We see dead projects

So I’ve got a confession to make…

Okay…

One time…

One time…

…when I was doing an interview for this show, I went about 20 minutes into the show and I realized that I did not hit the Record button.

Oh, gosh…

And it was a guest, a guy I had just met that day, and we were grooving, and we were having a blast… And then I stopped and I said “I am so sorry, because I’ve done this –” How many times have we 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 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 think 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, and… Relief. Sudden rush of relief when we realized we hadn’t wasted 20 good minutes. Now, that was a couple of years ago. And I got another confession to make…

Oh, gosh…

[laughs] And this is a confession to you, Adam, because you know this. This is a confession to our dear listeners… I did it a second time.

Yeah… Sure did.

However, this time we did not have any backups.

Nope… My QuickTime has been on the fritz… It keeps crashing on me, or something or other…

Mine’s running, but you can’t do it with one voice… So we just spent 45 minutes recording some dynamic dialogue for y’all.

Some really good stuff. Really good stuff.

Good stuff. And it’s now into the ether. We were piping it to /dev/null the whole time.

That’s not cool. That’s a bad place to go if you want the good stuff. The good stuff does not go to /dev/null, okay? It goes to maybe Temp, temporarily…

I would be very happy if we found that conversation in Temp, so we can restore it and give it to Jason and say “Edit this, would ya?”

“There it is! There it is, in Temp.”

But we think it’s /dev/null. We’re pretty sure, we’re a hundred percent sure it’s /dev/null. So…

Riverside should tell you this though, right? This should be a Riverside feature.

That would be a great feature. Like, “You’re getting dramatic. Are you recording right now?” It’s similar to how your watch says “Are you working out?”

Yeah.

Like “Hey, are you doing a run?” and you’re like “No, but sure –”

[unintelligible 00:04:50.16] Don’t make fun of me.

[laughs] 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 you do forget to start your workout, it’s pretty nice.

It is nice. It’s like, “Yes, I am.”

And had we gotten 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. We have to leave that one in the ether for you guys. It was really good.

It was so good, you guys. You were gonna love it.

You should have been there.

[laughs]

You should have been there. Okay.

Alright.

Let’s do a new show then. 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.

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.

[laughs] 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.

[00:06:06.14] Dreams come true…

Now, Thomas, you get what you asked for, a year later. And his idea was “Hey, what if you just go out and 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 would hurt too bad…

[laughs] 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 three by three 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. Have you had 298 software projects? Probably not.

No, that’s a lot. There’s one I didn’t even know was dead, right here.

Uh-oh, you just found out. Chromecast?

Yeah. I don’t even use it, but I’m surprised by that one.

I think they’ve decided they were going to discontinue that, and replace it with like some new Google TV thing, or something…

I think it’s still software, right? It’s still a protocol. You can still cast to your –

Oh, I think maybe Chromecast, the hardware.

It does say hardware.

The actual Chromecast.

Yeah.

Like, they’re not going to produce those anymore. But there’s so many things… Of course, the big ones, right? So the biggest one that kills me to this day would be Google Reader…

Yeah…

Killed over 12 years ago. Faithfully served from 2005 to 2013… And not only did Google kill Google Reader, it kind of killed RSS as a social phenomenon. I mean, except for us weirdos and hippies…

Never let it go.

…who are still using our RSS readers, like Feedbin, or Feedly, or what have you… But the death of Google Reader drove away what we’d say like mass adoption of RSS as a thing. And that was a big bummer… How about Google Wave? Heard of Google Wave?

A lot of buzz around that one, right? Was that –

It was so weird.

What was Wave? Was that the – oh, I was thinking about Google Plus.

Oh, that was a follow up. I think Google Wave came first…

Yeah, it attempted to –

Google Wave, released in 2009…

Yes.

It was like real-time, collaborative stuff… It was like a collaborative editing tool, kind of like in a social network fashion, of like “We’re all working on this thing together and talking to each other.” And that was the idea, like you hop on the wave and you ride it for a while. And it was very interesting. I think it was very much a single-page app, with all the Google engineering in it… Got tons of people super-excited.

I’m seeing a trend here when I’m looking at these. Maybe I’m just in a certain era looking at these, but… Killed almost 13 years ago. Killed about 13 years ago. Killed 13 years ago. Like, what happened 13 years ago, Google?

[00:09:59.08] Right.

Was there something going on? A lot of them died 13 years ago. Or just the ones I’m looking at. Yeah, Reader was a tough one, because - I agree. I mean, that one seemed misplaced; like, a thriving project… It’s like “Can you just figure out how to make some money from it, and not kill it?” But just like you can’t go to Target today and buy a Blu-Ray… What’s up with that? Who’s killing the Blu-Rays, man?

The death of physical media.

Physical media is dying. It really is dying. And my Plex catalog is suffering as a result. Such a shame…

Well, this website also catalogs open source things… AngularJS, killed almost four years ago… [unintelligible 00:10:46.05] 11 years old. You have Polymer –

Wait, AngularJS is dead?

AngularJS is dead according to KilledByGoogle.com.

That’s so wild, because I think I know somebody who was just talking about it. I won’t name them, because it’s embarrassing… And they weren’t aware of this. This is news to them.

So I think –

This is news – they were excited. They were like “Yes…!”

Oh, that’s not good. Sorry, y’all. No, I think they replaced it with Angular, without the JS.

Oh, okay.

So it’s I don’t know if that’s a spiritual successor, or a fork, or a rebrand… I don’t know what it is in relation, because I don’t play in this pond… However, AngularJS as a thing is dead, and Angular, which you can find at angular.dev, is probably the same thing…? I don’t know, some of our Angular fans out there can let us know what’s up with that. But AngularJS was killed. So I guess rest easy to Adam’s –

…unspoken of friend, yeah.

Yeah.

I mean, maybe they were talking about Angular.

Maybe they already know [unintelligible 00:11:48.24]

Angular.dev, versus AngularJS.org…

Right.

…to be clear. a lot of death over there at Google, you know? I think about two years ago when we were first talking about this AI hype wave, and what’s happening there… I think I was even bullish saying that Google would not win, and they would die… What’s wrong with me? But they’re all playing this catch-up game.

Did you say that on this show?

I just said they’re in trouble if they don’t change. I mean, they seem so far behind the ball…

Well, Google search as it currently existed certainly was being assaulted.

Yes. I mean, it’s dead to me. I only google something if I’m just trying to find a thing. So it’s usefulness has degraded to about 10% of my needs. It hasn’t become less useful, it’s just not as useful as other things. My needs have shifted. That’s just how it works. Well, I want a recipe… Give me the recipe. Riff with me. Make Mexican chicken with me, okay? Don’t point me to 17 Mexican chicken recipes… Let’s be chefs. Let’s make some food. Let’s go!

Good rant. I like that. That was a solid rant. Alright, so lots of stuff killed over there by Google, of varying qualities and interests… But that’s just kind of how they do what they do, and sometimes we feel a vein, and other times we don’t really know what it was. Google Revolve… I have no idea what that was.

Nah.

Google Now? No idea. Songza? Beats me. Google Flu Trends. Heard about it for the first time right now. It’s been dead since 2015. So lots of engineers over there, and lots of engineering… So let’s turn now to something else.

Ooph… Well, didn’t fully prepare for this one, but mostly; let’s just say mostly.

Okay.

I would say there’s been a lot of advancements on the frontend; one in particular that has evolved, naturally, is CSS. It lagged for a bit there… I think there were things like Sass, and… Gosh, there was something else. Less was a thing…?

Yes, Less.js.

[00:14:04.14] I’m just reminded about this in my brain just now… Don’t know about Less, it 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, and I’m so sorry… Created Haml, which was popular to Rubyists back in the day.

That’s right.

It was white space-aware, similar to Python is… And the original Sass syntax was just that, it was a pre… Pre processor?

Yeah.

It pre processed… Before you got the actual CSS. It was kind of cool. I’ve done a terrible job of explaining it, but… Sass, to me - I don’t know if this is true; don’t be offended if you’re still working on this… It’s kind of like dead, or dying. It’s evolved, it did its job, just like Quicksilver… You didn’t hear about that now, did you, listener? No, you didn’t… 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… Gotta bring that pun back. And then Sass did the same thing, I believe, in a great spirit to CSS. The goal was never “Let’s make Sass the thing”, it was “Let’s make CSS better by adding Sass to it.” You pre-processed this superset of CSS, is what Sass was. It sat above and on top of CSS proper. And it gave you a lot of superpowers you just didn’t have on the frontend. And I don’t know where it’s at right now, honestly…

Well, I went to sass-lane.com/blog, and it’s officially dead.

Is it?

Posted on the 23rd of October, 2025. It was five days ago.

Dang. How timely.

LibSass… Great timing for this.

This is breaking news!

Yeah.

Sass’es 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.

That is LibSass.

I don’t know what Dart Sass is, because I’m not in that world, but I think this is –

I can give you a little bit of that picture… So I remembered Hampton Catlin… So I remember the name. That’s the person. Created Haml, created Sass… I believe Sass was… I don’t know, C++ maybe? I know LibSass was definitely C++. And so the engine, I believe, for Sass was slow, and the challenge for a while, and the argument for many was “It’s slow. Can you make it faster?” And I believe they extracted out the engine, which was LibSass. And that was rewritten. It came many years after Sass was even a thing… I think it was originally written in Ruby, now that I remember; maybe it was Ruby. And LibSass was written in C and C++. It was meant to be the library that others could leverage to make their Sass better. So if you had a different Sass implementation, you had one in Ruby, you’d have one for Dart… Dart was a language. A Google language, actually, I believe.

Mm-hm. I should see if they’ve killed it yet. Hold on.

Let’s go see.

KilledByGoogle.com. Oh, Dart, you are safe. “Continue using Dart.”

[00:18:12.12] You are safe. That is not cake. And that’s pretty much it. You’ve got LibSass, which was the promise of the future… Get that thing extracted out of that slow Ruby world, if I’m speaking correctly… If I’m not, just assume it’s truth, okay? I believe there was a NodeSass, because the Node ecosystem needed a special way, and they leveraged LibSass to make NodeSass…

You had SCSS, which was more like the traditional SAS – or sorry, CSS syntax, which had parentheses, and curly braces, and all the fun stuff… Yeah. And now what do we have? We’ve got Tailwind, right? Is that what we have? I resist the frontend. If I have to go to the frontend these days, man, I’m just upset. I’m just upset. Put it in the terminal. Put it in the terminal. That’s where I live now, okay?

So Sass was the radically different-looking one, right? Like, it was no curly braces, it was white space-significant…

Right. It was the OG.

Okay. And just like Haml was very much a minimal and white space-significant templating language, Sass looked different than CSS. And so you kind of had to learn a separate – maybe call it a dialect of CSS, in order to use it. And then SCSS… So that was Sass. SCSS, which was also part of, I think, LibSass and the whole Sass community, was the, as you said, kind of the more CSS-looking version, that still got the features that you wanted. Because the features that it had, that you really wanted…

Well, you could drop it in for CSS. You can literally take a .css file and just call it .scss and you were using Sass.

It was a superset of CSS, correct.

Yeah.

But what you really wanted was nesting and variables. And then it got fancier from there, and people who are hardcore frontenders 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, variables and nesting… I’m a happy camper. I don’t need much else in my CSS.

Yes. 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 at 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. So I think to say dead, in this case… End of life, I believe, is the better term, like they did in the blog. Because it did its job. It was not meant to replace CSS, it was meant to augment and be a superset of it, to push it forward and give you features that you just couldn’t have otherwise. Because you could just use CSS. It was actually kind of hard to use Sass, because of the S-A-S-S version, because of the S-C-S-S version, because of the eventual LibSass… And just, it seemed like a challenging space to get into.

I will say, though, that the original syntax that mimicked the beauty of what Haml brought, which was this white space-aware, white space-significant world, you can easily see what you said before, which was important, which was the nesting. You could see how far based on indentation you were being, and you could tell pretty quickly while looking at a Sass file how terrible your CSS nesting and inheritance and tree essentially was, your specificity was just way off, or way too…

Way too deep.

Way too deep, yeah.

[00:22:11.04] Too specific.

Yeah. Too many selectors.

Yeah. And that, I think to this day, is the pro argument for white space-significant languages, is that you can at a glance see if you’re getting crazy with your indentations. That being said, while I understand that stance, I think you can also see it even with curly braces… I guess you’re not required to indent, but anybody who’s not indenting their code… Like, talk about death - you’re dead to me. Just don’t care. Don’t care to collaborate with you.

You better indent that code.

That’s right. [laughs] Or you may have another thing coming. So the indentation as advantage of white space significance is the scannability… I get it. I just think it’s not such an advantage. That being said, I’m not a white space hater. I’ve enjoyed – I liked Haml quite a bit, I like how clean it looked, and I appreciate Python at times, as well… Although even with white space-significant, I’ve seen some pretty gnarly Python code. So you can make a mess no matter where you are in the world… Just start being messy.

Break: [00:23:27.03]

You know, there’s nothing like a good project that was brand new, though, that can use the original, the OG Sass syntax… Because you didn’t have to compete with old CSS to move around and support. You could just start this new world with this indented world, and… Until you had to copy CSS from Stack Overflow or something like that, you were pretty happy. And then you had to like paste it, and then do a bunch of deletion, and we didn’t have AI back then to like automate this stuff… Gosh, that’s wild to even say that. “We didn’t have AI back then.”

Yeah, the bad old days.

[00:25:53.02] Yeah. Why would I write that code? It would do it for you, okay? It will do it for you. Listen, 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’ve 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 AMP, let me see what they did…” The 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… Shifted to production, and everything… Gosh! In a blink of an eye, I meant to think about it, and 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.

I just got a little bit concerned when you started talking to AMP, in kind of like a Marvin Gaye-style voice, you know?

“Let’s talk about it. Let’s think about it…”

[laughs]

Well, I was being – I really was like “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 going to 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 “Alright, cool.” I was kind of upset, because I really wanted to think about it, but it wasn’t necessary. It’s like “Gosh, man… I 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.

[laughs]

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… I will reveal to you right now what it is… I’ve been itching this scratch for a while to replace Pi-hole with DNS-hole. I like it a lot. It’s just got a good ring to it, and I can’t let it go. I failed – the first project version of it, just total failure. I was vibing all over the place, slinging code slop everywhere, Rust memories, unsafe things… Just –

Oh, my goodness…

Just dead code everywhere. If you’re a Rustacean, you’re just pissed at my code, okay? You’re just slapping my code all around. Well, I learned my lesson, okay? I just went ahead and said “That’s v1. It is now a reference-only project”, and just started off simple. And now simple is so much better. In fact, DNSimple is sitting at – it’s resolving all my DNS queries as we speak, on this podcast.

DNSimple is?

Sorry, DNSimple. DNSimple. DNS-hole. DNS-hole is resolving all of my DNS queries right now.

Oh, it is?

Yeah.

That’s pretty cool.

I’m excited about that.

I have a new phenomenon when I try to start a new project now where I don’t know what language it should be in anymore. I used to have opinions… But I do eventually resolve. So for instance, I started up this Static Site Generator thing that does some crawling, and blah, blah, blah… And I just tell it what I want. This is Claude 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. 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 “Nah. No, not really.” And I’m like “Let’s rewrite this in JavaScript. No TypeScript, please… Just pure JavaScript. Keep it clean, keep it loose. Give me options.”

No types. No types.

Do not constrain me. I’m vibe coding, okay?

That’s right.

[00:29:51.29] And so it ports it to JavaScript in the 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 is 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 bet on Go.

I like Go for that reason. You’re 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, in my shared hosting. It even embeds the HTML templates, and the CSS, and everything. You embed it all in one binary, dude. So cool.

One binary, to save them all. That’s how it goes right there.

And you either rsync, or SCP that sucker anywhere you want, and just run it. No dependencies, no package managers… Now, the code is kind of verbose and ugly… But who cares?

Do the tests pass? Does it work?

I’m glad that you said the second part… [laughs] Because I did not request any tests. Well, the thing about Claude Code is if you give Claude Code a task, it’s going to write the task, and – if you don’t specify “Please also write specs”, or whatever. If you’re just like “Do this thing.” It’s going to actually test that it does it via its own little shell scripts, and stuff. It’s going to output to a test directory, and it’s going to grep the thing, and it’s going to sed, and it’s going to awk, and it’s going to count lines, and whatever it’s got to do to make sure that it’s right. And then it’s going to say “Hey, I’m done.” And you’re like “That’s testing.”

That’s testing.

Now, I don’t have an automated test suite at the end of the day, but…

That’s right. Those tests are like one and done. Bye, see ya. But yeah, they work –

It’s all low stakes stuff anyways. That’s what I use it for, is like low stakes. I’m not resolving my own DNS with it, or something.

In that case, it is higher stakes…

Yeah, it is.

I’m [unintelligible 00:32:11.00]

Alright. Well, we’re upstream here… Let’s go to another project that is dead, for various definitions of the word ‘dead’. We aren’t declaring death. 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 MeteorJS.

Oh, yeah. Wow.

Remember MeteorJS…? Often imply called Meteor… It was 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 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 their own little 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 DeBergalis… I’m not sure how 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. It got lots of fanfare, lots of people trying it… It exploded in popularity… Kind of a typical hockey stick growth kind of thing. Raised more money. Raised more money. This is an MIT-licensed project that just keeps raising more money.

[00:34:19.13] 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 this Discover Meteor, which was either a newsletter or some sort of website, co-authored by friend of the show, Sascha Greif, and Tom Coleman. Of course, you know, Sascha was big into Meteor for a long time.

Really big, yeah.

And you know, things continued up and to 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 “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, it started to integrate Npm in 2015… They got worldwide production apps out there. So, I mean, not unsuccessful, by any means. I’d call this success, for many definitions of that word.

Things continued to go okay. In 2016 they were exceeding their revenue goals, according to them… But community debates arose. There was purists who loved the all-in-one 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 like 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, people started moving on… The State of JavaScript survey, of course, is Sascha Greif’s thing. He’s been doing it for a long time now. So the State of… The State of JS actually cataloged Meteor’s interest, and then you could start to see the decline. And sometimes that is a – what do you call it; a self-fulfilling prophecy… People started to see it declining in interest, and so they were 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. It got put into maintenance mode… I mean, more emphasizing maintenance than growth. And Tiny, which is the company that bought it, pledged they’d support it for long-term, like five years. Still out there…

[00:38:18.04] Still going.

Yup.

That’s wild to think about that, honestly.

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 Vue.js, 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, the original mission of Meteor as the future of web development - it’s gone with the wind.

You know, I never got on the train. I never understood it, personally. It was always confusing to me.

They did an interesting thing like munging server and client code early on, which of course, the React team has been struggling to make that a thing, in the public’s eyes, at least… They made it technically a thing, but in terms of adoption and people understanding it, it’s been an uphill battle for them as well. Meteor had that from the very start. I did use it one time for a hackathon. I had a lot of fun with it. Built a game for a hackathon… And it was very strange, knowing “Am I writing server code, or am I writing client code?” And they were syncing stuff for you… It was very kind of advanced at the time that it was out there. It had a lot of cool innovations in it. Maybe because of the time that I tried it, but also what I tried it for, to me it felt more like a toy than a tool that I would build a business around. Not that it necessarily was, but that’s just the way that it felt to me. So I never took it super-seriously, but obviously, lots of people did, and they’ve had lots of – they got 44,000 GitHub stars, 508,000 installs, 29,000 Stack Overflow questions… So, I mean, they definitely got out there, and people use it, but…

Yeah.

…no longer as culturally relevant, surely, as it was back in the good old days.

I mean, to be honest about the dead bodies in the JavaScript world is just plentiful.

Yeah, that’s true.

I mean, good luck winning that war..

Well, even when you’re winning, two years later all of a sudden they’re like “Yeah, we’re just not interested anymore.”

Yeah, “We need something new. We don’t even actually – you still solve our problems, you’re still good… We just don’t like you anymore, okay? We’re just used to getting new things. Where’s the new thing?”

Right.

That’s my feeling as a frontender in the JavaScript world, is just like expect change, expect churn… And honestly, it’s what tired me out of really wanting to play in that world at all. It’s just like trying to keep up is basically impossible.

That being said, if you look at it today, it’s basically the same thing as it was five years ago… I mean, what’s popular in JS land frontend world today? It’s React, basically. It’s Next.js. You have people raging against Next.js, you have other – I mean, there’s always experimentation, there’s always people saying “React is dead”, React is this or that… But it’s still just the 800 pound gorilla right now. And yeah, there’s been innovations along the way, and alternatives along the way, but it’s not like there’s been an upheaval. You know, there’s this HTMX movement, a return to form of multi-page apps, as they call them now… It used to just be how we build things, versus single-page apps and multi-page apps… Server-side rendering…

[unintelligible 00:42:02.16]

Full-stack frameworks, I think, are more interesting to the JavaScript community than they ever have been, but aside from that - I mean, it’s kind of been more steady, in the last probably five years, than it was prior to that.

I wonder if that’s because of entrenchment, though.

[00:42:19.26] Oh, for sure.

It’s just too hard, there’s too much sunk cost in the world of React and Next to move away from it easily. There’s too much investment into the infrastructure, the usage of it… I was thinking about this recently with Git. You make any piece of software, almost everything thinks you’re using GitHub to host it. It’s going to be there, and you’re going to have actions. And so CI begins with actions files, essentially. That’s where your CI begins if you’re starting fresh. And then everything’s assumed to be around Git and GitHub. And I love Git, I think Git’s great, but I just saw our friends over at Fallthrough talking about the newest thing, and I’m not even sure what it’s called, but there’s a new thing in town to check out.

Jujutsu?

What’s it called?

Jujutsu.

Jujutsu. There you go. Which I’m not even versed in, but I’m just thinking “How in the world do you even – how does the world move away from, at this point, Git?” It’s too entrenched. All you can do is really improve it… Maybe you have some people try and use - maybe the 5 to 10 to 20 percenters are going to try and use something different… But you’re just going against the current… And it’s just more challenging. Until it’s not, I guess…

Well, slowly, and then all at once. That’s what they say.

That’s true.

Slowly, and then quickly.

Slow, then fast.

That’s right. That’s how you get rich, that’s how you get poor, that’s how you get entrenched, that’s how you get disrupted…

Yeah, it’s so fast. It is. Things happen in one fell swoop, as they say. That’s what I heard. So many swoops… So Meteor - this is on your list, Meteor? Well, Angular…

Well, Angular I just saw on the website. But Meteor was on my list, yeah.

Meteor. Yeah. Is that how you spell the word Meteor? I don’t even know.

M-E-T-E-O-R. Meteor.

That’s how you spell Meteor?

Unless you’re talking about something that has more meat in it than something else. It’s Meteor…

Meteor.

That’s why I kept calling it Meteor, because I didn’t want to accidentally say meatier, like “Let’s talk about something with more depth to it.”

Right. Let’s get some meat on these bones in here.

Right.

Let’s get these Meteor out here.

You’ve got another one? You want to – we can loop… And I think it’s been long enough. We can loop back to Quicksilver. It’s going to feel fresh. It’s going to feel so fresh.

I think it might feel fresh, yeah. The pain is a little less fresh –

The pain of our loss is – yeah. Time heals all wounds…

I’ll give it another try.

Let’s talk Quicksilver. People want to hear it.

Take two. Here we go.

Okay.

Alright. Quicksilver, if you’re not familiar with this, is a software from [unintelligible 00:45:09.19] 2003. A long time ago. 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. It depends on who you ask. But Quicksilver was a app launcher for Mac at that time OS X. Not macOS. 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.

Mm-hm.

That likes Gab. That was a good one. Got lost in that one…

You just sidelined yourself with a stupid reference… [laughter]

A little Mac geek going on there…

It’s 2003, you’re running macOS X, probably like Lion, or Leopard, or some sort of a cat. It was a big cat.

Yes… Who the heck knows which one it was, but it was a kitty cat, yeah.

That’s right. And Quicksilver was launching your stuff, man.

[00:46:12.01] You wanted to launch things, and you would go and you would go to your app directory, or you would go to your dock 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.

Yeah, it’d 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 Quicksilver gave the world.

So many things.

So many things. It was made in 2003 by a fellow named Nikolas Ditkov. 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…

Copying it.

Yeah, absorbing it, copying it… Yeah. Absorbing I guess is probably not the truth. Copying the features, baking them into the operating system… First mover advantage did not work out for them. LaunchBar, Alfred, Raycast - that’s what I use… Now, I did use Alfred, and 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 and I’m not using the value anymore… Owner of a lifetime license of Alfred. I’ll never use it again… Just saying. Just saying.

There’s no chance they can win you back?

I mean, I just don’t think so. Never say never, though… My buddy Justin taught me that. But you know, I don’t see it happening.

Okay.

I don’t see it happening.

What about Quicksilver? Are you ever going to go back to Quicksilver?

I don’t even know where to go. Where would I find Quicksilver these days? I think it’s still a thing, but I don’t even know.

Brew cask install? I have no idea. I will say this… Quicksilver was the coolest. I mean, I’m not going to say I bought a Mac because of 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, 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, the actions you could take, beyond just – 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 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 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 power. I wanted the Quicksilver power. And so your story is basically competition… Why didn’t it get better? Why didn’t they win?

[00:50:07.11] I think it just died on the vine. I think it just – it got there, hung out 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.

[laughs] So then you could use it again.

Raycast, don’t die.

Well, Quicksilver was open source though, and these other ones aren’t.

Yeah. I mean, I guess –

Does that mean anything to you? 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’ve 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, you can swap models, you can save things… I think they could have done and should still potentially do a better job of that chat app. Like, it lives in Raycast, the launcher, so arguably, potentially the best thing they’ve built - a unified chat app for Mac that’s Mac-native is buried in somewhere you wouldn’t expect.

Yeah.

It’s like, you turn over that rock - boom. Gold. What’s up with that? Why is the gold under the rock? It doesn’t belong there. It belongs in my pocket, or my safe.

On the shelf, where you can see it.

Yeah. I want to see the gold. Don’t put it under the rock. So I don’t know…

Gold’s pretty valuable these days.

And Raycast is venture-backed… They could fail as a company for those reasons… I don’t think they will. In fact, I haven’t talked to him in a while, so I don’t know.

Yeah, but a lot of those Meteor users didn’t think they were going to fail either.

That’s true, man. That’s true.

Not that you all failed, but you know what I mean.

Yes. I did find Quicksilver though on GitHub. It is, naturally, GitHub.com/quicksilver/quicksilver.

Recent commits, recent activity? Anything going on over there?

This is a bot… So I don’t know for sure… 5,909 commits total, one yesterday from [unintelligible 00:53:24.10] integration bot. So I’m not sure about that one in particular. Although on the 23rd of October there was a merge pull request, pull request 3080, which was from the branch Quicksilver copy configuration build step. If you’ve got to use Quicksilver, there it is. It’s there for you, waiting. And it’s open source. What is the license…?

If you’re listening to this and you do use Quicksilver to this day, let us know in the comments. We want to hear from you.

Yes. Did you move at all? Did you come back? Apache 2.

[00:54:12.14] Are you still running macOS Lion…? Sorry, OS 10 Lion.

OS 10, yeah. 76 issues, two pull requests open… Not that bad. It seems to be tended. There’s not like a thousand issues, for example.

Are you wishing you took it off your dead list, though?

Nah, it’s still dead.

[laughs]

Nah, it’s still dead. It’s still dead. You guys are working on this, I can’t say that. It’s just rude. I’m sorry for being rude. It’s just good podcast nature to say “Hey, don’t –” If I was standing next to you in the hallway track of a great conference, I wouldn’t be that rude. It’s dead to me. I’m not going to use it. Can I come back? Man, absolutely. Give it a try. Raise some money. That’s the way you get to prosper, is raise some cash.

It sounds like that’s the way that you die.

Maybe.

Break: [00:55:05.26]

Speaking of, how about RethinkDB?

Oh, gosh, man… Blast from the past.

Remember RethinkDB?

Slava Akhmechec.

That’s right.

Akhmechet.

That’s right. [laughs] 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. It was active probably till 2016. The company actually declared bankruptcy in 2016, and the core team was disbanded. The repo was archived… I mean, this one’s dead-dead.

So dead.

[00:58:04.12] They did inspire some community forks, such as ElephantDB… RethinkDB - you can go back and find that one in our archives. Of course, we’ve had Slava Akhmechet on the show, I think probably at the beginning of RethinkDB, and I think at the end of RethinkDB. Wasn’t that right?

I’m thinking two times during its heyday, and one potentially on its wayday. Bye-see-ya.

Oh, I like that.

That’s the wayday.

The heyday and the wayday?

That’s right. You’re going on your way. Bye-see-ya. You’re on your way now.

Now, why did RethinkDB 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 NoSQL database department of the world. And Rethink just couldn’t get there. But didn’t they write a – 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 RethinkDB failed”? And didn’t we actually do an episode maybe even called “Why RethinkDB Failed”? I’m starting to remember something like this. Check out our last episode with RethinkDB, what it was called.

Yeah. I’m scouring the internet right now for information…

Okay… I think the long and the short of it is no product-market fit. I mean, that’s what the startup guy would say. They had some design choices that were odd, because they had their own language… This is why when the FaunaDB people came on the show. And in fact, I think FaunaDB 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 “Are you sure you guys want to do that? Because it’s very difficult to get people to just use your query language. Can’t you have it be SQL-alike?” That being said, 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.

Well, if this is true, founder Slava Akhmechet said that 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 ABOS. 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 ABOS, and adoption elsewhere. There was a lot of motion around databases, and so you had a lot of choice.

Yeah, there was a thriving market of offerings.

Now we just want Postgres, right? Postgres.

Postgres 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 know what’s happening.

Doesn’t everything continue 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 Claude trying to 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?

[01:02:02.22] “Fun fact. I see one of the discussions mentioned an episode from the Changelog about RethinkDB. You folks have quite the archive.” Come on, Claude…

It told you that?

That’s what it said.

Do you have that in your ClaudeMD, like “By the way, I host the Changelog podcast, and we have quite the archive”?

I think it knows. It knows who I am.

[laughs]

Yeah, it’s definitely pandering to me, because it’s like “Yo… Great job.”

That is pandering.

Yes.

But it worked. Did it work? You do prefer Claude now, to AMP.

So Fauna.com actually doesn’t go anywhere, which was the location of FaunaDB. F-A-U-N-A dot com. It’s either in my block list, DNS-hole, or it’s dead. I’m thinking it’s dead.

Mine redirected to 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 into Chrome… What does Chrome do…?

Yeah, Faunadb.org is what you’re looking for.

Ah, okay. So, okay, my bad.

“It’s a strongly consistent OLTP database with a hybrid blah, blah, blah…” Open source project, it’s 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? The last commit to the FaunaDB repo was six months ago. Oh, and the only commit was all six months ago. So I think they open-sourced it and moved on.

Yeah. They were like “Listen. Crunch the history, put it out there… Cash the – no, don’t cash the check. That’ll bounce… Get out of here. Just get out of here…! It’s over. We didn’t do it.”

That’s right.

I mean, they had some really good ambition, though. I mean, I remember that conversation…

Totally. Really smart guy.

The CTO was – no offense to him, but I recall him just being a little just monotone.

Monotone. He was.

Very monotone.

Yeah.

And I think he left months after our conversation. I don’t know. Some things are meant to live and some things are meant to die.

[laughs] That’s very final.

If we keep going, I’ll just keep making stuff up, okay? I’ll just keep making it up.

Well, should we keep going or should we call it? We do have a couple of things, just quick mentions. ICQ…

Oh, gosh.

Remember ICQ?

Oh, yeah. Let’s do a little leave it list.

Leave it list?

Yeah. Like, ICQ is cool, man.

For sure. And you are cool if you had a low ICQ number.

Yeah. I don’t know what mine was.

You used to be able to buy the numbers off of eBay and stuff, too. Did you know that?

You probably still can.

You probably can. I was not that cool. I came way late to ICQ. In fact, I used it because someone made me use it. Sometimes people are talking, they’re like “Yeah, we use this”, and I’m like “Dang it. I have to install a new app.” But I did have… There was an open source project that actually combined a bunch of chat messengers, of which ICQ was one. AIM… Others… I think it was called Pidgin.

Yeah, I recall Pidgin.

Yeah. Or Gaim…

It was a bird. It was a bird one. Is it the Pidgin, do you think?

Yeah, Pidgin was a bird. It also means like – it’s like a… It’s spelled not like the bird, but like the language feature… I think a pigeon language is like a language that multiple people use, as a crossover… Let me google that instead of just talking out my butt.

While you’re doing that, there’s a Reddit post, one year ago, so this is actually 28 years, not 27, as I read it… So it says “After 27 years, ICQ is officially shutting down today.”

Yeah.

[01:06:08.24] “End of an era for one of the first messaging apps.” So 28 years ago.

Okay, so Pidgin, at Pidgin.im - also dead. Last modified in 2020. It’s a chat program that which lets you log into accounts on multiple chat networks simultaneously. XMPP, IRC, Bonjour, Simple, Silk, Sefer… I mean, tons of these I’ve never heard of. Sametime. I thought there was an ICQ… Oh, and then, that’s a Windows, Linux and other Unix-like client. On macOS try ADM. Do you remember ADM?

That’s what I was talking about, yeah. The pigeon. I was thinking ADM.

That’s the that’s the bird, ADM.

Yeah, I used that one.

Yup. Jabber…

Oh, yeah.

Google Talk… Remember GTalk?

Yeah.

Google probably also killed GTalk.

There’s still some chat up in there in Gmail…

ADM is also open source…

GChat.

Let’s see… Latest release of ADM runs on macOS X. 10.7, 10.5, or newer. So this is also dead.

Man, all this dead stuff…

But it’s because this audience they serve is gone. I mean, who needs to connect to these different chat apps nowadays? Nobody.

Only fools.

Last commit here was like four years ago. So we’re finding all kinds of dead projects as we just talk. To go back to the word pidgin… Pidgin is a grammatically simplified form of a contact language that develops between two or more groups of people that do not have a language in common. That’s a great name then for a program that does what it did.

Yeah.

And so if you speak Swahili and I speak English, and we don’t have a crossover language, we would develop between our people groups this third simpler language, so we could actually communicate, and we’d call that a pidgin language. Interesting.

Very interesting.

So ICQ was rad, ICQ is gone… Skype was cool for a minute.

Dude, Skype was crucial.

It really was.

A lot of good days on Skype.

And now it’s gone, I assume… Does Microsoft still ship new versions of Skype?

It’s dead, man.

Who cares, man…

Teams. Long live Teams.

It’s irrelevant. Oh yeah, Teams…

I would never use Teams, personally, but…

What are non-Microsoft people using? I guess Discord… For voice it’s probably Discord at this point.

Even real time – I guess Zoom won that, right?

Zoom…

For a while it was Skype. You would skype somebody. You would never zoom somebody.

Right.

Then we would WebEx… Ugh, gosh…

[unintelligible 01:09:03.13]

I almost punched my screen just now, dude… Ooh, it was a bad situation. [unintelligible 01:09:10.05] It was almost a situation here. Yeah…

Like, you say WebEx, and that’s like worse than Rochambeau… You know, pick one. It’s like, yeah…

WebEx is just nasty. Nasty, man… You know, I do miss the fact that – gosh, I’m just like brain farting… Not ICQ. IRC… Is IRC still a thing?

Yeah, I mean, it’s as niche as ever. I think it had a moment when people were interested in it, but then like Freenode freaked out… I can’t remember what happened with Freenode, but like people that were running Freenode ended up being weirdos, or something… Somebody can fill us back in on that. And IRC just never had a chance in that way.

It’s a shame, because there’s a lot of people who found their love for the thing they do on a mailing list, or some sort of… What were the original like chat rooms called? They were like PDP hubs, or something like that, or… I don’t know what they were called. I’m not even from that era. But I kind of miss the days I didn’t take a part in, in a way.

[01:10:15.04] [laughs]

This one was simpler. Now everything’s commercialized. Like, if I can’t make money doing this, it doesn’t belong. And that’s just kind of sad, man… It really is. I think you should scratch your own itches. I’m just scratching itches, that’s all. It’s a pretty wild world out there these days.

You know, one thing I was taking note of when I was looking at Wikipedia for Quicksilver was how cool that logo was, man… That’s what made Quicksilver cool. It had a lot of cool things initially.

Yeah.

The cool HUD, the cool logo, the colors, the new thing… It just doesn’t work out sometimes. And if you go to Meteor’s – that’s how I said it too, Meteor… Meteor Software on YouTube - they’re thriving. Like, one month ago, 135 views. Seriously? I mean, four months ago, 3x faster builds, 409 views… They’re thriving on there. They’re just busting it up. Putting their content out there.

So should I take it back?

No, they’re dead, man. They’re dead, man. They’re dead.

[laughs]

That was me being politely facetious.

You know what I think is dead? I think this episode is dead.

You know, folks, I’m just really sad for you, because version one was better.

So good. It was perfect software.

It was, man. It was good stuff. And we did a good job with this episode, but I feel like we kind of botched it…

Well, we botched it when we didn’t hit the Record button, you know? The rest of the time, we’ve been kind of just…

…begrudgingly here, just podcasting. It’s like “We’ve gotta be here now”, you know…

We’ve just kind of been coping…

Yeah, it’s been hard.

So I guess, you know, sorry… But also, you’re welcome.

Yeah. It is what it is, as they say. In closing though, I do want to say one thing.

Okay.

Off the thread we were in there, but I did mention AMP. And I do want to say that AMP is free. You see this, Jerod? Ampcode.com…

I did see this. This is interesting. AMP is free… But free as in how?

Free as in you’re the product again, okay? They’re gonna advertise to you.

[laughs] You just went straight backwards on this thing. You’re the product again. Yeah, they’re gonna sell your eyeballs.

That being said, though, I don’t know what they’re tracking, so I can’t say that for sure… But it’s the first that I’m aware of where you can use a - I’ve said it before, my favorite AI agent, coding agent, for free. And I’ve hit limits with it, I think it’s a couple hours. I don’t know what the actual limit is, but after a bit, they’re like “Listen, we’ve only got six advertisers. You’ve seen them all.”

And the way it works is they show you ads in the terminal, right?

Right above – 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 the UI on this, too - is this a subtle text-based ad. It’s not even a bother, honestly. And I’ve actually enjoyed them. Like “Oh, sweet. They’re doing that? Cool. Nice.” I find myself actually wanting to see the – I want to see the ads.

It doesn’t bother me. I mean, you’re opting into this free thing, with – it’s all straightforward, the way it works. So if you need to use it for free, and you’re okay with ad-supported, I think it’s a great option. So I would say there’s no shame in that game whatsoever.

[01:14:04.14] If you want it to deep-think for you - well, ask nicely, okay? Because this probably isn’t gonna do it. [laughter] It’s like, “There ain’t no thinking in here. We just do around here.”

So if the whole time it’s deep thinking, could it show you like a video ad? That’d be kind of cool, turn it into like ASCIInema style in your terminal… That would be sick.

That’d be kind of cool.

I’d watch those…

Here’s Star Wars. ASCII Star Wars.

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 ASCIInema with some sort of software program, and feed those into your terminal. That’d be sweet.

There’s a dog that hunts there, I think, maybe. But I don’t know.

We should talk to our friends at Sourcegraph and at the very least get a Changelog ad up in there. Like “Hey, while you’re sitting here waiting for this thing to go, why don’t you listen to the Changelog, dudes?”

Absolutely. So my prescription though for people is this: “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, so does everything else that it has… It’s nothing different about it that 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. Take it.

Alright. Well, thank you for hanging out with us…

Yes, friends, thank you so much… Remind me, is this our spooky episode?

Spooky, man.

This was spooky. Was it a little spooky?

Well, you’re acting a little spooky there. You’re like “Dead, dead…!”

[unintelligible 01:16:14.11] I was just trying to like lean into it, okay?

[laughs]

Trying to play a role here, do my job.

You got the sixth sense “I see dead projects” up in there.

That’s right.

Although that got probably cut before we started hitting record.

Yeah… You might have it, though…

Let’s say goodbye, friends, before we –

I think you need to say it twice though, so that’s cool, though.

Oh, good.

Enjoy your Halloween, eat lots of candy, brush your teeth, and support open source.

Stay spooky. Bye, friends.

Stay spooky.