Pod Insights Syntax - Tasty Web Development Treats 971: Stackoverflow and Firefox are Dead? About this episode Is Stack Overflow actually dying, and what does that mean in an AI-driven dev world? Scott and Wes break down the latest web dev news, from Firefox’s AI crossroads and Apple’s browser engine changes to new tools, docs, and spicy browser updates.
Show Notes
00:00 Welcome to Syntax!
02:36 Stack Overflow is Officially Dead
05:40 AI’s Impact on Software Development
07:56 Brought to you by Sentry.io
08:20 Micro QuickJS for Embedded Systems
13:03 Open Workers: A Cloudflare Alternative
20:09 React Aria has new Docs
24:12 Firefox and the AI Dilemma
The Mozilla Announcement
31:11 Apple’s Browser Engine Changes
Using alternative browser engines in Japan.
36:12 Fractured JSON for Better Readability
37:45 New Chrome Permissions Dialogue
Chrome Network Access
41:15 Sick Picks & Shameless Plugs
Sick Picks
Scott: TRMNL E-Ink Display
Wes: ACEBOTT
Shameless Plugs
Scott: Syntax on YouTube
Hit us up on Socials!
Syntax: X Instagram Tiktok LinkedIn Threads
Wes: X Instagram Tiktok LinkedIn Threads
Scott: X Instagram Tiktok LinkedIn Threads
Randy: X Instagram YouTube Threads
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Wes Bos? Welcome to Syntax. Today, we've got a whole bunch of web development news for you. Stack Overflow is officially dead. Is that the case? Why did it die? There's a new JavaScript runtime in town, which is not what you would expect. There is an open source CloudFlare Workers that's written in Rust. Kind of interesting. Can Firefox survive all of this new, in the new age of AI? Fractured JSON, new spec for formatting JSON. Apple now has to allow browser engines in Japan, which is kind of interesting. And then a new pop-up you may have seen popping up in your Google Chrome recently. We're going to explain what that is. So some pretty good stuff.
0:42 Wes Bos? My name is Wes Boss. I'm a web developer from Canada. With me as always, Scott, how are you doing today?
0:47 Scott Tolinski? I'm doing good, man. Yeah, just hanging out, feeling good. Just working on little projects here and there. I'm getting a lot of joy out of the stuff that I'm building all the time right now. So the problem is I have too many little projects, which...
1:02 Wes Bos? Man.
1:02 Scott Tolinski? Oh, it's a killer.
1:04 Wes Bos? Story of my life. I like, I was like sitting down, like, like looking at like what I should do during the year. And like, that's like the, my biggest Achilles heel is that I love and am interested in too many things. And I love learning. I love going down all these rabbit holes, but there's like a certain balance of like actually finishing stuff.
1:23 Scott Tolinski? Yeah, finishing stuff. And actually like dedicating all of your focus to one particular thing for a period of time. Like, over, over the break, I probably spent like five days in my, my basement, in one room in my basement. And I, you know, went through all of, it's like our storage room, which we don't store a ton of stuff. But I like went through every single tub we had, pulled it all out, did all that stuff, built a workshop, did everything organized. 3D printed a ton of stuff. And it was just like, I spent so many days just dedicated to that one thing. And it was like the, the most focused time I've had on a single project in quite a little while.
2:06 Scott Tolinski? And the results spoke for themselves. Yeah, it came out really good. So.
2:10 Wes Bos? I think the AI has made the software project, like, ooh, look a squirrel.
2:15 Scott Tolinski? It's made so much worse. It's made my stress around it worse.
2:19 Wes Bos? Yeah.
2:21 Scott Tolinski? Yeah, I get like stressed out by how many term windows and stuff and projects I have running at once.
2:26 Wes Bos? Ain't easy, ain't easy.
2:27 Scott Tolinski? Ain't easy.
2:36 Wes Bos? All right, let's get into the first one, which is Stack Overflow officially dead. This is from Stack Exchange website itself. Stack Exchange is the company that owns Stack Overflow and the thousand other like question and answer websites. And they have the ability to just simply query the data. And this is from when Stack Overflow started to current day, how many questions per day have been asked? And it's crazy to see that it's just been spiraling trending to several thousands per day today, that 321, that's, that's just today, right? But like currently only having thousands of questions asked per day, whereas like at
3:20 Wes Bos? the peak, you can see like the peak of COVID during lockdown, they were getting what almost 200,000 questions and maybe in 2014, they had, had that as well. And they've just been spiraling right on the way down, which is kind of sad, but also just like, that's the way the cookie crumbles. And I'm curious, what do you think? Like, why did this happen?
3:43 Scott Tolinski? Yeah, well, you know, obviously the answer is AI. Most of the things that you would ask on Stack Overflow, the way that it would work is that you'd post on Stack Overflow first, well, maybe you'd search on Stack Overflow and maybe there would be an answer, or maybe there would be, maybe there would be a question and then possibly an answer, possibly an answer that is accepted and not correct, or accepted and not relevant, or no answer accepted at all. Or you would see, okay, this question hasn't been asked, let me post it on here and then wait like a week for somebody to possibly respond to it. And if they did potentially help you, like that, that loop has been replaced by just asking AI, which honestly is, is better.
4:30 Scott Tolinski? In my, my opinion, the way I use this stuff, I, I found it the loop to be so much more efficient. And you're not going to get some snarky nerd who thinks they know everything, who wants to use this as an opportunity to, you know, flex instead of be truly helpful.
4:46 Wes Bos? Now, Stack Overflow. That's the thing.
4:48 Scott Tolinski? Yeah.
4:48 Wes Bos? Is that I stopped using Stack Overflow probably, I don't know, maybe eight, 10 years ago. That was probably at the height, you know, like 10 years ago, that was what, 2016? This is probably when I stopped using it.
5:02 Scott Tolinski? Yeah.
5:02 Wes Bos? And I stopped using it just because of the moderators would, would be like, didn't ask the question properly or not enough info. And they would just shut it down or duplicate question, which is the most frustrating thing in the world. And a lot of, I think a lot of those people moved to Discord, Reddit, all of that type of stuff. So while it was a really good spot, I used to spend all day on Stack Overflow, answering people's questions, trying to get my points up. And it was so fun to try to beat people to it. The questions were answered very quickly on there, but.
5:37 Scott Tolinski? Yeah.
5:37 Wes Bos? I think the combination of AI, like when, when did the AI stuff really pop off?
5:43 Scott Tolinski? Oh, God.
5:43 Wes Bos? Like the middle of 2013? Yeah, that's the big.
5:46 Scott Tolinski? Yeah, that's the, yeah, you could see right on this graph exactly where. Because even like for simple questions, and then once the AIs got access to the internet, once they got access to documentation, it was so much easier to, to, to get a question answered that way by pasting in the link to the, the source code and asking questions of the source code. Because I, I do think that is like such a thing for me, I was admittedly always on the cutting edge of stuff in a way that was negative on Stack Overflow. Because if I wanted an answer on this beta software, there's no, I'm, I'm, my answer is to look in the source code myself and figure it out.
6:26 Wes Bos? Yeah, you're not finding anyone to, to help you there.
6:29 Scott Tolinski? Yeah.
6:30 Wes Bos? One other thing. About AI answering the questions for you is that like we've said it ourselves on this podcast many times is if you want help, you have to make a distilled example of the, you got to reproduce it outside of your project and you got to like make a whole case and like figure it out. And often, often while you're doing that, you'll find the bug. But AI is really good at finding the bug in context for, for what you're actually doing. And sometimes that's often that's way better.
6:59 Scott Tolinski? Yeah, I agree. Stack Overflow, it is sad because that's such a place in time. What could Stack Overflow have done to have prevented this? They could have implemented an AI solution that would have answered a, or attempted to answer the question instantly or something, right? And then make that public. But ultimately like-
7:22 Wes Bos? It's odd that they provided all the data for everybody to, to sort of eat their lunch. They were sitting on a gold mine and doesn't look like they did much with it, unfortunately.
7:33 Scott Tolinski? Yeah. Yeah, it's crazy. I, you, you got to wonder like what they could have done differently because it does, from an outsider's perspective, it seems like they just kind of sat there and watched the, uh, the trend go downward. Like the, the writing's been on the wall for years for them. So I think they've had an opportunity that, uh, either they just didn't care to, to solve or weren't able to come up with something. And if you want to see all of the errors in your application, you'll want to check out Sentry. At sentry.io/syntax. You don't want a production application out there that, well, you have no visibility into in case something is blowing up and you might not even know it. So head on over to sentry.io/syntax.
8:14 Scott Tolinski? Again, we've been using this tool for a long time and it totally rules. All right.
8:20 Wes Bos? Next one we have here is micro quick JS. So this is from the creator of FFmpeg. And we've talked about Quick JS, uh, in the past. So Fabrice Ballard, he's the author of FFmpeg. He just said, um, I'm just going to write an entire new JavaScript engine called Quick JS. Absolute genius, like really knows how, how a lot of these lower level things work. And now he has put out another thing called MicroJS. And this is a JavaScript engine targeting embedded systems, meaning that you can then run JavaScript on a microprocessor.
9:01 Wes Bos? So imagine you're running this on like an Arduino or ESP32. We've talked about these very like, like, like five, $10 little microcontrollers are running in very memory constrained applications, which is absolutely nuts. It's not full JavaScript, right? There's no browser in it. It doesn't even have ES6. So it's a support of a subset of JavaScript close to ES5. It implements stricter mode, which is more error prone and inefficient. So why? Why would somebody want to run JavaScript on a microcontroller? Do you have any, any thoughts there?
9:44 Scott Tolinski? Because they like to put JavaScript into everything. Honestly, I don't do any work with microcontrollers. So I'd be interested in what you would have to say here more than what I would.
9:53 Wes Bos? Yeah. So in the microcontroller world, if you want to program one of these things, hold on, let me grab something. All right, so I got, this is something that I got, I got for Christmas and it's basically a little robot that can walk and run around.
10:09 Scott Tolinski? Very cute.
10:10 Wes Bos? And it, it has like lots of different I/O outputs, right? There's a little LED screen. There's two little proximity sensors on it. There is a little speaker on it. There is four servos. You can kind of hear it so that it will be able to walk. And there's a whole bunch of different input outputs. And the whole thing is powered by this little microcontroller. This is a, what is this? Raspberry Pi Pico, right? And the code that runs on this type of thing, and the same thing with the badges at GitHub Universe. Do I have, I have mine.
10:44 Scott Tolinski? I got mine somewhere too.
10:46 Wes Bos? The badges that run on from GitHub Universe, this is running a RP32 something. And similar, the years before, on GitHub Universe, they had another one, which also has a similar Raspberry Pi controller. Like, these are not like the Raspberry Pis that run Linux. These are like lower level microcontrollers. And in order to code on these, you sort of have a couple of options. You can go like the Arduino route, which is what a lot of people do. And in that case, you're writing some C in order to make it happen. Or you're going to be loading them, the GitHub badges run MicroPython.
11:27 Wes Bos? And the benefit to that is when you're running MicroPython, it's a little bit easier to actually write this stuff. Because not everybody is as well versed in writing and compiling C and being able to reach out for libraries. But they are pretty good at writing Python.
11:45 Scott Tolinski? Yeah.
11:45 Wes Bos? So I think the whole idea with this micro quick JS is that you'll be able to run this on on microcontrollers with writing JavaScript. So a lot of people in the comments of the release of this thing were saying like, this is a good plugin scripting language. So if you want to allow people, like, let's say you're writing an LED controller and you want to allow people to write a plugin that will do things. They could then author that in JavaScript and you could just run it. There's no compile step, but you're not going to upload C to a microcontroller that's already running. But you might upload a little bit of JavaScript to that, right? It's a really nice scripting language. So it's easy to write.
12:30 Wes Bos? It's sandboxed, meaning that like, there's no network, there's no file system. You can allow people to just give you JavaScript and it will run it in a sandbox. So it's kind of an interesting space. I'm sure in the like industrial controls automation world, aside from like the like hobbyist tinkerers that we are, I'm sure there's a lot more use cases for it as well. So I'm going to keep my eye on this.
12:53 Scott Tolinski? Yeah, yeah, I'm interested, but I also, man, there's just one more thing that I don't have time for is honestly what it is. Holy cow.
13:03 Wes Bos? Yeah.
13:03 Scott Tolinski? All right. The next one here is Open Workers. Have you liked the idea of CloudFlare workers, but also like to self-host? So Open Workers, what it does is it lets you run JavaScript in a V8 isolate in your own infrastructure. It's basically the exact same experience as CloudFlare workers, except for you're not locked in to the CloudFlare platform. Which again, I think this is like a nice for the, the type of person who might be interested in something like Coolify, the self-hosting, owning your own infrastructure, those types of things, while also keeping
13:45 Scott Tolinski? the more modern kind of development flow, being able to have individual functions that are, are running the way these are. It also does more than just, it seems like it does KV storage, it does Postgres, it does R2, S3, cron scheduling.
14:01 Wes Bos? bindings.
14:02 Scott Tolinski? Yeah, it, it does feel like a CloudFlare, an open CloudFlare replacement of sorts. Pretty neat. I, I actually hadn't seen this before you brought this to my attention, but is this something that you would take on yourself, Wes?
14:18 Wes Bos? Kind of interesting because people always, people always talk about like the Vercel lock-in. But like CloudFlare, as much as I love them, and I'm, I'm a CloudFlare stan currently.
14:29 Scott Tolinski? Same.
14:31 Wes Bos? It, it kinda is lock-in, meaning that like, you can run a lot of stuff locally. They've allowed you to like reproduce it locally and whatnot. But when you want, go to deploy, that, that's kind of your, your option, right? You gotta deploy it to CloudFlare. And, and the beauty of CloudFlare is that they have all these bindings, right? You need a database, you need key value, you need like blob storage, someplace to put your files and your images. They have all these things that would sort of just work together and it, it works really well and it's really, really nice. But like the, the thing that a lot of people don't talk about is that like, yeah, you're, you're kind of locked into hosting on CloudFlare and.
15:11 Wes Bos? That has not been a problem for like, probably not anyone, I'll say, but for a lot of people, like my bill is, is almost nothing. We went over that a couple of shows ago as well. Um, but like, there's always that thing of like, what happens? Um, if they decide to three times the price or whatever, you know, they, there's, if there's any rug pull, um, and you're locked in now.
15:34 Scott Tolinski? It doesn't look, Paul, I'm, I'm toast with CloudFlare because yeah, I hosted a lot of stuff there.
15:39 Wes Bos? But, but here's the thing that I think about this stuff is that the way that I'm building my apps that run on CloudFlare is I'm not directly using the like CloudFlare async fetch and env.kv.get and like all of their explicit bindings directly like this. What I'm doing is I'm building them in some sort of standard framework. Hano, Eliza, Remix, whatever. And then I'm using an adapter that will then expose those things. Whaku is what my own currently, or Wakku is what my currently, my website runs on and that's hosted on CloudFlare.
16:21 Wes Bos? There's a thin layer that sort of exposes those bindings to the thing. So like, I'm, my database adapter, I'm not using env.db.
16:32 Scott Tolinski? Yeah.
16:32 Wes Bos? I'm using, um,
16:34 Scott Tolinski? Yes.
16:35 Wes Bos? Drizzle and, and then I have a, an adapter that Drizzle will connect to the CloudFlare binding. And then if I wanted to move that thing somewhere else, no problem. You know, I build everything on web standards. CloudFlare implements web standards on all this and I would be able to move that thing somewhere else with very little work. So this is cool. I like this. Um, but I don't really see a, a huge use case for this type of thing.
17:03 Scott Tolinski? I think it's just people who want to own it. Yeah, they just want to own it. They want to throw it up on a, a VPS and, and own it. They want to throw it on their own machines and own it, but they don't want to
17:12 Wes Bos? Yeah.
17:13 Scott Tolinski? have to host everything as a straight up node app.
17:15 Wes Bos? I can still do that with all of my stuff and that's just because you're not coding things. Uh, anytime I use like a, an API that is thinking, looking like login, you just like, look at like, how do I? How do I make this so that it is, has an abstraction layer so I can switch it to something else?
17:33 Scott Tolinski? Yeah.
17:34 Wes Bos? This is really cool that somebody figured out how to do this as well.
17:37 Scott Tolinski? The one that's been getting me lately.
17:38 Wes Bos? Production yet.
17:39 Scott Tolinski? Yeah, yeah, well, yeah, I mean, this is definitely for hobbyists. It's for experimenters. It's for, for, uh, the one that's been getting me is hyperdrive. I have to use hyperdrive if I'm deploying something on workers that uses a postgresdb outside of, uh, like, uh, Neon or something like that, something that's not a, uh,
17:59 Wes Bos? Yes.
17:59 Scott Tolinski? Distributed DB. And that's obnoxious as hell. I'll tell you that because you need to get the hyperdrive URL from the event like process in, in the time of like, you, you can't just grab it from an env variable. I find the hyperdrive thing to be pretty annoying, uh, personally, but yeah, that's one thing I could do without. And sorry as Cloudflare goes.
18:20 Wes Bos? Yeah. And that's because Cloudflare doesn't have sockets in their implementation. So if you want to connect to something outside of the Cloudflare ecosystem, you either have to use HTTP, which is sometimes okay, but usually not for stuff like real time databases and whatnot, or you have to use their, their hyperdrive adapter.
18:45 Scott Tolinski? Yeah.
18:45 Wes Bos? But this also solves the problem of your database living somewhere where your code is not running because that's where a lot of latency happens.
18:54 Scott Tolinski? Yeah, yeah.
18:55 Wes Bos? And turns out Cloudflare, um, has fat pipes to most of the internet. So they're able to sort of, uh, make those connections as extremely fast between the two.
19:07 Scott Tolinski? I just wish the, the workflow development, like, I don't want to use Wrangler locally while I'm building locally for some reason. Like, I just don't want to do all that stuff.
19:16 Wes Bos? That's my feedback for Cloudflare team all the time is that I don't want to do weird stuff locally to make this work.
19:24 Scott Tolinski? Yeah.
19:24 Wes Bos? Don't make me do weird things. Don't.
19:26 Scott Tolinski? Cause then I gotta do an if statement, if it's local, then use the normal one. Otherwise use hyperdrive and yeah.
19:33 Wes Bos? Luckily, a lot of that, they're implementing a lot of their APIs. When Wrangler runs locally, it runs, it actually runs their JavaScript engine locally. Um, and why, why am I frigging the name of it? Worker D, that's what it is.
19:48 Scott Tolinski? Worker D.
19:49 Wes Bos? So Worker D is their JavaScript engine that runs. And that is open source as well, so that will run locally. But the, all of the bindings, most of them will run locally, but some of them still have to, like browser rendering will still have to reach out to the, the cloud and, and run it there.
20:10 Wes Bos? Next one, React Aria just got some new docs and I think this is fantastic. So React Aria is a React library for building components. And it has absolutely everything, you know, like autocomplete breadcrumbs, button, calendar, checkbox, color, like there's
20:30 Scott Tolinski? Yeah.
20:30 Wes Bos? Probably 80 of them here. And the cool thing about React Aria is it's 100% headless, meaning that it simply just brings the functionality and then you, you bring the divs. You bring the, uh,
20:46 Scott Tolinski? Yeah.
20:46 Wes Bos? You bring the CSS. And they just redid their documentation to be a bit more like, cause like people, people want that, but they also just want to slap a ShadCN
20:57 Scott Tolinski? Yeah, yeah.
20:58 Wes Bos? Component in there and, and call it a day. So they've redone, you've always been able to do that with React Aria, but they've redone their documentation so that it is dead nut simple on how to get this to work. So you can, you can choose vanilla CSS or Tailwind. And then it also uses ShadCN to install the component, which I thought, ShadCN is so interesting is people love ShadCN installing things.
21:24 Scott Tolinski? Yeah.
21:25 Wes Bos? Now, things that are not ShadCN are now using the tool to actually just like, what it does is it downloads the component to the component folder and then you can go in there and poke around and, and change it however you want.
21:37 Scott Tolinski? That's really the way to go, that rather than importing something from a package and using it that way.
21:42 Wes Bos? Exactly, yeah.
21:43 Scott Tolinski? I, man, I had a slight spat with somebody on Twitter over the holidays where
21:48 Wes Bos? Oh, boy.
21:49 Scott Tolinski? Uh, somebody was like, there was a whole thing about using, um, components to do
21:56 Scott Tolinski? Buttons as links and all this stuff. And either way, long story short, they they wrote this really obtuse and just like nasty looking React component just to simply output a link that could be styled as a button like that's it. And I I like, I've been left with this thought lately that it's like this component could just be a class like, come on, I get the isolation, I get all of those things. But when you have a component that is simply just a link styled as a button, you have lost the dang plot. Like this list box, this stuff is exactly what components are for.
22:40 Scott Tolinski? But stuff that is just a button link, like, come on, you're importing a stupid class like, just make this component could have been a class. That's the hill I'm gonna die on.
22:51 Wes Bos? I hate the over abstraction of components. Where it's like, you make like, oh, you want to have a link? You want to have text? You want to have like, the browser comes with these things built in. You don't need a component for that. It's yeah, just slap a class on it or slap some styles on it if you want, but don't over abstract these things because it's such a pain in the ass to debug.
23:16 Scott Tolinski? Yeah, yeah, it's crazy. And and it had some like really dismissive things to say about my opinion on it. Like this person just clearly doesn't understand like modern web development. It's like, brother, it's a freaking link. It is a link that's styled as a button. Uh, get over yourself. Uh, it's crazy to me. Yeah.
23:39 Wes Bos? I I have a component somewhere in my personal website where I have like an as. Uh, and I can, I can pass an as property and then it will render as a, I think a div or a span. Or it can also be a link. And in that case, it's it's good because it's a stylistic some, it needs to look like something. But in some cases, it needs to be like a proper link. And in that case, I think that's fine. But yeah, in so many cases, no, no, thank you.
24:11 Scott Tolinski? Yeah, no kidding. Cool. All right. Next one. Firefox. Can Firefox survive? Firefox had a big old thing about putting out, well, AI into their browser. And that pissed off what like the, the, the one, two, 1.2%, 2% of people who are still on Firefox, which is like their core audience is exactly the type of person who doesn't want AI shoved into the browser. And so they were pretty tone deaf in just saying, guess what, everybody? All the, the people who are still stuck on Firefox were putting AI in your browser and you're gonna love it.
24:54 Scott Tolinski? And then they had to walk back and say, okay, you can turn it off. But you have a whole thing about the new Firefox CEO here in addition to that. I, I'm just like Firefox shot themselves in the foot. Yeah.
25:07 Wes Bos? So the new CEO comes in to sort of like save the currently plummeting Mozilla Firefox. You know, they just been, haven't been doing great for a number of years. And then they just released this huge rebrand. And I don't know, maybe six months ago, they also released, they changed their terms of service to say, we will not. What was it? It was, it was personally said, we will not sell your personal data. And they changed it to something like Mozilla doesn't sell data about you. And people lost their minds over that, which is, I understand that as well. So like the Mozilla has not been, been doing great for people using that. And then the new CEO comes in here, ready to shake things up.
25:50 Wes Bos? And he's like, look, like we're an AI, not we're an AI browser, but like, that's essentially what they're going to be doing.
25:57 Scott Tolinski? We're putting AI.
25:58 Wes Bos? And everybody lost their minds.
26:00 Scott Tolinski? Yeah.
26:01 Wes Bos? And honestly,
26:03 Wes Bos? I don't think you guys should be giving Mozilla a hard time to be doing this thing, because I think there's probably two outcomes here. Mozilla keeps, and I say this as a huge Firefox user, I love Mozilla as like a browser engine. I think we as web developers need it and it is so important to the web ecosystem. But most people don't care. They do not care about that type of stuff and people are not choosing Firefox because of their flex box frames per second animations or things like that. People are choosing browsers based on on features, on what they do for them. And if Firefox is going to be able to survive this, right? There's two options.
26:46 Wes Bos? Firefox continues to be the browser that they are and they're going to slowly lose money. Most of the money from Firefox comes from Google Ads. More people are using AI. They're not searching Google for these things. So that money is going to go away and then Firefox will just cease to exist. Right? So we could keep going that way, but that will probably only be another year or two. And then Firefox will no longer be a thing at all. The other option is, is like, can we actually try to get some users that care about this stuff? We have the neck beards and the techies, but you can't run a business on those people. You need actual people who like regular users who care about features of the browser.
27:30 Wes Bos? And privacy is one of them, but unfortunately, regular people just want to do things like shop and get discount codes and have pages summarized for them and all of these types of things as well. So as much as it kills me, literally every other browser is going to be doing this as well. And I think it's kind of necessary in order to to keep it up.
27:54 Scott Tolinski? Yeah, Firefox is damned if they do and damned if they don't. Because if they add this stuff, their core audience, which is the only people still left using Firefox, they're going to get pissy about it. They're not going to be happy about it. And all of the paper clips in the comments are going to go nuts. If the
28:13 Scott Tolinski? If they if they don't add this stuff, like you said, the users are just going to keep on leaving because other browsers, more better features, more hype, whatever. What were their choices here? Like if they, like I said, they don't do it what they're holding on to this. It's the same kind of predicament that like Arc had where Arc was a niche browser for nerds. And I loved Arc. Arc is my favorite thing. But the Arc team, the browser company realized that they were never going to get this big market share in their current nerd focused way. Now, their goals could have been, oh, we'll just make a paid browser.
28:54 Scott Tolinski? We'll have this thing serve a niche community, but they wanted to reach more people. So they took an approach and made something different that had more mass appeal. I would argue that it's not been successful for them. But like in the same regard, that was the choice. It's like, do you keep your nerd audience who, you know, demands privacy? They demand, you know, all of the things that Firefox brings you. Or do you try to go out and get a bigger piece of the pie? And they made their choice. So, yeah, I don't know. I wasn't going to use it either way.
29:33 Scott Tolinski? So for me, this is like, I, this to me is like, they're, they're trying at least. They're trying to survive.
29:43 Wes Bos? Totally. And I think from a browser engine perspective, I really hope they make it out. Cause as much as I get mad that Firefox is lagging on CSS anchor. We don't want to be in a world where there's only one browser engine. I wish it was just Chrome and then as soon as you get that, oh, now this one company has, has control over, over everything. And I trust Google for that type of stuff, but that happened with IE as well. We don't want that for the web. And I don't know that a lot of these new browsers are being good stewards of that. You know, like, there's a lot of people forking Chrome right now, which.
30:26 Wes Bos? Which is taking a lot of people off of using Chrome. And like, what does that mean for the resources that are like, are they, Google spends billions of dollars on engineers to actually develop Chrome, you know? Like, hopefully, OpenAI is actually putting some engineering talent on, on pushing this stuff back, you know, and, and upstreaming a lot of the, their fixes back into Chrome. I don't know if that's the case yet or not, but we'll see.
30:54 Scott Tolinski? I know. I feel like.
30:55 Wes Bos? Although nobody's using the OpenAI browser, you know, that thing sucks.
30:58 Scott Tolinski? Yeah, that sucks. Yeah, I, I know. I, like I said, I don't even download it. I'm, I have, I have literally 30 browsers downloaded on my computer and I don't even download the OpenAI one. I do not care about it. Yeah.
31:09 Wes Bos? Yeah.
31:10 Scott Tolinski? Yeah.
31:11 Wes Bos? Speaking of alternative browser engines, Apple now is allowing alternative browser engines in Japan. So, under Japan's Mobile Software Competition Act, Apple must permit iOS browsers in Japan to use browser engines other than WebKit, such as obviously Blink or Gecko. Those are the other two big engines. Blink is, is Chrome's, Gecko is Firefox. Um, so this is kind of interesting, although, let's talk about it. This happened, I looked it up, two years ago.
31:46 Scott Tolinski? Two years ago, yeah.
31:47 Wes Bos? The EU mandated that this exact same thing in the European Union. And guess how many alternative browser engines are our? Not browsers, browser engines, meaning that like alternative browsers that use different, something not WebKit on, on iOS. Guess how many?
32:03 Scott Tolinski? I'll tell you.
32:04 Wes Bos? Zero.
32:04 Scott Tolinski? Yeah, and there's a good reason for that. There's a good reason for that. It's that because that's the EU and now it's the EU and Japan. So maybe there's a bigger opportunity there. But until the US has, has this available to them, there is no, there is zero incentive for Google to develop a web, all right, not a WebKit, a Blink version of Chrome just for those markets. Right?
32:32 Wes Bos? Yeah, you're right. That's two different.
32:35 Scott Tolinski? Now they have two, now they're forking their browser with two different rendering engines, uh, just for mobile. I want WebKit, uh, gosh, I, Google used to, Chrome used to be WebKit. I'm still like, my, my brain is still there. I want Blink.
32:48 Wes Bos? You can't change that.
32:49 Scott Tolinski? I want Blink on iOS. I do. I seriously do. And maybe Gecko, I don't really want Gecko on iOS if we're being honest, but I want Blink on iOS. And like, ah, come on. In the current US political landscape, there is 0% chance of any regulation happening to make Apple do this. And Apple is not going to do this without a regulation telling them to, because there's no incentive for them. They want to own the mobile space on iOS. And as long as Apple owns the mobile rendering space on iOS, you're missing out on the world of browser APIs and experiences that you could have off of WebKit.
33:33 Scott Tolinski? Like, come on.
33:35 Wes Bos? Bluetooth?
33:37 Scott Tolinski? Mini, all kinds of stuff. Like, yeah, it drives me nuts. And then, and then we all have to support, you know, Safari's jank and issues all the time on every mobile browser. And there's really no incentive to use other mobile browsers other than Safari unless of course you like want to sync your bookmarks and stuff and then you just pick the browser. They're all the dang same. It's annoying.
34:05 Wes Bos? Yeah. I will say, one good thing that has also come out of these things is the part of these mandates is not just being able to use different browser engines, but it's also to make the users have a choice. You set up an iPhone, you choose which browser you want to use. You don't just default to Safari and you have to go download a new one. And I think that's good. I think that's, even though they're still rendering with WebKit under the hood, I do think that there's
34:33 Scott Tolinski? Oh, of course.
34:34 Wes Bos? way better experiences when you're switching between the two. I use Edge on everything right now, but maybe I should switch for for 2026.
34:44 Scott Tolinski? Yeah. You use Edge on your phone?
34:47 Wes Bos? Yes. Yeah, I do that because I just, I want my history synced, you know? If I'm, if I'm like, oh, I was on a website the other day. I want to be able to start typing it in and
34:55 Scott Tolinski? Yeah, I get that.
34:56 Wes Bos? I found that. Or if I have something on my phone, I can open it up on my desktop really easily.
35:01 Scott Tolinski? I was using Arc Mobile, which honestly sucks. They had like a so weird feature where like a phone dialer would pop up even, it's like their Arc. Assistant, oh my God, whatever they were doing with it, it was just idiotic. Uh, but I, I've just been on Safari and I don't get my bookmarks synced and I'm just been, you know, living with that.
35:22 Wes Bos? You've been fine. The one killer thing that happened maybe six months ago is you can now use Apple Pay in other than Safari.
35:30 Scott Tolinski? Okay, yeah.
35:31 Wes Bos? Best. I hate when, when web checkouts don't have Apple Pay. And then they're like, the one password credit card never works on a mobile, you know? And then the like Chrome or I, I have the like edge Chrome autofill as well for my credit card. And then it asks me for like the three digit number. I don't know that number.
35:51 Scott Tolinski? Bro, yeah.
35:52 Wes Bos? Fill the whole thing in. Just give me Apple Pay. So beautiful.
35:56 Scott Tolinski? Yeah, I know. Man, I'm using uh, Helium as my, my browser on a day to day basis. And I would love just a Helium on iOS. Cause really what's it, what's it gonna do? It's just sync, just to sync my bookmarks. That's specifically it, but yeah.
36:10 Wes Bos? Yeah. On a lighter note, this is really simple. Fractured JSON JS is a JSON formater that formats JSON in a way that is, is human readable. And I absolutely love this because the one thing I hate about prettier and, and all of those like prettier compliant ones is they put arrays on like one item per line and that drives me nuts. Especially when you have things like, like latitude, longitude, or like you've got coupled values. This formats it in a way that is much more human readable. And, and as humans, we often read JSON. So this is just a little.
36:55 Scott Tolinski? Your code still?
36:57 Wes Bos? Uh, sometimes, you know, sometimes I read it. Yeah.
37:01 Scott Tolinski? That's a little joke. Um, is, does this work with prettier or any of that stuff? Or is it just a tone thing?
37:07 Wes Bos? I was just looking into it and I was like, someone's gotta write like a plugin for Oxlent. Or what is the other one that we're? That clip got pretty popular over the holidays. Biome, yeah. Someone's got to write like a, it's in JavaScript. Someone's got to write a plugin for Biome or Oxlin that does this. Cause I love this. And let me choose if I want my arrays to go on one line. Or like, I don't necessarily want them on one line. Sometimes I do, cause they wrap. But I, I want to set like, I don't know. Put, put five items on a line and then wrap to the next line.
37:41 Scott Tolinski? Make it a dang matrix. Let me look at that.
37:43 Wes Bos? Yeah. Exactly. I thought I would cover this, cause I've seen this pop up several times. This is a new Chrome permissions dialogue. Look, and they'll say like, this website wants to look for and connect to any device on your local network. And that's kind of spooky when that pops up. So this is a new-ish.
38:02 Scott Tolinski? Wired.com wants to connect.
38:03 Wes Bos? Been rolling out in Chrome. Permission for local network access. So when a website wants to be able to ping, fetch whatever, like a, a local domain name on your network, it's going to now ask for that. And you might be saying like, what the hell is the point of that? Turns out, bad people were, were using this for, for malicious intent and for ad tracking. Because a website like Wired.com, I don't know why this is popping up on Wired.com. But they are obviously doing a fetch request to like a 192.168 or like a .local domain name or a local host.
38:48 Wes Bos? And if things on your network are not locked down, or, or if you simply just want to check if they have them there, right? Because you might get like a cores error, but you know, ooh, that device does exist there. Then you know things about that, your user's local network. So like local routers, you could ping IOT devices that you know have security vulnerabilities. You could ping that don't have cores set up. Lot of stuff like that. So this, in my opinion.
39:19 Scott Tolinski? Interesting.
39:20 Wes Bos? Is a very good, cause.
39:22 Scott Tolinski? Oh, yeah.
39:23 Wes Bos? Almost never. Does a website actually need to access your local host? And if it does, then, then you would know, yeah, of course. Like, I'm using this W LED app and it's going to search my network for other instances of this LED controller. Or I'm going to try to connect to like a local printer. Or I'm, I'm running this thing on this port, but my backend is on like local host 4568. And that's in development mode. That's, that's going to be fine.
39:54 Scott Tolinski? Yeah, this is great. Yes, I do. I do like this as much as I hate pop ups for all kinds of things. I certainly want to know if wired.com is trying to access something on my local network for some reason.
40:03 Wes Bos? I do wish that, like, you know, on, on Mac OS, when you get a prompt or on iOS, when you get a prompt, they have a little description of why they want to use this. I do wish you could put that in, but then I guess like websites probably could maliciously be like, we, we do this to pet bunnies.
40:24 Scott Tolinski? Yeah. We're not trying to do it. We're not trying to footprint you or whatever. Yeah.
40:30 Wes Bos? Footprint? Fingerprint?
40:34 Scott Tolinski? Don't you leave footprints?
40:35 Wes Bos? No, well, I guess, but the word is fingerprint.
40:40 Scott Tolinski? They did not leave footprints.
40:42 Wes Bos? Footprints, bread, there's breadcrumbs. Is that what you're thinking about, like Hansel and Gretel?
40:46 Scott Tolinski? Leaving breadcrumbs, footprints in the sand. I was thinking footprints in the sand.
40:51 Wes Bos? Yeah. I can see it.
40:53 Scott Tolinski? Yeah. Hmm. So that is the news for you today. What do you think? Do you think Firefox has shot themselves in the foot? Or do you think it was totally necessary? Do you think we'll actually see an alternative rendering engine on iOS sometime soon? Let's say even within 2026? Leave your predictions below. Let us know what you think. Let's get into sick picks and shameless plugs. Uh, I got a sick pick for you today, Wes.
41:22 Wes Bos? All right.
41:23 Scott Tolinski? I. Got a little tur- uh, it's a terminal e-ink screen.
41:28 Wes Bos? Ooh.
41:29 Scott Tolinski? Um, and this thing i- is dope. There's a little battery that lasts a long time, like months. And the reason why this lasts months is because this thing is not updating often. It is like hitting the web once every X amount of times. And the less you have it hit the web, the more, uh, the more battery you're going to gain from it. But I have it set up with a number of plugins and it will change the screen every 15 minutes or something like that. This one is our YouTube stats. And so this will have our YouTube stats. It will have my calendar. It will have, uh, there, the, the web, there's a web app where you can connect plugins.
42:14 Scott Tolinski? You can write your own easily enough. I have my GitHub contributions. I have how many days are left in the year. I have when is the next hockey game? I have all kinds of stuff on here, uh, that it just sits on my desk and it's a, just a nice little e-ink screen and it's just kinda cool. It's a nice little thing to have there with, with no backlight and no concern about keeping it charged like at all. I've charged it once since I got it. It's just been sitting there. Really cool.
42:41 Wes Bos? Those e-ink screens are so, so cool. I often, I look at them at the, at the grocery store, you know? They've got them as the price tags at all of our grocery stores.
42:51 Scott Tolinski? Yeah.
42:51 Wes Bos? And you, you guys have that, right? All your, your price tags are in e-ink now?
42:55 Scott Tolinski? I don't think so.
42:57 Wes Bos? No?
42:57 Scott Tolinski? Now, uh, now where I've been shopping. Yeah. Not at, at Koger or Costco.
43:03 Wes Bos? In Canada, I would say, not, not Costco, but pretty much the rest of our stores have e-ink, which also drives me crazy because it means that they can run an algorithm on how much are people buying? And just change the prices.
43:16 Scott Tolinski? Change dynamic pricing.
43:18 Wes Bos? Willing, yeah.
43:19 Scott Tolinski? I, you know, I haven't been to Whole Foods or as my in-laws call it, Whole Paycheck. Uh, I haven't, I haven't been to Whole Paycheck in, in a little while. So there's a chance that they have it there. I don't know, but yeah, at, at Koger or Costco. I don't see them.
43:33 Wes Bos? But I see them at the grocery store all the time and I've, I've looked into whether you can just use those because they're relatively cheap if you just put them in your pocket. Um, I'm joking. They're on, on, you can buy them on eBay relatively cheaply as well. Um.
43:48 Scott Tolinski? You stealing ink, Scrims?
43:50 Wes Bos? No, but they're like, it's encrypted and nobody has cracked the encryption, which is probably also good that someone can't come into a store and just send like data to these screens.
44:03 Scott Tolinski? Oh, yeah, I always love it when you see like, uh, somebody has hacked the road signs to like, change, uh, you know, change it to say fart or something, who knows? That's always really funny. Yes.
44:14 Wes Bos? Um, I'm gonna sick pick this little Arduino kit that I got for Christmas from my mother-in-law. This was really cool to build, so it's a little kit that allows you to build a cat and it comes with a Arduino. It's actually an older Arduino, um, but it's, it's still worked totally fine. And then it comes with like tons of different I/O inputs and I, this was a very good learning opportunity. Um, I'm fairly comfortable with this stuff, but being able to, people are always asking me like, what do I, what do I do to, to learn like microcontroller stuff? And I feel like this is probably a good kit, um, because it takes you step by step, but it
44:58 Wes Bos? also gives you appreciation for writing the code, um, with all of the different I/O inputs, you know, like how do you control a speaker with a microcontroller? How do you control servos? Um, how do you control a little LED matrix? There's, I don't know, probably seven or eight different I/O GPIO inputs on here. And it was really fun. The kids loved doing it with me and, uh, it's one of those things that now that it's done, I'm sure I'm just going to start pulling parts off of it for other stuff that I want to work on.
45:29 Scott Tolinski? Sick. Cool. Let's shamelessly plug the Syntax on YouTube. If you're not watching this on YouTube, check us out at Syntax. Uh, we're at Syntax FM on YouTube. We do more than just the video podcast. We do all kinds of stuff. We do fun competitive coding games. We do, uh, just hijinks. CJ Reynolds does incredible videos where he goes deep on topics. He, he's got, uh, one on dev containers. He's doing all kinds of really cool stuff this year. So follow us on YouTube. And if you're watching this on YouTube and you're not subscribed, go ahead and just, just pound that subscribe button, like, subscribe, thumbs up, hit the hype button. There's like 18 buttons now you can hit, hit them all. That'd be great for
46:14 Scott Tolinski? us. Thank you so much. Base.
Previous 970: Why Did Anthropic Buy Bun? Next 972: These Things Make Your App Feel Like Crap on Mobile Page updated October 31, 2026 · transcript and summary are machine-generated