We See Dead Projects  ## Metadata - Author: [[The Changelog: Software Development, Open Source]] - Full Title: We See Dead Projects - Category: #podcasts - URL: https://share.snipd.com/episode/31712547-ed1d-4ad3-aad7-fa4530ffce78 ## Document Notes ## Highlights - **Recording Mishaps And Lost Conversations** - Adam Stacoviak once recorded a 20-minute interview and realized he never hit record, but the guest had a local backup. - He repeated the mistake later and lost 45 minutes with no backup, highlighting recording fragility. Transcript: Adam Stacoviak So I got a confession to make. Okay. One time. One time. When I was doing an interview for this show. Okay. I went about 20 minutes into the show. And I realized that I did not hit the record button. Gosh. And it was a guest. I had just met that day. And we were grooving. Oh, man. And we were having a blast. Then i stopped and i said i am so sorry because i've done this how many times we've done this just so many so many in fact i think my personal episode count has broke the 1000 mark at this point And i think you're getting close to that i wouldn't doubt it yes and i said i am so sorry i'm not gonna i'm not gonna dox him and i said but i have not been recording this now i said hopefully You're recording a local of your own and he said i am and i said i'm recording a local of my own and he said okay i we're good then. And we stopped and made sure that he was recording and it was good and you could hear him and mine was recording and it was good. ([Time 0:02:13](https://share.snipd.com/snip/3ba34b0c-3de3-442f-a707-c6c4aece0148)) - **Always Verify Recording State** - Use recording platforms that surface clear recording-state notifications so you don't lose interviews. - Check backups and local recordings immediately after starting to prevent silent failure. Transcript: Adam Stacoviak Should tell you this though right like this should be a great feature like you're getting dramatic are you recording right now it's similar to like how your watch says are you working Out yeah like hey are you doing a run and you're like no but sure i'm just taxed on this walk okay don't make fun of me i know it's kind of insulting sometimes you're like i'm not actually Working out i'm just a little bit out of shape and i'm breathing heavy okay but when it does get it when you do forget to start your workout it's pretty nice it is nice and had we got a notification Even like three or five minutes into that diatribe we would have been so happy so happy because you can't just go back and redo the same conversation no we can't. Jerod Santo We have to leave that one in the ether for you guys. It was really good. It was so good, you guys. You should have been there. You should have been there, okay? You should have been there. Okay. All right. Let's do a new show then. Adam Stacoviak Let's do a brand new show. Setting all of that aside, which none of you guys know what we're talking about, but that's fair. We do. If you see it on our faces right now. You can't know what we're talking about. ([Time 0:04:40](https://share.snipd.com/snip/ea17f518-79c6-4c0b-9401-8cdf6738e1bc)) - **Corporate Priorities Kill Projects** - Many beloved projects were discontinued because companies reprioritized or couldn't monetize them. - Google killing products like Reader or Wave shifted broader adoption and cultural momentum. Transcript: Adam Stacoviak If I'm like a little white, like a ghost. That's why. Today's episode is all about death. Oh, yeah. So much death. The death of projects. The death of open source projects. The death of closed source projects. This show is called I See Dead Projects. And it is an episode request from a good friend, Thomas Eckert, who requested it last year for last Halloween, and we didn't do anything about it. Dreams come true. Now, Thomas, you get what you asked for a year later. Find some of these stories of open source and other projects that have gone by the wayside, whether they were abandoned, whether they were beat out by competitors, whether money destroyed Them, or maybe they were just finished, whatever it was, find some stuff and tell those stories. Now, we're not going to tell you all about Quicksilver. Can't do it. Because we just told each other about it. It hurt too bad. And it hurts too bad. It hurts so good. But that gives us more time to talk about other things that aren't Quicksilver. And I got off onto a tangent while researching this. I got off onto a killed by Google tangent. Because when I think of the death of many pieces of software that many people have loved, don't you just think Google's murdered so many of them? Well, find your way to killedbygoogle.com, where we have a 3x3 grid maintained by Cody Ogden on GitHub, an open source project that loads a cool website from a JSON file with 298 things That Google has killed over the years. I haven't even built that much software in my life. ([Time 0:05:41](https://share.snipd.com/snip/c22a60e7-6cfa-49e4-8c33-baed08cf8524)) - **Tools Can Outlive Their Purpose** - Sass evolved into CSS features and implementations like Dart Sass, making older engines obsolete. - LibSass reached end-of-life as the language matured and native browser CSS improved. Transcript: Adam Stacoviak Oof. Jerod Santo Well, didn't fully prepare for this one, mostly let's just say mostly okay i would say you know there's been a lot of advancements on the front end one in particular that has evolved naturally Is css it it lagged for a bit there i think there were things like sass and gosh, there was something else. Less was a thing in my brain just now. Don't know about Less. Never really caught on personally for me for some reason. But the original Sass was created by the person, which I can't remember his name in the moment, but he's a close friend. I'm so sorry. Created Hamill, which was popular to Rubyist back in the day. It was white space aware, similar to Python is. And the original SAS syntax was just that it was a pre-processor. Yeah, it was a pre-processed before you got the actual css it's kind of cool done a terrible job of explaining it but sass to me i don't know this is true don't be offended if you're still Working on this it's kind of like dead or dying you know it's evolved it did its job just like quicksilver didn't hear about that now, did you, listener? No, you didn't. Ah. That's okay. Just like Quicksilver, we discussed that in the Ether show, so to speak. Quicksilver enabled this first mover phenomenon to help Apple essentially bake a lot of features in that Quicksilver launched with, got to bring that pun back. And then SAS did the same thing, I believe, in a great spirit to CSS. Like the goal was never let's make SAS the thing. It was let's make CSS better by adding SAS to it. And you pre-processed this, you know, this superset of CSS is what Sass was. It's set above and on top of CSS proper. And it gave you a lot of superpowers you just didn't have on the front end. Adam Stacoviak And I don't know where it's at right now, honestly. Well, I went to sass-lane slash blog, and it's officially dead. Who is it? Posted the 23rd of October, 2025. This was five days ago. Okay. Libsass. How timely. Great timing. Jerod Santo This is breaking news. Adam Stacoviak Yeah. Sass is end of life. Libsass has reached end of life. Libsass and the packages built on top of it have been deprecated since October of 2020. So that's five years ago. In the five years since we made that announcement, the Sass language has continued to evolve to support the latest CSS features like color spaces and embedded Sass, made it easy to run Dart Sass across numerous different languages and platforms. Dart Sass now meets essentially all the use cases that LibSass once did. At the same time, development of Libsass has faltered. There hasn't been a new commit to the source code repository since December of 2023, and there are numerous issues languishing unaddressed. The time has come to be clear. Libsass is no longer maintained and will receive no future updates. So that's Libsass. ([Time 0:13:38](https://share.snipd.com/snip/99070e9a-8610-4932-ae99-45a16c1c2e95)) - **Prefer Native CSS Features** - Use language features in native CSS when possible instead of relying on extra preprocessors. - Migrate incrementally as standards add variables and nesting to reduce tooling debt. Transcript: Adam Stacoviak What you really wanted was nesting and variables. And then it got fancier from there. And people who are, you know, hardcore front-enders and designers who love CSS and all the features, you could, like, use it as a polyfill, basically, to get new CSS features before They existed. Precisely. And that's what the attraction was, was that. Totally. And now most of that's pretty much built in. I mean, yeah. Variables and nesting. I'm a happy camper. Like I don't need much else. Jerod Santo Includes and imports and a lot of that stuff was all, I mean, hats off to everyone involved there. Natalie, I think there's Chris, I forget his last name in the moment. These were dear friends of mine over the years, just paid attention closely to that. I mean, thankfully, all that work went into it because it all poured into eventually a better CSS world and it's no longer needed. ([Time 0:19:56](https://share.snipd.com/snip/e1e0d3ac-7d50-4296-8e05-51e9711e5503)) - **AI Finished My Feature While I Pondered** - Jerod described asking an AI (AMP) to 'think' about a feature and returning a finished implementation instantly. - He found the agent completed work faster than he could deliberate, changing his workflow. Transcript: Jerod Santo Ai back then yeah like the bad old days yeah like why would i write that code it would do it for you okay it would do it for, today, I took a little break. I just want to think about a feature. I just want to think about a feature. Here I am in AMP, and I'm like, hey, AMP, can we talk? Can we talk? I got an idea for the future, and I just want to think about it with you. Man, I dropped that idea in there. I'm not kidding you, dude. I got an email, went and read it, responded back, was happy. Okay, let me get back to my anthem. Let me see what they did. Feature was done. I said, let's think about it, okay? Let's talk about this. Let's not do the work. Made a bunch of changes all correct and stuff. Shipped the production and everything. Gosh, in a blink of an eye, I meant to think about it. The feature was done. What a world. What a world. That's not even a joke, man. That's not hyperbole. That's true. That was not fabricated for podcast material. That was straight up truth. Adam Stacoviak I just got a little bit concerned when you started talking to AMP, kind of like a Marvin Gaye style voice, you know. Jerod Santo Let's talk about, let's think was being i really was like can we just i know you're a doer okay i know you're a doer amp you just go and do stuff can we just talk about it no i'm a doer amp says I'm gonna go do this i can't plan i only do it plans well. It's got the Oracle comes back with good stuff. But I went and got that email taken care of, came back, and it was done. And I was like, all right, cool. I was kind of upset because I really wanted to think about it, but it wasn't necessary. It's like, gosh, man. Just want to think about that feature for a second. No, I can't. It's done. No input. Yeah, but was it good? It was a pretty simple feature. It didn't require a lot of thinking. I'm just trying to incrementally inch into this one because I had a whole failed version of it. ([Time 0:25:50](https://share.snipd.com/snip/e785c11e-3015-45d7-a98c-ac3df67a3eb1)) - **Choose Languages For Deployment Constraints** - When unsure which language to use, iterate quickly and port code to the environment you must deploy to. - Compile to single binaries (e.g., Go) for easy deployment on constrained servers. Transcript: Adam Stacoviak Like I used to have opinions. Jerod Santo Yeah. Adam Stacoviak But I do eventually resolve. So for instance, I started up this, it's like static site generator thing that does some crawling, blah, blah, blah. And I just tell it what I want. And I'm just like, you know, this is a cloud code. And I'm like, you just do your thing and come back to me. And it came back to me with a Python script, you know. And I was like, well, I guess I didn't specify. I didn't want to specify. But now I'm looking at this and I'm thinking, yeah, no, not really. And I'm like, let's rewrite this in JavaScript. No TypeScript, please. Just pure JavaScript. Keep it clean. You know, keep it loose. Give me options. No types. No types. Do not constrain me. I'm vibe coding, okay? That's right. And so it ports it to JavaScript in a blink of an eye. And I'm like, that was cool. And then I build out some features and I get going. I realize I want this actually to run on a remote server, which's a shared server, so I don't have access to NPM. I have Node there. It's an older version. I can run it, but I can't run NPM. And of course, this thing Vibe coded up a couple of packages. I think mostly around requests and stuff like that. Maybe like HTML parsing and query stuff. So I'm like, well, that sucks. And so I'm like, you know what? Port yourself to Go. And then, bam, Go. And I said, compile yourself to... Always bid on Go. I like Go for that reason. You just compile yourself to Go. I've got everything in one binary. I just SCP that sucker to the remote server. It's compiled against the Linux that they expect over there on my shared hosting. It even embeds the HTML templates and the CSS and everything. You embed it all in one binary, dude. ([Time 0:29:10](https://share.snipd.com/snip/709161b2-4583-413b-93b1-de55cbdc3699)) - **Momentum Shapes Framework Fate** - Meteor.js rose fast with VC backing but declined when core focus shifted and new frameworks emerged. - Community perception of decline can become self-fulfilling and accelerate abandonment. Transcript: Adam Stacoviak In fact, I know this one still lives in a sense, but it doesn't live in the way that it originally was pitched to live, which was as the future of web development. Yes, I'm talking about Meteor.js. Remember Meteor.js? Often something called meteor is open source it was a full stack JavaScript framework came out around 2011 back in the early days of jQuery backbone node stuff like that initially developed By some folks at Y Combinator. Yes, this was one of the reasons why it's now being declared dead, because it was VC-backed from the start. But built by some cool devs, Jeff Schmidt, David Greenspan, he co-created Etherpad. Remember Etherpad? Rings a bell, yeah. Etherpad was one of the very first things that was like collaborative editing, like Google Docs. And it was open source and self-hostable. And so people would stand it up and they would use it for like their own little, you know, real-time collaborative editors inside their web apps. And it was very cool. I'm not sure what's going on with Etherpad. That could probably be another story there. Matt Dibergalis. I'm not sure how to to say your last name, Matt. And Nick Martin. That was the team dropped huge fanfare. April 2012, largest Hacker News debut of all time, according to this. Probably since then, there's been larger ones, but at least at the time, it was huge. Got lots of fanfare, lots of people trying it. Exploded in popularity. Kind of a typical hockey stick growth kind of thing. Raise more money. Raise more money. This is an MIT licensed project that just keeps raising more money. In July 2012, they secure $11.2 million in seed funding from A16Z. They hire people, they buy infrastructure. By 2013, they release a package manager by Tom Coleman that was actually integrated into the core. There's Discover Meteor, which was either a newsletter or some sort of website coauthored by friend of the show, Sasha grief and Tom Coleman. Of course, you know, Sasha was big into meteor for a long time. Really big. Yeah. And, um, you know, things continued up into the right. 2014, they raised $20 million, Series B round. And they've totaled over $31 million raised. FathomDB comes out, so they have a database, a cloud database. Galaxy, they launched Galaxy, which is like a hosting platform for Meteor apps. So that was kind of the play, was like, you know, it's all open source and free, but if you want it to be hosted by us on our galaxy, that's how you pay money. And performance monitoring. So they started to productize around 2013. Continued to grow. I mean, the package registry had thousands of community packages. Started to integrate NPM in 2015. They got worldwide production apps out there. So, I mean, not unsuccessful by any means. I call this success. For many definitions of that word, things continue to go, okay, 2016, they were making like, exceeding their revenue goals, according to them. But community debates arose. There was purists who loved the all-in simplicity. Others demanded deeper modern tooling integration. So the community started to disagree on the tech itself and the direction that it should go. They started moving towards GraphQL, and that laid the groundwork for Apollo. Remember Apollo GraphQL? That's still out there, which was a spinoff. And then what happened basically was just new JavaScript frameworks came out that were more interesting to more people. And because they were so fueled by VC money, they needed to grow faster than they could grow eventually discover meteor grew outdated you know people started moving on the state of javascript Survey of course is a such a grief thing he's been doing for a long time now so the state of yeah the state of js actually cataloged meteors uh interest and then you can start to see the decline. And sometimes that is a, what do you call it? A self-fulfilling prophecy. Like people start to see it declining in interest and so they are less likely to be interested in it. And just that momentum downwards can actually be a problem. And so the transparency around that actually perpetuates it. But it still got worked on. And then in 2019, all of a sudden, they changed. MDG, which is the company behind it all, pivoted from Meteor to Apollo GraphQL. They saw more money in that, I think. This deprioritized Meteor's core development, frustrating people. And then in October 2019, Tiny Capital, who are owners of Dribbble and other things, acquired Meteor.js and Galaxy, that's their hosting service, assets for an undisclosed sum. So it got sold off and renamed Meteor Software. Got put into maintenance mode. I mean, more emphasizing maintenance than growth. And Tiny, which was the company that bought it, pledged they'd support it for long-term, like five years. Jerod Santo Still out there. Still going. Yep. That's wild to think about that, honestly. Adam Stacoviak They call it a renaissance. Meteor 3.0 and beyond. People still working on it. It has a small community. And so by no means dead. Legacy, perhaps. The influence endures. It helped birth UJS, Apollo, obviously, other things, Storybook. But by no means did it take the world over as it was supposed to. And so in that sense, you know, the original mission of Meteor as the future of web development, it's gone with the wind. ([Time 0:32:26](https://share.snipd.com/snip/7cfc26f5-fe39-49a1-a5c6-1b3a946badf5)) - **Quicksilver: First Mover, Later Forgotten** - Quicksilver launched in 2003 as a powerful Mac app launcher with extensibility and HUD-style UX. - The project faded as Apple and competitors integrated features and users migrated to LaunchBar, Alfred, and Raycast. Transcript: Jerod Santo Take two. Here we go. Okay. All right. Quicksilver, if you're not familiar with this, is a software from your 2003. Long time ago. Okay. That's the beginning of the internet, essentially. It's not actually the beginning of the internet, but it's pretty close to it. No, it's not at all. It's not at all. It's about 15 years after the beginning, maybe more, depending on who you ask, but Quicksilver was an app launcher for Mac. At that time, OS X, not Mac OS. This is how far back it goes. Or OS X, as a lot of people call it. OS X. OS X, yes, unless you're a Mac geek that likes Gab. That was a good one um got lost in that one you just sideline yourself with a reference i just a little little mac geek it's 2003 you're running mac os 10 probably like probably like lion Adam Stacoviak Or leopard or some sort of cat. It was a big cat. Jerod Santo Yes. Who the heck knows which one it was. It was a kitty cat. That's right. And Quicksilver was launching your stuff, man. You wanted to launch things and you would go and you would go to your app directory or you would go to your doc bar and you would click the button for your app. You'd see it bounce. You'd wait for it, and that was cool. No, that was not cool. It didn't bounce forever. No, that's not cool. You had to do command space to conjure this new thing called a HUD, a heads-up display, and it would give you the ability to conjure these things to happen apps to launch math to be done Terminals to be opened finder windows to be discovered is what quicksover gave the world so many things so many things uh it was made in 2003 by a fellow named nicholas ditkoff Made in 2003, released several versions of it. Always seemed to be in beta, basically in beta for 10 years. But it was the darling. It was what launched literally the ability to have these launch bars, these launch apps to do this. It didn't win, though. It only led to innovation being done by Apple absorbing it or our friends over at LaunchBar. Yeah, absorbing it, copying it. Yeah, absorbing it, I guess, is probably not the truth. Copying the features, breaking them into the operating system. First mover advantage did not work out for them. Launch bar, Alfred, Raycast, that's what I use. Now, I did use Alfred. I'm still a semi-happy. How would I be? I'm not sure if I'm happy or not. I guess less than happy because I spent my money. I'm not using the value anymore. Owner of a lifetime license of Alfred. I'll never use it again. Just saying. Just saying. If you want it, I'll give chance that you can win you back. I mean, I just don't think so. I don't. Never say never, though. My buddy Justin told me that, but, you know, I don't see it happening. Adam Stacoviak Okay. Jerod Santo I don't see it happening. Adam Stacoviak What about Quicksilver? Jerod Santo Are you ever going to go back to Quicksilver? I don't even know how i don't even know where to go where would i find quicksilver these days i think it's still a thing but i have no idea i will say this quicksilver was the coolest i mean Adam Stacoviak I'm not gonna say i bought a mac because it, but it was certainly in the calculus. I was like, I want a Mac because I want to use Quicksilver. It was minimally attractive, you know, like it was aesthetically pleasing, but it was also minimal, almost invisible at times. I mean, it's obviously invisible when you don't invoke it, but it had a cool kind of purplish, translucent vibe to it. And then the combination of things you could do, like the actions you could take beyond just, you know, if it was just an app launcher, then I don't think it would have been quite as popular, But it had this like extensibility, extensibility and like combinability that made it really powerful. And you could like take a file, you know, find a file, pass it to a thing, have it output something, transmogrify it, I don't know, capital case it, and then spit it back out the other side. And so the power users and the nerds, they would like show you how many cool stuff you could do with Quicksilver. And that's why I wanted to run it. Now that being said, I only ever used it basically for math and for app launching which you know to this day is what i use all my app launchers for which is why i'm a spotlight user it does Both those things just fine but i wanted the the power i wanted the quicksilver power and so your story is basically competition why didn't it get why didn't it get better why didn't it Jerod Santo Get competed why didn't they win i think it just died in the vine i think it just uh it got there i'm not in beta it seemed to not move and so when something seems to not move and it says beta You kind of think well maybe it's not moving maybe there's no true maintenance behind this thing and so your appreciation and belief in something kind of wanes launch bar comes around Seems to be newer seems to be fresher more mac native feeling same thing i think with alfred alfred is still amazing software. No offense to Alfred whatsoever. I'm not going to use it anymore because, well, I moved on. But I suppose if Raycast died, I'd probably fall back on my free lifetime license. Anyways, we'll see. You could use it again. Raycast, don't die. Adam Stacoviak Well, ClitSilver was open source, though, and these other ones aren't. Yeah, I mean, I guess. Does that mean anything to you? Jerod Santo Does that not sell you on it more or make you want to support it or use it? I think in the altruistic terms of open source, of course, but I'm not going to personally maintain it. So the fact that it is or is not doesn't personally matter to me because I'm not doing the maintenance of it. And I'm not going to care enough to step in. It's not a problem I want to solve. That's why I'm thankful for Raycast. They've done a great job. I think Raycast is by far, they got a great free version of it. You can pay to use it as well. And I think it's the best version of an app launcher. They've got a lot of cool stuff in there. And obviously, they even have, I think it's like, what is it called? I don't even use it too often. But here and there, they've got Raycast AI, a whole chat window. And I believe it ships free with all the free models. Now, if you actually have Raycast AI, I believe you can pay to use other models. ([Time 0:45:04](https://share.snipd.com/snip/4f2ad22f-2efd-43de-91ff-2990d6252db9)) - **Databases Fail For Market, Not Just Tech** - Many database startups failed from poor market fit, wrong optimization goals, or dominant incumbents like MongoDB. - Open-sourcing after company shutdown preserves code, but company failure still halts mainstream adoption. Transcript: Adam Stacoviak Speaking of, how about RethinkDB? Ooh. Gosh, man. Remember RethinkDB? Blast from the past. Slava Akhmachek. That's right. Akhmachet. Akhmachet. That's right. Whichever way you say it, I'm going to say it's right. They're both right. Rethink was a very interesting database. A scalable NoSQL, back in the craze of NoSQL, database with a declarative query language for real-time apps, launched in 2009, was active for probably until 2016. The company actually declared bankruptcy in 2016. The core team was disbanded. The repo was archived. I mean, this one's dead, dead. So dead. They did inspire some community forks, such as ElephantDB. RethinkDB, you'll go back and find that one in our archives, of course. We've had Slava Akhmachet on the show I think probably at the beginning of RethinkDB, and I think at the end of RethinkDB. Jerod Santo At least two. Was that right? I'm thinking two times during its heyday, and one potentially on its wayday. Basia. Oh, I like that. It's a wayday. Adam Stacoviak The heyday and the wayday? That's right. Jerod Santo You're going to go on your way, Basia. You're on your way now. Adam Stacoviak Now, why did R db fail i would only speculate at this point i really don't know my speculation and this is just speculation i just think they couldn't capture market share i think mongodb Was just too dominant in the no sequel database department of the world and rethink just couldn't get there but didn't they write i mean now we're going deep into my brain archives because I'm starting to remember didn't they actually even write up a thing why rethink db failed and didn't we actually do an episode maybe even call why everything could be failed i'm starting To like remember something like this check out our last episode with RethinkDB. What was it called? Jerod Santo Yeah. I'm scouring the internet right now for information. I think the long and the short of it is no product market fit. Adam Stacoviak I mean, that's what a startup guy would say. They had some design choices that were odd, I think, because they had their own language. This is why when the Fauna DB people came on the show. And in fact, I think Fauna DB also at this point, can we consider that one dead? I'm not sure, but I haven't heard about it in a long time. They also created their own query language. And I remember saying like, are you sure you guys want to do that? Because it's very difficult to get people to just use your query like can you have it be SQL like that means that Mongo did it so I mean they're all kind of following the Mongo playbook but Mongo really did succeed where so many others failed. Jerod Santo If this is true founder Slava Apichet said there were two main problems they picked a terrible market and then they optimized the product for the wrong metrics of goodness so they optimized For correctness simplicity and consistency which are all great engineering goals but those were not what the market valued this is also in a day when a lot of things were happening on AWS. A lot of databases were popping up. So we're now past a lot of this. You had CockroachDB. You had a lot of databases come out of seemingly nowhere to solve problems. Some of them open source, some of them not. Obviously the ones who are not just didn't have a chance. And then ultimately, it came down to just competition at AWS and adoption elsewhere. And there's a lot of motion around databases. And so you had a lot of choice. Adam Stacoviak Yeah, there was a thriving market of offerings. Jerod Santo Now we just want Postgres, right? Postgres. Postgres forgres for life man why not you know why would you choose anything else than postgres i just don't know they were eventually acquired by the linux foundation the cncf a few Months after the shutdown in 2016 and it continues to be an open source project to this day but i don't know who's using it i don't i don't know what's happening doesn't everything continue Adam Stacoviak To be an open source project to this day though i mean it's not like the source goes away otherwise it's like you could say that about anything that's dead is this i want to know if this is Jerod Santo Claw trying to get get into my good graces considering what i just said a little bit ago about AMP. At the very end, it says, Why is that? Fun fact. I see one of the discussions mentioned an episode from the ChangeLog about RethinkDB. Adam Stacoviak You folks have quite the archive. Come on, Claude. It told you that? That's what it said. Do you have that in your Claude MD? By the way, I host the ChangeLog podcast. Jerod Santo I think it knows. It knows who I am. Yeah, it's definitely pandering to me because it's like, yo, great job. Adam Stacoviak That is pandering. Yes. But it worked. Did it work? You do prefer Claude now to AMP. Jerod Santo So fauna.com actually doesn't go anywhere which was the location of fauna db fauna.com uh it's either my block list dns hole or it's dead i'm thinking redirected the fauna robotics Which is a different company yeah i'm using safari so mine don't redirect man safari does its own thing okay it really does let me go here into chrome what does chrome do yeah fauna db.org Adam Stacoviak Is what you're looking for uh okay so okay it's a strongly consistent ol2b database with a hybrid blah blah blah uh open source project is still very early stage yeah you're probably Looking at it used to be.com. Well, you're probably talking about the company and I'm talking about the open source and the company might be gone, but the project's still open source because they all are right. ([Time 0:57:13](https://share.snipd.com/snip/1394ab16-10ac-4e2a-a9c7-f3a9b8bd7420)) - **Exploit Free AI Tiers For Low-Stakes Work** - Use free AI tools that trade attention for access to get low-cost coding assistance and documentation. - Leverage daily free allotments for tasks like tests, docs, and code reviews to save time. Transcript: Jerod Santo So you have your prompt area. Then you have the chat area. That's been what's been going by. And right between those two perfectly sandwiched. I think they've done a great job with UI on this too. Is this a subtle text-based ad? It's not even a bother, honestly. And I've actually enjoyed them. I'm like, oh, sweet. They're doing that? Cool. Adam Stacoviak Nice. I find myself actually want to see the ads. It doesn't bother me. I mean, you're getting what you, you're opting into this free thing with like, it's all straightforward the way it works, right? So if you need to use it for free and you're okay with that supported, like I think it's a great option. So I would say there's no shame in that game whatsoever. You know? If you want it to deep think for you, well, ask nicely, okay? Jerod Santo Because it's probably going to be going to do it. Adam Stacoviak It's like, there ain't no thinking here. We just do around here if the whole time it's deep thinking it could it show you like a video ad like like that'd be kind of cool like turn it into askinima style like in your terminal watch Jerod Santo Those here's star wars asky star Wars. Adam Stacoviak No, not Star Wars. That's not an ad. It has to be like Dawn Soap or something. Even worse. All they need to do is take typical advertisements, like TV commercials, and they need to turn them into Ask Cinema with some sort of software program and feed those into your terminal. That'd be sweet. There's a dog that hunts there, think maybe but i don't know talk to our friends at sourcecraft and at very least get a changelog ad up in there you know like hey while you're sitting here Jerod Santo Waiting for this thing to go why don't you listen to the changelog dudes so my prescription though for people is this is like don't sleep on this free it doesn't have to be your daily driver But what about free docs who wants some free docs because this thing will just wind up and do whatever you want for a couple hours unfettered every day so you basically have two to three Hours per day of amp code that you can just use it still has the Oracle. It still has everything else that it has. It's nothing different about it than I can tell. It seems just as smart, just as able. And get yourself some free documentation, maybe some free thinking, some deep thinking. Have it write some tests for you. Stuff you don't really want to do, and maybe you're not going to do. Maybe do a code review for you every day for free. Boom. You're wasting it. It's like free labor, just sitting there. Scoop it up. ([Time 1:13:20](https://share.snipd.com/snip/ec19e4c2-301e-468f-bd61-746ba9b147e1))