hey, Adam, could we update from Java 17 to Java 18?

1:04:46Those are the kinds of problems that people typically have. That's the status quo. You ship people more powerful laptops to improve the build times of your projects. You try to reduce the complexity of your products instead of simply leveraging better hardware. We believe that the future is leveraging the cloud for a lot of these things. You can get more powerful instances in GCP or AWS that can make the build times faster instantly. you can let one developer create a standardized environment and then distribute it to a thousand so that when you're updating from Java 17 to 18, it's just a simple pull request.

1:05:17You can co-locate your servers right next to something like S3 or a database they're using in development so that you get immediate data transfers and it's not slow. Many of our customers, which is a crazy thing to say, but they use absolutely massive monorepos and they get clones that go from like 10 minutes or 20 minutes or an hour to simply like a minute or 30 seconds. It's just a lot simpler when all of your engineers are standardized on one centralized piece of infrastructure, and then one person can impact the lives of hundreds of engineers. And with that, we don't believe that everything belongs in the cloud.

1:05:49We think that some workloads are really amazing for it, and some are absolutely terrible. Coder should be a self-serve offering to your engineers. It should not be prescriptive where you migrate all pieces of software development into the cloud. Only the things that really get a lot better by running them in this cloud-native way do we really promote moving. Well, it might be time to consider a cloud development environment. And open source is awesome. And Coder is fully open source. You can go to Coder.com, get a demo or try it right now, or even start a 30-day trial of Coder Enterprise. Once again, Coder.com, that's C-O-D-E-R.com, Coder.com.

1:06:48well there's a terminal app i haven't seen visuals of this yet how how cool is this terminal app have you seen it i'm excited for a terminal app i think that's a it's very hacker i like that yeah if you go to zil.com slash apps you'll get it you'll see a link to it There you go. Okay. Terminal beta. Cool. It's very TUI-like, Jared. Obviously, it's an application. I do like TUIs. It's an official terminal client written in Python. Seems like Zulip is almost entirely written in Python, except for that Flutter part and that web app, of course, has to be TypeScript. But you guys have Python roots.

1:07:25Yeah. Zoolop is one of the first kind of major projects to be using MyPy static typing in Python. So we're engineers. We're part of developing that. Awesome. I'm just staring at your terminal UI now. Same. I've seen a squirrel and I've become distracted. I forgot to continue talking to you. What I'm seeing on the side, though, if I can talk through a little bit and see if you're following me, Jared, is that it seems like you've got the channels, of course, and it seems like those are topics beneath it potentially. Obviously, it's not as full-featured as an actual web UI or an application UI. Do you find that people actually use this terminal app a lot?

1:08:09Is it one of the primary client set that you have in your stats? What do you think the usage might be? I don't have a number handy for you. I mean, folks do use it. Definitely not as much as their other clients, but it, you know, for sure. I guess sort of philosophically, I would say one piece of it is that, you know, we've talked about how, just how much time folks are spending in chat. And so having that chat experience feel pleasant and natural and sort of do what you want, I think is really, really important. Like you don't want to be like annoyed and frustrated by something in an app you're using, you know, every day.

1:08:46And so we do believe in like giving folks flexibility and options and configurations and different ways to experience Zulip that sort of matches well with their workflows. And I would say having a terminal app as part of that, just like for some folks that is really like the natural way for them to engage with a piece of software and it feels really smooth and kind of how they want to experience it. And so I think that's really valuable just because people are different. We can't make an app that is just one way and works perfectly for everybody. There has to be flexibility for folks to engage with it in different ways.

1:09:26If we can use this GitHub repo as a proxy for usage, I would say there are people using this. It has over 600 stars, but most notably 871 merged pull requests and 165 open pull requests. So people are working on this. People are collaborating on this. And of course, people only work on and collaborate on software if it's useful and being used by folks. This is not an afterthought. This is very much an officially supported thing with 77 contributors. So pretty cool. Yeah, and we had multiple interns working on it this summer. So yeah, it's definitely interactive. That's awesome. Tell us about the team.

1:10:05Tell us about the company and all the people involved. Yeah, so we have a pretty small kind of core team of folks who are paid full time to work on or full-time or part-time, I guess, to work on Zulip. And we do think that's really important kind of as part of our model that there is a team of really kind of really talented expert engineers and other folks who are, for whom this is their day job. It's really hard to run a project where it's kind of a side gig for everybody. So with this core team, we've also invested a lot and to making it really easy for folks to get started contributing to Zulip.

1:10:48So there's been a huge amount of investment into creating the space for a really active, really lively community around it as well. And that comes in terms of tons and tons of documentation. I think you saw some of our production documentation. There's also tons of contributor-side documentation from, as you mentioned, how systems work, but also just the contribution process, what a good pull request looks like for us and everything about that process. And that's really something that we put a lot of thought into. What is that process of contributing and how do we make that a really excellent experience both for us in terms of reviewing the work as well as for the contributors themselves and make that a really great positive experience, great learning experience for folks.

1:11:34So for example, with a team of, on the order of 15 paid team members, we had 124 people contribute to our last major release. So that's like about a six-month cycle. So it's a lot of folks who are either doing, some of them are doing kind of a formal internship program with us. We've been participating in Google Summer of Code for the past, for a number of years now. I don't know if you're familiar with it, but basically Google funds internships for open source projects, as well as kind of managing that overall structure of helping folks find projects to work on. So that's been amazing for us.

1:12:13We have generally, most years, we have about 15 to 20 interns, most of them mentored by kind of alumni of the program or other community members. And that's been another really great way for us to bring folks into the community. But it's, you know, Zoop is open source, not just in the sense of like the code being open, but really just in our whole model of how we develop the product and how we engage with contributors, how we engage with our users. One time, I guess one of our folks who joined recently, he started out as an intern and then joined as a full-time team member. And he commented that he was surprised when he got added to kind of all our private company channels, just how little traffic there is in those channels.

1:12:58He was thinking that when we were giving him feedback on things he was working on, maybe we're somewhere off on the side discussing that amongst ourselves and then like providing the summary version he was like oh wait no that's not how it works i was like no no no yeah we if we're talking about how the product should work we just talk about that in the open and you know that way everybody can kind of see understand the decisions can can contribute to the decisions like yeah we're like non-hierarchical in terms of it's really about what your ideas are and how clearly you communicate them and explain to them, not what your title is or how long you've been involved with Zulip or anything like that.

1:13:38It's really about kind of working together to come to the best decision we can about how something should work. Yeah. Let me know if I didn't quite answer everything, all the parts of your question. Has she answered your question, Tara? Yeah. Okay. What's stopping you from, or have you considered raising funds? I know you had grants in the past, but I'm not sure what your angle is. I mean, there's obviously this idea of commercial open source companies out there. We're very anti-Rugpool, not cool here around these parts, which means don't change your license once you've gotten to critical mass because it's against your future business objectives.

1:14:19Hopefully I paraphrased that well enough for you, Jared. I think there's an opportunity. I'm just curious, have you? Why haven't you? What's the status on that front? Yeah, absolutely. Yeah. So we have intentionally not raised VC money and do not plan to raise VC money. And what we're in terms of the business model, what we want is just to build a sustainable company on top of this open source project. So we've discussed some paid plans we have on the cloud side, on the self-hosted side, you know, services we can provide. And so that's really our strategy to have our users pay for the software.

1:14:54And then that funds the development of the project and the product. And kind of a key reason we don't want to go the VC route is that we feel that kind of misaligns the incentives. There's a misaligned, kind of an inherent misalignment of incentives. So for us, we're not going to take 100 swings at this. We're not going to try to build 100 different products and see which ones land and abandon ones that don't. We really are building Zulip because we think it's a better way to work. And we're really, really committed to making that around for our users for the long term. So as I mentioned, we still have users from 2013 who are on Zulip now.

1:15:34And we want that software to be around for the long run. And so we want to just take that one single bet and make it work. Whereas VCs, their incentives are, you know, they're looking for like the next, you know, your next Facebook, your next like giant company that just explodes. and they're willing to take big risks in order to have that probability of a really remarkable, amazing return. Whereas for us, we want to take very small risks and have a very high probability of success without necessarily aiming for that galactic outsized return. Our main priority is really to get to a point where the software, we have enough, we're making money to really continue to develop the software and have the staffing and the team that we want.

1:16:25And it doesn't have to be stratospheric. And of course, we would like to reach as many people as we can. And we think it can benefit lots and lots of different kinds of organizations. It's a huge market. There's definitely tons of opportunity. But just the kinds of risks we're comfortable taking to get there are very different from the kinds of risks VCs would feel comfortable with taking to get there. What if that's not true? Which part? all of it what if there are venture capitalists that align with open source which is becoming a thing what if there are venture capitalists that see your idea as the way and they want to fund companies that have prive by cold at hands aspects to open source would your would your tune change well i think it's not just about open source like i i think there are now starting to be vc firms that are focused on open source and and really buy into that model but it's also just kind of the structure of how and how you do that investment right so do you try to like hire up really quickly spend tons of money you know in marketing even if it's the return on is not there but just to get that growth curve you know like what are you what are you trying to do right and like what is your strategy to get there you know I'm not like I'm not going to tell you 100 % never in the next 100 years will Zulip take VC money.

1:17:49Whatever. We're a small company, right? Like we do to some extent, like kind of like make decisions about things when we need to make them, not, you know, planning things for 50 years ahead. But just that has been our kind of strategy so far. And we have not been approached by a venture investor who we think would be completely different from all the other venture investors such that we would start thinking about it. I think the reason why I come with those questions is less to challenge you by any means. It's like zero about challenging. And it's more like if Zulip is the best and it is open source and it is superior in so many ways, in so many models even, of how you can use the software, not just in your cloud or in the self-hosted version, the exporting, the non-fettered access to it to be able to move and all those things.

1:18:45If it's superior, I would want to, if it were me, I would want to do all I could to ensure everyone could use it more. And the way you get there is generally the reason why people raise money is not because they literally just want money. It's because they can leverage that money as a resource to go faster to the roadmap. And we talked earlier about Flutter. We talked earlier about some different areas and maybe you're slow and steady and that's okay. And there's nothing wrong with that. I just wonder if a little funding that was in alignment with your morals, values, et cetera, towards open source, the way you run your company, if that money didn't challenge those values, if things would change.

1:19:27Because if you truly are better, and we've seen even in our own Slack, a person say infinitely better than X. So we hear that ourselves even. If that's truth, then I would want to do all I could to get that truth to many people. Yeah, and we're definitely, so we're not currently raising money, but we definitely are currently exploring sort of kind of different strategies on the go-to-market side. And that's something that we're thinking very actively about, the sort of how do we increase that reach and grow faster in terms of, you know, kind of finding different ways to introduce folks to Zulop and to reach more people.

1:20:05So it's definitely a major priority for us right now. Yeah, that has to be one of your biggest challenges is like nine out of ten people don't know who you are. Yeah. Right? Yeah, yeah. No, it's true. It's true. Yeah, it is a major challenge. No offense, but I mean, even most things, nine out of ten people don't know what it is. For sure. Yeah, yeah. Yeah. I mean, there's tons of things we're trying. I like the free for open source education, et cetera, that you already discussed. What are some of your other ideas? What are some of the things you're thinking of trying to get more people to know what Zulip is, to make Zulip a household name?

1:20:37I mean, some of them are kind of standard things, like paid advertising, going to conferences and various kinds of events and sharing Zulip that way. One thing that, another direction is kind of content. So we've had blog posts on various topics. We're starting to, you know, one of the things that I talked to, you can probably see my excitement about is this kind of side of community management and getting folks engaged in an open source project. So, for example, like we're working on some partnering with some organizations on blog posts around that kind of thing. And so just kind of getting the name out there in whatever way.

1:21:19because I think, as you were saying, the brand recognition and just awareness matters so that when not everybody's all... People aren't constantly in the market for a new team chat, but we want to be top of mind when they are starting to think about it and when it does come up. But yeah, I would say we don't necessarily have something unique other than we do have this open source angle and things engage in the community and the open source community more broadly and sponsoring OpenStars projects is definitely one angle for us that we're investing in. Well, it's one of the hardest nuts to crack.

1:21:57And everybody out there is trying to crack that same nut, aren't they? And so there's a lot of noise. There's a lot of competing voices. And you definitely have a lot going for you. I think leaning in on community and open and I think moderation, as Adam said earlier, as you guys continue to flesh out the product, those are all good strategies if there was a magic carpet that you could go on that would automatically get you to brand awareness of course we'd all just hop on that magic carpet but in general our style is just try to be really as clear and direct as we can that's really our focus for all our marketing and so on we think the value is there for folks and if we can communicate that clearly we don't need to get super marketing, super sales just tell folks what's there very cool Adam anything else from you just to add on to what you're saying here I think probably without digging into the data I will hypothesize that probably the biggest challenge first is awareness that you exist and then obviously once they realize you exist you know the opportunity for superior feature sets then I would say the very next thing is like okay now what which is our requests for information on hypothetically what it would take to move, what it would take to go from a Slack or a Discord.

1:23:21I feel like if you could do content around that subject, not just documentation like how to, but like good stories of folks who've moved and their journey and to demystify the scares and concerns. Like my main scare is, is that a proper adjective? I don't know. I'll allow it. is that, or I guess anxiety point is, is will we lose the people that we have in our community? Will they, will they bounce? You know, if you can showcase the, what's on the other side of the, of the wall, rather than me assume as somebody who is not happily, but happily using Slack, given the things we've already said, still like Slack, it's still amazing.

1:24:02It's just, they got warts for people like us, communities like us. I feel like that's the content I would personally, I would look at the data And I think that would be the hypothesis. Get awareness, show off the amazing feature set that really captures 80 % of who likes you most, and then show how easy it is to move. And almost make it like, you should be doing this. Like, it should happen today. We can help you. And if there's money to invest, in quotes, money to invest, could be time, could be people, could be people hours, is to guide and assist certain organizations on that path. Yeah, and some of what you described, we do have case studies on our site where a lot of folks talk about starting initially with something else and then moving over to Zulip and sort of that experience.

1:24:49But parts of what you said, you're kind of reading off of the to-do list I was working on yesterday. Just yesterday? Okay, cool. Yeah, literally just yesterday. Yeah, I was thinking, you know, we have some content in our help center about that migration path, but we definitely need more clarity on just kind of bring all those pieces of information together and like coming from different kinds of tools here. Here are the steps you take and just, yeah, like folks are busy. There's a lot going on here. You know, we can make that easier for people. It can make a big difference. If I had to divide my time up into fifths, I'd take two fifths of that time and dedicate that kind of content.

1:25:28If not more. Because fourths is like whatever, you know, like 25%, 25%. I mean, that's pretty easy. Like one fourth. Yeah. I feel like two fifths sounds better to me. Two fifths of my time would be focused on awareness and showing off the better world, the FOMO. Yeah. You're missing out. Yeah. On freedom, control, access, enjoyment. Privacy. And then obviously your dev team and engineering teams can be focused on all the surface area, Flutter, that migration, finishing out those applications, polishing the peripherals. Your dev team does a great job on documentation compared to what I've seen in a lot of projects.

1:26:09We see a lot of open source projects. The documentation is really good. The read-mes are very deep and detailed and organized, thoughtful. And so obviously you want your dev team to be devving. That's what they're there for. But as much as they can write about what you're doing technically, decision-making, architectural stuff, not just in documentation form but in content form, I think that would pay off dividends as well and obviously can also double as documentation in a certain way. Cool. What's next? Yeah, exactly. What is next? What's next? For you, listener, are you going to go to Zulip.com?

1:26:49Got the.com. It's a big deal. It is.com. Yeah. It is a big deal. It's a five-letter.com. Free, open source, cloud, or self-host, unfettered. Do it today. And if you think we should switch to Zulip, hop in our Slack and tell us. We'd be happy to at least try that Docker image. I mean, I'm going to give Adam a to-do, you know? See if you can get it running on Docker. on your home lab or fly and just toy around with it and see how it feels. Try it on for size, you know? Or if you want to just try it, it literally takes less than two minutes to create an organization's little cloud and then you can just poke around and experience it for yourself.

1:27:29It's almost too easy, Adam. It's almost too easy. Yeah, I feel like we should try cloud out first. And if we like how it feels, take the next step. Yeah. That's half the battle, right? Because sometimes that switching of the UI and everything, it can be jarring the ideas and the features that may be there but maybe maybe it feels weird I don't know and then give us feedback that's the other thing if there's anything that feels off or feels confusing just come by the development community and tell us and we'll try to fix it well very cool well thank you for this time thank you for going through all the details with us it was awesome thank you for the great set of questions

1:28:09in a world where open source is eating software faster than software is eating the world, there's Zulip, the open source chat that has the potential to unseat the giants, to at least unseat the giants based upon features that really matter to users. And the thing is, is that they have so much potential. What exactly is potential? Potential is kinetic energy stored waiting to be released. and after this conversation I'm so hopeful for the team at Zulip but at the same time I know it's kinetic stored energy potential not realized now there are a lot of people who love Zulip and there are a lot of people who don't even use Zulip or even know about Zulip but now you do so what are you gonna do well I say go to Zulip.com check it out try it out self-host it use their cloud contribute be a part of the community all the things that open source provides Now, I, for one, am very hopeful and very happy Zulip exists.

1:29:13But Microsoft, Slack, Salesforce, they're massive. And so they need us to step in, to use, to try, to contribute all the things. Well, make sure you check them out, Zulip.com. And Ali was inspired by this conversation to create a brand new guide called Moving to Zulip. And that'll be linked up in the show notes for you. Okay, sponsors for today, big thanks to Sentry, Sentry.io. Use our code, Changelog, to get three months, almost four months of the team plan for free. Again, Sentry.io, and the code is Changelog. And also to our friends at Fly, we love Fly, Fly.io. It is the platform where you can do pretty much anything.

1:30:03and Tigris is one example of that. Check them out at tigrisdata.com. We're using it and we love it. And to our friends at Paragon, useparagon.com. All these SaaS integrations you need for your B2B SaaS. Again, useparagon.com. And to our friends over at Coder, coder.com. Cloud development uncompromised. they're the number one self-hosted cloud development environment out there I checked it out I think it's so awesome what Coder can do check them out Coder.com and of course the Beat Freak in residence Brake Master Cylinder bringing those beats every single week much love BMC much love ok so no bonus today but I do want to mention cause hey why not changelog.com slash plus plus it's better it is better today it's not better because there's no bonus but hey other weeks other shows bonuses galore for our plus plus subscribers that is where you go to get the ad free version of our show the way to directly support us to get closer to that cool change law of metal get bonus content not this week but hey you know next week and i know because we recorded next week's show today and we have a very lengthy very awesome bonus content for you you'll love it if you're a subscriber once again changelog.com slash plus plus okay that's it this show's done thank you for tuning in and we will see you on friday

1:32:12Game on.