Open Models Replace Big AI ⟡ Every New Browser Feature ⟡ Death of Stack Overflow ⌁ Syntax Weekly ⌁

[00:00:00]Oh, welcome to Syntax. Today we are talking about all kinds of stuff. Wes has been gone for two weeks and now he is back and the entire web world has forever changed. Everything is different.

[00:00:14]>> Holy smokes.

[00:00:15]>> Yes, everything is different, folks. Uh we're going to be talking about all kinds of stuff in this episode. Vue 3.6 has entered release candidate and there's something called alien signals.

[00:00:27]That sounds cool as hell. We uh we're going to be talking about how you can stay on top of every single browser update that happens. We're going to be talking about an uh what is this? A a work tree alternative. Work trees. My gosh, those are one thing that I can't seem to get a handle on. So, I'd love to see that there is an alternative idea.

[00:00:49]And we're even going to be diving into openweight models. There's been a lot of new openweight models released. There's been a ton of crazy conversation around that and we're going to host a funeral for Stack Overflow. With me as always is CJ Reynolds and Wes Boss. What's up, CJ?

[00:01:07]What's up, Wes?

[00:01:08]>> Yo, doing good. Um, Wes, how's it going?

[00:01:11]You've been gone. How was your vacation?

[00:01:12]You feeling good? Ready to be back?

[00:01:13]>> Fantastic. Yep. Went to New York, went to the cottage. Uh, I've I was gone for what, two and a half weeks, and I feel like like I don't know what I don't know what soul is. Um, Fable is back apparently. Kimmy, is that Kim Kardashian's model? The hell? Like, I can't Antelli's how much stuff when I was gone.

[00:01:34]>> Hey, uh, Randy, we got some notes of some echo in the chat. So, >> let's see.

[00:01:45]>> Um, let's work this out.

[00:01:48]>> Let's work this out.

[00:01:49]>> Yeah, I don't hear echo. Um, let me pull up the stream really quick and then >> Okay, >> we just got one guide.

[00:01:58]>> That guy's probably in a big room.

[00:01:59]>> Folks, let us know if there's any audio issues. Sound is good here. All right, that's all I need to know. One person says it's good. I'm good. I think that's great.

[00:02:09]>> Yes, >> Yuri. Yuri had two tabs open. Okay, Yuri, double check next time. Okay, >> double check. Hey, that's me with two tabs open all the time. Uh, yes. Kim Kardashian's AI model, Kimmy Anteneelli's AI model, all that good stuff. Wes, do you know Kimmy Antonelli is? Mr. Lando Norris himself.

[00:02:31]>> No idea who that is. Who's that?

[00:02:33]>> He's the the current first place Formula 1 driver and he's like 19 or 18 years old. He was a Formula 1 driver before he had his driver's license. That's crazy.

[00:02:42]>> Holy smokes.

[00:02:43]>> So, he's a he's a little young guy. Uh but he is going to win Formula 1 probably. What's up? Uh CJ, I think we're going to get into your story first. Talking about the latest release from the Vue.js folks. I'm stoked just because it seems like it's got some cool naming stuff and I'm down to hear about some cool naming stuff.

[00:03:03]>> Yes. So, for the uninitiated, Vue.js is the progressive JavaScript framework.js.org.

[00:03:10]Uh they're one of the OGs. They came out around the same time as React. They've been here all along. Um they're not new.

[00:03:18]And um they've been working hard on on some new stuff. So you >> I've never heard of it.

[00:03:25]>> I'm sure most people have, but yeah, you know, >> it looks vibe coded. Is there this big gradient on the text? What is this new?

[00:03:32]>> That's fascinating.

[00:03:33]>> JavaScript framework.

[00:03:34]>> We can look at the web archive. This was probably before LLM even existed.

[00:03:38]>> Yeah, it's created by Evan View himself.

[00:03:40]>> Evan View. Yes. Okay. But version 3.6 RC has been released. So RC means release candidate and that basically means um that the API is stable. It's ready for people to try. So if you're a user of Vue, if you just want to try it out, you can try out the RC1. But there's two big things that are happening in version 3.6. Uh the first is a major refactor of Vue/Ractivity. So when Vue moved from version two to version 3, there was a lot of internals that were reworked and they essentially made it so the reactivity piece, the piece for state management is actually a separate package that you could use without Vue if you really even wanted to. Um, but they've been hard at work refactoring it using a library called Alien Signals.

[00:04:23]And so Alien Signals is a signal library that's really fast, >> really, really fast. And there are other uh frameworks that are adopting it under the hood as well. Um, but the idea being that and and you'll also notice that this is not a major version bump, right?

[00:04:39]We're not there's no breaking change major breaking changes. We're not going to version four. It's just version 3.6.

[00:04:44]So once this is finally released, all view apps will instantly be just a little bit faster because it's using alien signals under the hood. Um, so that's one one bit of good news. And then the other bit is vapor mode. And they've been they've been working on vapor mode and talking about it for a long time. But the idea with vapor mode is you can essentially build your Vue components, your Vue apps into a bundle that does not include the virtual DOM, does not include any extra stuffs. It's just as little JavaScript as possible. Um, this idea was popularized by Spelt in SolidJS, but now we're getting that inside of Vue.js.

[00:05:27]Um, little things about uh vapor mode.

[00:05:30]So in Vue.js, JS we use SFCs or single file components and in version three they introduced this setup keyword essentially on your script you throw setup and that allows you to just define all of your state variables and functions at the top level >> and now you can add vapor on there and it's opt-in per component but it's also optin per app. So you could actually have a regular plain old view app where it works the old way with the virtual DOM. I will say an optimized virtual DOM. It's a faster virtual DOM than React. Um, but if you have, let's say, one component that needs to be really performant, let's say it's like a a data table or uh an analytics dashboard or something that that requires a lot of updates and would basically have overhead if there was the virtual DOM, you can just throw the Vapor keyword on there and instantly that one component will now use the the native compilation and be a whole lot faster.

[00:06:21]>> Um, and so it's opt-in, which is cool.

[00:06:24]And then the other aspect of this is if you're writing an app from the ground up, you can write it all with vapor. So instead of saying create app, you say create vapor app and then you essentially get a similar build pipeline as with spelt or solid. Whereas when you run v build, this is creating a really small bundle that doesn't include all of the view runtime, all the virtual DOM stuff.

[00:06:44]>> So why is this faster? Like like what's what's going on here? I understand like with spelt it doesn't have the entire uh virtual DOM to be able to diff everything. It it simply just compiles to divs or or elements that can be updated. Is that the same thing?

[00:07:03]>> Yeah, exact same idea. Essentially, uh right now if you're building a Vue app with Vue version 3, um it actually compiles into render functions. So the the the output of a a view build is very similar to React where you have these uh functions that get instantiated and then added to the DOM and then whenever an update happens like a state changes or an effect something changes inside of an effect the virtual DOM reconciliation kicks in to do the diff and then update the DOM. with vapor with vapor mode that it's very similar to spelt in that the built code essentially includes inline calls to update the view whenever a variable's changed. So that way there's there's no extra runtime that it needs to shift.

[00:07:47]>> Okay.

[00:07:48]>> Yeah.

[00:07:49]>> So instead of saying where this variable has changed, figure it out and update it, it's compiling it down to this span will need this number updated. And when that happens, call this function.

[00:08:03]>> Yeah, >> exactly. So it's more it's more fine grained.

[00:08:07]>> Um and so faster overall. Um >> I've always liked that spelt can can compile to that because just like the output of it is is like let alone performance, you know. Yeah.

[00:08:17]>> But the simp simply the output of this that they don't have to carry along the whole diffing library.

[00:08:24]>> Definitely. And I would say like early early days I wrote an app with Vue. Um, it was I wanted to have like the the hacker effect of like you type in a password and then it like slowly looks like it's revealing each letter like it's it's hacking into the thing. Um, but it was really slow.

[00:08:41]>> Yeah. One letter at a time, you know, like movie style. Uh, but it was really slow with Vue because of all the virtual DOM updates. And back then I was like, I'm just going to try rewriting this in spelt. I rewrote it in spelt almost like a one to one port, but spelt doesn't have a virtual DOM. it builds to more uh performant codes uh direct update codes and yeah >> the moment I built it with Vue it was instant it was like spinning really fast like I didn't sorry the moment I built it with spelt it was super fast out of the box so we're going to be able to do similar things with uh Vue vapor mode so I'm super excited >> sick uh love uh more improvements like that to Vue and I think more frameworks can embrace not having a virtual DOM so thank you to the Vue folks uh Let's get into the next story here. If you are done, CJ, are you done?

[00:09:29]>> I'm done. I'm done.

[00:09:30]>> Oh, yeah. All right. Get Work Trees. So, folks, before we get into this next story, which is Rift, get work trees.

[00:09:38]Uh, I want to say what we're doing here is we each brought some stories. What's been going on in the web ecosystem, what's been going on in the dev space, we're going to be talking about all kinds of stuff. Stick around because we're going to be talking about open weight models, which has been a huge topic of conversation in just a little bit. This show is brought to you by Sentry. Sentry is the perfect place to track all of your errors and exceptions and fix them. Uh, I'm currently tracking down a problem with our Slackbot, Tasty.

[00:10:09]Tasty told us this morning that he didn't have database credentials, which is BS. Uh, and I looked into it and it turns out the sentry log showed that we are on the free plan of Neon and therefore we have run out of space uh already. So, I got to figure out what the heck Tasty's been doing with all of our free database space. Oh, agents, we love them. So, let's get into the next story. But first, >> yes, >> the three of you, Wes, CJ, do you guys use work trees?

[00:10:46]>> I've tried.

[00:10:46]>> But CJ, you go ahead. Sorry.

[00:10:48]>> Yeah. Yeah. Yeah. I So, I have tried and issue with want like needing using work trees is you have to be um you have to really plan your app. You have to know what are all the things that I need to be working on and what can be worked on in parallel because the idea with work trees is you can basically two two different agents can be working on the same codebase and not interfere with each other. So I've tried it but because the amount of planning it requires up front I've kind of just settled back into just one sub agent at a time but essentially you know multiple agents can work on the codebase without interacting with each other. That's that's the idea with work trees right? Yeah, I've I've used git work trees more with with non-agenta coding where I simply just need to have two of the same code bases working. So, think of work trees as if you were to have two branches. Um, but you can literally have those both open at the same time. You don't have to switch between them. The way that it works is that it creates a secondary folder on your file system that has access to all of the files of that branch and then you can you can work on both of them. So, it's kind of nice when you can like reference both of them at the same time or if you're like uh if you need to do like a quick hot fix but you don't want to like commit or whatever, you want to have both of them over at the same time, then you can do that. So, I've I've used those plenty and I do reach for them every now and then in agentic coding. Um, but I don't find myself running so many things at once where I need to have them all run at running in parallel. Um, unless it's something like if I have like six or seven git issues, I want to fire them all off at once. Yeah, that's a good use for for a work tree where it could go and do that, right? But >> CJ and I 67.

[00:12:38]>> Oh, I didn't even know.

[00:12:40]>> I didn't even know. I'm so brain rotted.

[00:12:41]I didn't even know that I said that.

[00:12:43]Okay. Um, now let's talk about Rift. Um, Rift is from Anomaly, which is the folks behind Open Code. So, they kind of know what they're working working with here.

[00:12:56]And they have implemented a better alternative to get work trees. So, I saw this on on Twitter today. If you don't know what Rift did, Rift did an experimental alter alternative to get work trees. It creates isolated writable workspaces using copy on write file system primitives instead of copying your repository. So, one of the downsides to using work trees is that um your git blobs and storage and all of that stuff that is shared between um branches and and work trees in git. But what is not shared is it literally copy pastes every single file. And even if you don't have a particularly large um like like 10 gigs worth of code, you could have thousands and thousands of files and you know that copy pasting many many files can can take a little bit of time. So what this does is it uses some primitives that are in your file system. So on Mac OS um APFS most modern Macs are are using APFS or I would say all of them. Um and what that does is it uses clone file meaning that it will clone the file but it will not actually make a duplicate version of that file until it absolutely needs to.

[00:14:15]Um and what that allows you to do is is you simply can make a work tree much faster. It's not like oh wow this is amazing. I just think this is pretty nifty. It's not something that's built into Git just yet. So, people are looking at alternatives to it. Looks like it works on some Linux, all of Mac, and I I don't know if Windows has a primitive like this or not. I was looking into it. Maybe chat can let us know, but it be it I I don't see why you wouldn't dip into something like this other than it is experimental.

[00:14:48]>> Oh, that's a perfect uh thing for me to dip into. got the experimentals. I love it.

[00:14:53]>> Yes.

[00:14:54]>> Yeah. Just your description of it sounds great. It's like duh, why isn't Git doing that already? Um, one of the other issues I had with work trees is the fact that that sub agents would have to do a PNPM install in every single work tree.

[00:15:06]I wonder if this has any solutions for that. I'm I'm guessing not because maybe you you do need a separate node modules folder depending on what the agents are doing in each of the the work trees.

[00:15:17]>> Yeah.

[00:15:18]>> But yeah.

[00:15:18]>> Uh yeah, I I don't think so. Here there's some some hooks post create PMPM install frozen. So no, you still have to do a PMPM install. Although if you have all the same dependencies, that shouldn't be a major, >> right?

[00:15:32]>> Um time thing because you would have all of those files on your system already cached >> instant.

[00:15:38]>> Yeah.

[00:15:40]>> Nice. Have you tried it out yet or just early days like?

[00:15:42]>> Yeah, I uh tried it out this morning and like it works. I don't I don't know what to tell you. Um, it makes a uh folder here. Let me let me bring up my terminal here.

[00:15:58]>> So, I just did rift init uh riff create name rift test. I had a pseudo it of course. This didn't work right off the thing. And what it does is it creates by default it creates a folder in the parent directory called riffs. Um, that's always a weird thing about git work trees is like by default it will put it in the same folder that you're in. So like what you're working on is now a subtilder. I think that's part of the reason why people get confused.

[00:16:24]Ideally you would you create the work tree in like a parent folder or you have a separate work tree like temporary folder that you're working in. Um, so I just simply did that and I have a riffs folder and each of my riffs which would be a work tree are now now live in there. All of your node modules are in there as well. Um, all of your your git history is available to you. So like, yeah, it just worked. It wasn't anything like, "Holy smokes, this is hard to learn or this is amazing." It's like, "Oh, wow. That that worked. That was very fast."

[00:16:55]>> Cool. All right.

[00:16:57]>> If you look if you look at the docs for work trees, like I got lost in it. My agents got lost in it. Like like you were saying, I had to figure out, oh, create a folder, get ignore that folder, point it at that folder. Um, this kind of just takes care of it all built in.

[00:17:09]So, I like that.

[00:17:10]>> Yeah. Yeah. That's nice. So, we'll see if it uh it's again all of these like new ideas. You can't put too much weight into them. It just means somebody had a cool idea and then typed it into a box and it's now a thing. Um but you got to think like the folks at Open Code are thinking about how do we make our product better? And one of the downsides is that work trees for some people are very slow to create.

[00:17:33]>> Nice. Sweet. So, if you're just joining us, this is Syntax Weekly. Uh, we're getting into news, new libraries. We just talked about Rift. We talked about Vue 3.6RC. We've got some upcoming news about open weight models that have um are on par with some of the Frontier models. And uh, but before that, Scott has some stuff to talk about. Um, this is about staying up to date, right, Scott? What do you got for us?

[00:18:00]>> Yes, people ask me all the time about staying up to date on web stuff. And one of the resources that I have always been using is this MDN's update page. I'll share the link for this in just a minute here. But I want I I talked about this in my conference talk in Amsterdam and I had so many people being like I looked everywhere for that page. Where is it?

[00:18:24]Uh and then I was like, you know what, I thought it was easy to find. And I started pawing around the MDN website for about 15 minutes and could not find this page. So, I would assume that most people don't know about this or cannot find this page. I'll post a link. It is basically developer.mosilla.org/plus/updates.

[00:18:45]This page contains every update for every API JavaScript, HTML, and CSS as they get added to each individual browser as well as things like node. And if you're into consuming a lot of data or creating process to understand what's being added here, this is a really great resource to have. I wanted to call out a couple of really neat APIs. I saw that just were added into this. Um, well, before I do, it looks like I have a sandwich delivery. Sorry.

[00:19:17]>> Scott gets a sandwich delivery every single episode.

[00:19:21]>> Guys, this sandwich >> got some great swag. It's It's got It's got pesto and pushcido and matzah on here. Oh my god.

[00:19:29]>> Yeah.

[00:19:29]>> What a sandwich.

[00:19:30]>> We need an overlay with sirens and confetti when Scott gets >> sandwich delivery.

[00:19:36]>> Courtney opened the door and was just like laughing cuz she knew that I was going to be calling it out immediately.

[00:19:42]Oh, that's so fun. Um well, thank you to my lovely wife Courtney. Um so one of the cool APIs I wanted to show on this was text fit. And text fit does exactly what you would think it does. This is the property of my dreams. Scales the font size of tech no text nodes to perfectly fit the width of its containing box in CSS. It's in Chrome, baby. This is in Chrome. It's in here.

[00:20:10]Uh, it's only in Chrome right now, obviously. But here it is, folks.

[00:20:15]>> What? I had no idea this was coming.

[00:20:17]This is like a like a fit text, right?

[00:20:20]This is a fit test.

[00:20:21]>> Some might call a fit text a text fit and they have it's a text fit. Uh this was added to Edge July 1st. So just to give you some context there. There's so much cool stuff in here like flex wrap balance, uh flex float balance, text fit. This is just a really awesome listing to keep your eyes on at any given point and just scroll through here and just see what is arriving in the browser. Or again, you could always take this information, distill it down into like an agent table of things that you could tell your agent to use whenever it's supported, blah blah blah. You could do so much with this. Um, either way, I I just wanted to call out this specifically as a nice resource for you to have to to check this out. Rule row rule row rule was added as well. And then I clicked over. Oh, a popover hint was added to uh web view Android, which means that popover hint is being added to more things. Popover hint for making tool tips with popover uh nice and easy with low effort. pop over hint. Um, background border area, man. Like, what is some of this stuff? Every time I go through here, I get curious and fun. So, check this out. We'll have a link in the show notes description, folks. It is for um and and also I when I posted this on Twitter, people were calling out this is being an AIdriven design consider heavy border left on these things. I do want to say that this page has existed before that was an AI trope and I can say probably pretty confidently I think it's it's not that >> maybe MDN was like patient zero for this little fingernail border.

[00:22:08]>> Possibly. Yes, but very possibly.

[00:22:11]>> Yeah, this design needs to be able to collapse every single one cuz like I don't care what just dropped in Android Opera.

[00:22:21]>> Oh yes. Yeah, you kind of filter that.

[00:22:24]Um, you kind of got to know what you're looking for a little bit or at least go through here. Yes, there's too many things. And then also, uh, when there is a property like ro rule, it's like easier to just >> Google it than it is to >> than it is to like actually go somewhere from this. But just a good nice little viewpoint in what's coming up in the browser.

[00:22:46]>> Looks like a bunch of the sanitizer API stuff just got added as well. That's an interesting one I'm going to dip into.

[00:22:51]We've been waiting forever for like a native sanitization API for the DOM.

[00:22:58]>> I'm in. I'm in. I'm in. What also I'm into is CJ, you're going to talk to us today about open weight models. Uh it seems like when I was gone, absolutely everybody released some sort of new one.

[00:23:10]And you're going to give us a rundown of what? Inkling, Quinn, Kimmy, View, what else?

[00:23:17]>> That's the plan. Yes. So, if you're just joining us, this is Syntax Weekly. We're going over news and such. And the big news is that just just the in the past five days. Okay, so first of all, Kimmy K3 launched and this was on Thursday.

[00:23:33]Um, if you're not familiar, you might have heard of Kimmy K 2.5. Essentially, it's an an open model.

[00:23:38]>> Kimmy Kardashian.

[00:23:40]>> Yeah, exactly. But it's an open model that you can run on your own hardware.

[00:23:45]Um, but there all are al al there are also cloud providers that will give you access to these AI models. Um, and if you're excuse me if you're new to this space, essentially there are AI labs that are have been training models and releasing them out in the open uh for people to use and run themselves. Um, a lot of them are Chinese companies. So Kimmy is a Chinese company. We're going to talk about Quinn. They're a Chinese company as well. But on Thursday they released Kimmy K3 and they're calling it open frontier intelligence. So that's why we care about this. That's why this is news is up until now most of the open models have been decent but have never really been able to be on par with like Opus level models or GPT55 level models.

[00:24:30]But now with Kimmy, we're there and the benchmarks look good and the rumblings of people using it on Twitter look good as well. Uh but this is a 2.8 8 trillion parameter model. It has a million token context window. Uh, and it's natively multimodal, meaning you can pass it images, PDFs, documents, and it can figure it can basically figure out what's in them and respond accordingly just like you're used to when you're talking on chatgbt.com. It can do all of those things. Um, and the the thing to look at are these these coding coding benchmarks. Um now again benchmarks are just benchmarks but >> but if you look at these charts Kimmy K3 is sitting for a deep SW. So that's one of the benchmarks. It's sitting above GPT 5.5. It's like.5% better and just below Fable 5. Um and a few points below GPD 5.6 Soul.

[00:25:28]>> Wow.

[00:25:28]>> Yeah. And so on all of these benchmarks, it's either it's like in the top three and that's comparing it to Opus and Fable and GPT56.

[00:25:38]So the fact that you can eventually run this on your own, they're saying that open weights will be released by July 27th. So right now, if you want to use this model, you do have to talk to Kimmy's servers, which are hosted in China. So um don't do it on anything you wouldn't want them to have your source code of. Uh, but on July 27th, there'll be an open version, so you can run it on GPUs in the cloud. This is a massive model. This is not something you're going to be able to run on like a Mac Mini or anything like that. You need lots of GPUs. Um, but once they release the open weights, we're going to be seeing probably a lot of people uh making this model available. And uh, for instance, um, GitHub Copilot made Kimmy K2.5 available via their API. And so eventually you're probably going to get access to Kimmy K3 as well. Uh so yeah, that's the news with Kimmy. Um thoughts, comments? Yeah.

[00:26:32]>> Yeah. Well, there's a lot of chatter today about uh the Trump administration apparently considering blocking openw weight Chinese models. So, uh get your downloads in Canada, folks.

[00:26:46]>> Yeah.

[00:26:48]>> Um in the chat, I see people asking about quantized versions. um because no no model has actually been released for people to run themselves yet, we're not going to see any. But the idea with quantized versions is you can run bigger models on lesser hardware. But but even with a 2.8 trillion parameter model, the the best you're going to get is maybe like a 300 gigabyte model file. And that means you need like at like a at least 300 gigabytes of video RAM. So this not necessarily something you're going to be able to run yourself yet. I I think the like running it yourself is cool, but I think the exciting bit here is that this is significantly cheaper than the Frontier models and seemingly just as good. Like to give you some some pricing here, if you look on Open Router, the the weighted average, so they do open router has this thing called effective uh let me share my screen here.

[00:27:53]Open Router has this thing called uh effective pricing where they kind of like show you what the price effectively is because um how tokenization works and and pricing and caching and all of this type of stuff um makes the pricing kind of hard to compare. But if you look at it, it's about 50 cents input versus Fable is is $268.

[00:28:17]Maybe not totally fair to com compared to Fable. Um, and then the output $15 per million tokens whereas Fable is $50 per million tokens. So significantly cheaper and seemingly almost as good. I guess we'll we'll we'll see with that type of stuff. That's pretty exciting to see that type of stuff. And I will tell you, I've been in DMs. I'm in a couple NDAs right now of testing some like upand cominging models and I last week I got two emails from people saying, "We've delayed our model." Um, meaning that this came out and they said, "Oh, >> okay. We got to make ours a bit better."

[00:28:57]>> Yeah.

[00:28:58]>> Nice. Uh, a few more stats to share and then and then we'll talk about uh Quinn as well. Yeah. Anything to add, Scott?

[00:29:04]No, I I I think the openweight models like many people are saying like that's clearly the the future of things, but I think um just the word open kind of leads people to think that they could run these on local hardware and for the most part most people aren't going to have rigs that can run these things to get them anywhere near the uh the uh performance that you're looking for.

[00:29:25]>> No, but where I imagine is first of all, you saw we saw the pricing that that Wes just showed like it's significantly cheaper. And so if you're paying API costs at enterprise for cloud or open AAI >> this looks a lot more interesting to be able to run this yourself um because maybe you're an enterprise and you can afford the hardware or renting the hardware. So that's something to think about as well. It's it's really going to be it's going to give the the uh big companies OpenAI Anthropic a run for their money in terms of they have to figure out how to compete on pricing as well. Um >> can you speak on this comment? Someone says open weight does not equal open source.

[00:30:04]>> Thank you for teeing that up because I do want to talk about that. Yes. So o open source in the traditional sense is about licensing of software. It means uh for open source uh licenses like MIT or Apache it basically they boil down to you can look at the source code you can modify it you can do whatever you want with it. And um and then there are some open source models or open source licenses that say um you can look at the source code but if you make any modifications you must release those as open source as well. That's things like the GPL license. So that's about source code. Here we're talking about open weights. So the the process of actually training a model is long and intense, right? They have to they have to get all of the training data. They have to run it on GPUs and they run >> they have to distill all of anthropics models. Exactly that. We're going to talk about that next. But yeah, so basically there's this whole big upfront process that they do >> that at the end of the day results in a file that has billions of numbers inside of it. And so when we say open weight, they mean you can look at the weights.

[00:31:07]But the thing about neural networks is just looking at the weights doesn't tell you how it was trained or what they did special with it. You just can see the numbers. So that's what we say when we mean o open weight. Um, and then on top of that, these models also have various licenses like yes, you can run this for profit or you you can only you you can run this model, but you uh can't charge people for it. So yeah, that's a good distinction to make. We can't see into how they trained this thing, but we can see the numbers that the training process produced. Um, so something to think about.

[00:31:38]>> And like I'm curious like like you obviously wouldn't be able to take this model and then start charging people or I'm not sure what the license actually is, but like what about like if you were to like slap this into your product that uses AI, you know, there's the rules around that as well.

[00:31:54]>> Exactly. You'd Yeah, I'm I'm not a lawyer. You'd have to look at their their terms of service and licenses for each of these models, but that that's the kind of thing. The other cool thing you can do with these openw weightight models is fine-tune them. Uh and so like there's a whole ecosystem of uh taking an openweight model and then further training it on specific types of data to get specific outputs. So >> that's the other cool thing about openw weight is like we could take a a specific model and then train it really really good on all of the the syntax podcast transcripts and then it would be like and this is as opposed to something like retrieval augmented generation where you just like paste in the transcripts. You could literally fine-tune the weights to know more about how Scott and Wes talk and what topics they talked about in what episode. Um, or if you work at an enterprise, you could take all of your internal documentation and and everything else and fine-tune a model on that. So instead of having to wait for API calls, you have a model that knows your company uh inside and out, that kind of thing.

[00:32:54]>> Yeah.

[00:32:54]>> Remember everybody remember that was what everybody wanted to do. They're like, I can't wait till I can train my own model. Um, and that doesn't seem to be >> as as they got better and as rag got better. I don't it doesn't seem to be as popular, but I still I certainly can see the the use case for it.

[00:33:11]>> Definitely. Yeah. Somebody asked about the architecture. So, this uses a mixture of experts and it says uh spar sparsity effectively activating 16 out of 896 experts. Um, so a quick plug for my video on how AI works. Um, in that video I talked about the basically transformer blocks where data passes through the blocks, but there's a a different architecture that basically bolts stuff on top of that be like in inside of each block. And the idea is instead of activating all of those for every single prompt that goes through, they train it so that it only activates some of them. And what that means is that even though it's a 2.8 an 8 trillion parameter model. If there's only uh 16 active experts, it actually can run on uh lesser memory than would be required if all of the experts were were active all of the time. Um so there's that that aspect of it. Um I want to keep moving because I've been talking about this a long time. The other well another bit of news is u 48 hours after it was released they cut off access because it was getting so popular. Um, so I don't I don't know if they've reopened access since then, but yeah, it really took the world by storm.

[00:34:23]Um, it's exciting news. But on top of that, just uh three days later, Quinn, which is another AI lab, I believe by Ali Alibaba, which is like the Chinese Google of sorts. Is that accurate to say?

[00:34:36]>> Yeah. Yeah. They're they're they're huge big company. Yeah. They're they're in e-commerce, but also they're kind of like Google as well.

[00:34:46]Yeah. So they released Quinn 3.8 um and this is a 2.4 trillion um parameter model and so on par with this one that we just saw from Kimmy and um again they haven't released it for people to run themselves yet but you can call their API to to interact with it and um we're seeing similar similar benchmarks. Um, the other bit of news that's more interesting in terms of if you're from the US is this company called Thinking Machines. They released an open model called Inkling. And what's cool about this is they are a US-based company. So both Quinn and Kimmy and Zai and all of these others are China China based models that have trained on distilling larger models. So that's what Wes Wes talked about earlier where they're not supposed to do this because it's in the terms of service of like the anthropic API and the OpenAI API. But essentially what they do is they make requests to let's say Claude Opus 4.6 six with a bunch of of like questionans answer pairs to see how Opus 4.6 responds and then they use all of those uh question answer pairs that are in the style of Opus 4 4.6 to train their own model. So that's called distilling. You basically like prompt another model a bunch to get the data out of it and then you train another model using that data that you got out of it. And as far as we know, that's how these Chinese companies are able to produce their models and do it so cheaply is they're basically piggy back piggybacking off of all of the work that OpenAI or Anthropic have done. Um, but the cool thing about uh >> who is hold on who is piggybacking on the work that the rest of the world has done and put on the internet.

[00:36:35]>> Absolutely. I didn't I didn't want to lead into it, but you're totally right.

[00:36:38]And I'm I'm not saying that in from I'm not I'm not I'm not backing the the the the big AI companies because they also are have stolen the data or borrowed the data, however you want to say it, but >> it's kind of a weird world, you know, and like I I'm I'm busting their anthropics balls here, right? And like obviously they should be mad about this.

[00:36:56]I I would be as well if you spend billion dollars making Fable and then somebody just cheats on your homework.

[00:37:02]But >> Wes with a top rope ball buzz.

[00:37:05]>> Yeah, me. I'm just a little sore that my 300 blog posts got scraped. So, can't continue.

[00:37:11]>> Yes.

[00:37:12]>> Yeah. Um, so, uh, Inkling is a new model that was trained from scratch by a US company. So, the former CTO of OpenAI, I believe, uh, Meera. Yeah, previously a CTO at OpenAI, >> um, is now working at Inkling.

[00:37:29]And they released this, it was trained from scratch and they released it as open weight. So you can run it yourself or run it on hardware that you rent. Um, and it's not as big or as capable as Kimmy and Quinn, but it is from a US company and it was trained from scratch.

[00:37:47]It wasn't distilled as as as much as what as what they're saying. Um, >> and if you look at their initial uh benchmarks, uh, the one that they led with was actually design arena and that puts Inkling really high up. So design arena arena is a benchmark where these models build out websites and then the resulting websites are actually ranked by humans in terms of how good they look and how well they work. And in this one inkling is right between GPT56 soul and cloud opus 4.6. So in terms of design it's really good and it's open weight.

[00:38:19]Um once you start to look into like the software benches though it's not as good. Um it's probably more on par with Kimmy K 2.5.

[00:38:29]um trying to find a good good chart here.

[00:38:36]>> Yeah, it's a little bit all over the board in terms of all all of the benchmarks, but ultimately the news is we we have an openweight model from a US company. It was trained from scratch and this is also going to give OpenAI and Anthropic a run for their money because >> yeah, nice. you can run this and and also there there won't be necessarily like uh the restrictions being placed on Chinese models if you're in the US. So, >> it's good news.

[00:39:01]>> That's cool. That there's a company in here in Canada called Coher that does um they're making their own models as well.

[00:39:07]They're like the Canadian AI model maker that everybody is touting. And look, just looking at their website, it seems like they are really gunning for the enterprise because like that's going to be a the people that spend a lot of money and b these companies are not going to be allowed to use um a Chinese model. A lot of them are not. They're going to have to use something that is like like my bank, you know, my bank's not going to be able to hook up my bank account to uh Kimmy even though I probably already have. Um they're going to need to use something that's from Canadian soils.

[00:39:42]Definitely, >> man. This uh new Mac Studio with 1.5 terabytes of RAM needs to drop.

[00:39:50]>> It's looking real good right now, right?

[00:39:52]And I think Yeah, I mean, it's going to be a super expensive probably like minimum 10,000, maybe 20,000, but it's an investment in your future because you'll be able to run these openweight models on your your local Mac Studio, at least maybe some of the quantized versions eventually.

[00:40:09]>> Yeah.

[00:40:09]>> Yeah. Full weights uh for Inkling require 1.9 terabytes of just disc space. Okay. Where where's the RAM?

[00:40:19]>> Impossible. Apple doesn't even offer that.

[00:40:22]>> That usually corresponds to how much video RAM you need, but that also depends on um like the mix mixture of experts. Uh I do have numbers here though. So the they released two models.

[00:40:33]Inkling's big bigger model is 975 billion parameters with 41 billion active parameters. So it's a similar idea what we talked about before where not everything has to be active, meaning it doesn't have to use necessarily as much uh video memory when it's processing, but as a size comparison, this is like less than half the size of Quinn 3.8 and Quinn 3. So we know that they're not necessarily going to be uh or Inkling isn't going to be as good as those, but still pretty decent. And then they released a small model which was 276 billion parameters with 12 billion active parameters.

[00:41:09]>> All right, sick >> inkling and good branding. Good good looking branding.

[00:41:13]>> Yeah, I like the little Inkling stuff.

[00:41:15]So Scott, um I see you have some some some apps for us or >> Oh yes, and I ain't talking about appetizers here. I'm talking about whole ass meals of apps. You if you want to call this that. Uh I I have a couple of really cool apps. one of which is a CLI um called Mole. Now Mole has been talked about quite a bit all over the web, but Mole is really great because it's a Mac cleaning CLI that you can install with Brew and I like this so much that I was really stoked when they released a fullon app. Um then the app is really great. So, um, as far as like a better system process viewer, like I I use BTOP a lot to do my system process checking. Uh, but like nice guey for everything. Find and remove old files and uninstall stuff and delete caches and that type of this type of app has been done for a billion years with like clean my Mac and some of these other ones. And this one just feels so nice to use. And if you just want to use the CLI, you can. But I just wanted to shout out Mole at mole.fit.

[00:42:28]Really great little app here. Um >> this >> Yes, >> it's everybody. I I was I posted about I forget what it was, a tool that I use to delete old node node modules cuz I'm always always out of space on my computer. And usually what I do is I'll run this like CLI tool and I found like a 100 gigs of old node modules I didn't need. And then I would run Daisy disk uh to see like are there any like random weird files that are using up 80 gigs, you know, like I found old Screenflow caches from three years ago that I could delete. But like this looks like it puts it all in one single tool. Everybody was recommending it so I am installing this right now.

[00:43:08]>> Yeah, sometimes when everybody recommends stuff, I can shut it out. But like to me, when I tried this, I was like, "Oh, it does everything I want and it has it all in one spot and it's very nice and easy."

[00:43:16]>> So I've used the CLI tool. I didn't see that they launched this this guey.

[00:43:20]>> Um, and initially I was very hesitant because the CLI tool has full access to your machine to delete cache files and stuff, but >> I ended up throwing a model at it to audit the source code and then I was like, I'll take the risk and then it freed up a bunch of gigabytes on my my hard drive.

[00:43:36]>> Oh, one thing the reason why I bought the UI because you have to buy the UI >> is that it's a $19 one-time purchase. No subscription.

[00:43:45]>> Thank you, Mo. Yes. you release a CLI and you give us a one-time purchase.

[00:43:52]Thank you, Mo. You rule uh the developers behind this. Um another app here I wanted to shout out was Moshi. I had tweeted out yesterday, are there any uh terminal apps for iOS that support touch? Because my god, when you have a system that like herder to manage all of my my t-mucks kind of stuff, everything is based on leading keys and prefixes and uh keyboard shortcuts. And then you pair that with a stupid mobile keyboard and you got to be kidding me. Like I I'm going to be using my mobile keyboard to do controlB or even set up like uh macros to do that for me. I need to be able to click on stuff. And I couldn't believe just how none of them did. But uh Moshi was suggested to me and I installed Moshi and it worked first try.

[00:44:44]This is get Moshi.app. It is a terminal app. It works for Apple and Android.

[00:44:50]It's uh has touch support. It even has built-in Herder support so you can get directly to your Herder tabs if you are using Herder to manage your agents. Um, I don't know why it it has dictation as it a feature. I guess that's fine because like but I I mean the phone has dictation built in div viewer chat viewer. All I know is that this thing allowed me to connect via mosh which is an SSH alternative to my server. It allowed me to use my herder and allowed me to ding prompt from a dang phone and it did so in a much better way. Uh so Moshi getshi.app really enjoyed this.

[00:45:30]>> So what's your stack now? You got Herder mosh.

[00:45:36]What's the M one that we were doing?

[00:45:38]That's a T-m alternative.

[00:45:40]>> That's Herder.

[00:45:41]>> Herder. What about Okay, Herder is the one we did the episode on. Man, like right before I left, we did an episode and I was like, I love this. And then I came back this morning, I was like, what was the app that we were using?

[00:45:55]>> I'm also been using from our good friend and former head of syntax, Ben Vinegar.

[00:46:01]Uh, I'm using Hunk from modem and I'm using that as a diff viewer as another tab in my herder setup. So, I am using Hunk for my diffs.

[00:46:12]>> I just opened up my Herder from before I left. And this is this is why it's so good. I opened up all my projects. I have all my sessions.

[00:46:22]Click on what I want because that was the thing. I opened up my terminal. I was like, what was I working on before I left >> my herder? Everything is still there.

[00:46:31]It's amazing.

[00:46:32]>> I know. You can restart your computer.

[00:46:33]You can do all it. Just your sessions, your tabs. It's It's the best.

[00:46:39]>> Beautiful.

[00:46:40]>> Yes. Yes. Beautiful. But also beautiful is Stack Overflow. Rest in peace. That's beautiful. I It's not beautiful. It's actually kind of sad. CJ, you had a story on here for RIP Stack Overflow.

[00:46:54]Can you talk about this? Yeah, it's it's nothing official, but uh this was actually shared on Hacker News. So, um if you're not familiar, Stack Overflow actually releases all of their data that you can then query. Um if you're a teacher, it's really good for teaching SQL because they give you like tons of tables and you can do joins and stuff like that. And they also have a data explorer. But this query uh shows us how many new questions were created each day. And this graph just shows the decline of Stack Overflow. You can see that in in just just yesterday only 880 questions were asked on Stack Overflow.

[00:47:33]Whereas if you look like at their at their peak back in uh 2017, they were getting as much as 200,000 questions asked per day.

[00:47:43]And we we know why. We we've talked about this, but basically you can see right around 2022. So this is right around when chat GPT was released.

[00:47:52]>> People people just started asking chat GPT instead of posting their questions on Stack Overflow. And you so you see this slow decline. So in 2022 we're at about 115,000 questions per day and then slow decline. In 2024 we're at 40,000 questions per day. By 2025 it's only 15,000 questions per day and now less than a thousand questions per day. So, um, first first F in chat. Throw your Fs in chat for for Stack Overflow. Um, I I actually like I don't like seeing this because, uh, that means there are a lot of questions that people have that aren't getting indexed on the web, right? They're having a one-off session with ChatGpt.

[00:48:35]>> They solve their problem, but then when someone else comes along to solve that same problem, they have to pay token pricing, right? like they don't they don't get to benefit from a forum with all of these questions that have been asked.

[00:48:46]>> Well, that's Stack Overflow is trying to do that. They're they're they're they released like Stack Overflow for agents >> where like agents can post what they had, but it doesn't seem to be catching on. At least not yet. But you're right.

[00:48:59]It's just like, oh, you have to everybody solves the same problems over and over again. And is is that a problem?

[00:49:06]>> Yeah. I mean, to me that that seems like a step backwards. Um, >> yeah, >> because it's it's more compute, it's more powered, it's more resource, it's more potential time spent because an agent is spinning to try and solve that same problem.

[00:49:17]>> Yeah.

[00:49:17]>> So, what you mention, I hadn't heard of what you mentioned, but that could be cool where where agents are putting their their questions and answers on a on a forum.

[00:49:26]>> Um, >> what if we just generated all of that every single time we needed it and boiled the oceans because of it? like h >> you know what makes me think is weird about this is that so often I would post something to Stack Overflow and I would get >> why are you doing it this way? Um and and sometimes that's the that's frustrating but sometimes that's the best thing if you're debugging something with like a co-orker or or posting it on Stack Overflow. often the answer to your problem will be like you're you're coming apart approach you're approaching this wrong and I don't know that like the quick fix from a lot of these AI agents is often let's take a step back and and redo this often the fix is like all right well here here's how you can fix it in the like the hole that you're in you know we can patch it here we can add an if statement when this thing pops in and that'll cover that use case and you start to make a bit of a mess so that kind of makes me bummed because like people are not going through that like hm what's the best way to approach it and and coincidentally that's one of the highest skills you can have right now is like not just writing the sheer code but how do you tackle these problems >> have you ever considered that chat GPT is correct and Wes you are absolutely correct in your ideal implementation here and you made no mistakes with this and I of course was wrong for my uh my suggestion >> yes and I think you can get those things out of an agent but I I don't think that those are those are automatic. I'm just looking at it. Agents.stackoverflow.com.

[00:51:01]I don't know. Seems couple hundred hundred. Here we go.

[00:51:05]Agent activities 1,200 posts created. That's in all time.

[00:51:11]And this launched a couple weeks ago. So yeah, nothing >> nothing really.

[00:51:17]So, I wonder if this is is it like an MCP server where your agent will post after it solves an issue or something like that or the agent's actually actually asking questions.

[00:51:26]>> Yeah, it's a skill. Yeah, it's a skill.

[00:51:28]So, they just post things. Yeah, it's weird that they ask questions like why would you >> Oh, there's a today I learned from an agent. Today I learned VLM's KV cache DT type auto silently promotes compressed tensors.

[00:51:42]>> Yes.

[00:51:44]>> Fascinating. That's some of the best learnings is like, okay, well, you figured it out. It took you three hours.

[00:51:50]Now summarize it and post it online.

[00:51:53]>> I think if people could get in the habit of doing it, this could be good. But this is also another avenue for just AI slop.

[00:51:59]>> Just posts that actually don't work >> and then they're trying to game the system to get more points. Yeah, >> one one thing I'll mention is so Stack Overflow is actually part of a bigger community of question answer websites >> and um I'd be curious to see the data for these other ones as well. So there's like ask Ubuntu mathematics um there's a specific software engineering one >> plumbing ones. Yeah, there's plumbing, there's like pet owners, movies and TV.

[00:52:25]And I I would be curious if they're seeing a drop in questions asked because like that's the other aspect of what was great about the internet before AI is is people talking and people sharing and giving real people opinions. Very similar to like what we try to use Reddit for now, but that's slowly going away because it's overridden by bots.

[00:52:44]Um, so yeah, rest in peace.

[00:52:46]>> Let's let's check it. Okay, can we run the same stat?

[00:52:50]>> I think so. Let me see. Um I think I can just copy the URL because this is >> change >> instead of stack overflow for instance.

[00:52:58]Let's try just server fault.

[00:52:59]>> Yeah, if you change the stack overflow in the URL to math, you are also seeing the exact same drop.

[00:53:11]>> See, I'm not a robot.

[00:53:15]>> Blender WordPress. What's something that's not destroyed by AI?

[00:53:22]>> Oh, that's an impossible question.

[00:53:25]>> Well, so this is for law.

[00:53:31]>> So, yeah, for server fault also, we see a drop. I'm curious like um yeah, what would be a good one to look at? How about English language and usage? Because that has nothing to do with computing.

[00:53:46]It's just on the English language.

[00:53:51]>> I mean, I I'm not I wouldn't be surprised to see that have the same exact drop. I mean, I I'm all my English questions I'd be asking to AI or or Google.

[00:54:00]>> It's the same drop as well. They I mean, at their peak, they were getting 1,600 questions a day. Now, they're getting 50 questions a day.

[00:54:06]>> Yeah.

[00:54:06]>> Yeah. Yeah. I I I've asked a plumbing question in the past on whatever the stack exchange plumbing one is. Um, and then the last time I had a plumbing question, I just went on Reddit.

[00:54:17]>> Were they just like, "Are you sure you want to do that?"

[00:54:20]>> Well, I knew somebody immediately would, which I I had an answer within like three minutes. You go to my Reddit and look at it, but >> I think Reddit is a better avenue for this stuff.

[00:54:30]>> Reddit. Yeah, for sure.

[00:54:31]>> Or like Facebook groups. I know you guys hate Facebook groups, but there are some cracked people in these very niche Facebook groups.

[00:54:39]>> Some cracked 90year-olds answering your questions. Yeah.

[00:54:41]>> No. No. There's there's a lot of like dudes wearing white sunglasses.

[00:54:45]>> Okay.

[00:54:45]>> Like making doing the laugh emoji to everything. But there are some smart people in there.

[00:54:51]>> Nice. Okay. So, that's all I had.

[00:54:54]Funeral for Stack Overflow.

[00:54:56]>> Uh I'm going to pass >> Rest in peace.

[00:54:58]>> Yeah. RIP. RIP. Uh Stack Overflow. Yeah.

[00:55:01]Pass it off to Scott. Uh we have >> I don't know if you about Yeah. Have you talked about your foot pedal on the channel yet?

[00:55:07]>> We talked about it in the Dax episode. I got a foot pedal. Uh, when you Google computer foot pedals, man, they're all from like the 1980s. They're crazy bad.

[00:55:17]But Elgato of all people actually made a really good one, which the reason why it's good is because one, it actually doesn't look like complete ass. Two, it has three pedals instead of just one.

[00:55:30]And the three pedals is really great.

[00:55:32]So, what the heck am I using a foot pedal for? Well, I use the center one to push and hold for doing dictation. Uh, this one does enter and this one monitors my herder. Randy, if you could can you throw me and my herder up on screen.

[00:55:49]Okay, let's see. So, ideally, what I would do is I would push this one like this and that sends me to the next tab.

[00:55:58]And then I would push this one to speak and then I would push this one to hit enter. And I would just do that. And that's how I that's how I operate my uh that's how I operate with my foot pedal.

[00:56:13]Um but yeah, this thing's dope. I uh I like it quite a bit. I was I was considering making >> dogs out and like didn't even flinch.

[00:56:22]>> Yeah, for those listening to this uh bare feet live on stream. So >> Oh yeah, for free.

[00:56:28]>> For free. Yeah, you didn't even charge >> for free. Sorry folks. Yeah. Well, I had to show the dexterity needed to require uh for this this foot pedal. I can use it with shoes, too, but I prefer not to.

[00:56:38]I got a nice little foot feel on there.

[00:56:41]>> Disgusting.

[00:56:43]>> That's all I got for the foot pedal, folks. That's how I use it. The software is all Stream Deck software, and you can do macros or whatever you want with it.

[00:56:51]So, >> sick.

[00:56:53]>> I couldn't have just used my hands, Max.

[00:56:55]I couldn't have. I needed to show how it really worked. Uh CJ, you got a bunch of other stuff. Do you want to get into it?

[00:57:01]>> Yeah, I got a couple more links to to wrap us up here. Um, the first one is called Decoy Font.

[00:57:09]So, this is a font that hides what you're typing from AI. And, uh, the idea is when uh, this won't work with just like plain text on a website. Basically, if you give an image to AI, any of the main models for anthropic or OpenAI or Google or whoever else, if you give them an image that has text in it, it will basically be able to read the image because it's been trained on image. It's not exactly OCR, but it basically can pull out the text and understand the text that's in the image that you gave it. But if you use this font, they showed in their tests that even the most frontier models can't actually see the hidden message. And basically what it does is it puts your your real hidden message in a kind of like blurry behind the uh text on top that the the robot the AI will actually see. And if you zoom out, you can see the actual text that you you had written. So >> for instance, if if you're publishing PDFs online and you don't want AI training on them, um or images and you don't want AI training on them, you could use this font instead. And when AI pulls this image in, it sees just the text, sorry, robot. Whereas if you're reading it just at a normal distance, you see the text happy human. Um, so pretty cool. It's it's a an interesting idea, but the whole the whole thing is basically from at a distance you can see it blurry, whereas the the model the AI models only see the higher level text. and they showed a bunch of example um threads where they're like asking the AI, "Are you sure you don't see the the that other message?" And it it couldn't figure it out. So, >> that's funny that somebody figured that out.

[00:59:00]>> Um >> and I'm just curious how easy it would be to figure help the AI write a script to reverse it. Well, I think if it Yeah, if the AI knows it's a thing, it could literally just down sample the image, like zoom it out.

[00:59:16]>> Oh, yeah.

[00:59:17]>> Able to read it.

[00:59:18]>> Um, but like in this example, you can see the embedded image says this sentence is written in decoy font and then when the AI tries to read it, it's just gibberish, just a bunch of like random letters.

[00:59:29]>> So, >> yeah.

[00:59:31]>> Is this just like a is this just a funny thing or like a somebody has like >> I think it's probably a funny thing. I mean, it's funny, but at the same time, if you're publishing Yeah. I mean, if you're publishing content on the web and you don't want AI trained on it, you could use this font. Like, but it specifically has to be rendered documents. Like, it can't be text on the website. It has to be a PDF or an image.

[00:59:50]So, >> so you just give accessibility out the window >> pretty much.

[00:59:55]>> So, what you're saying, CJ, is that you hate accessibility and you think it's wrong.

[01:00:00]>> Do you hate blind people, CJ?

[01:00:03]>> I knew what I'm saying. I knew it.

[01:00:06]>> Moving right along. I got another story for you.

[01:00:08]>> Let's cancel CJ.

[01:00:10]>> Uh, this is this one's really fun. Um, so it's from Fabian Sangard and they went deep on Jurassic Park. So, Wes, have you seen Jurassic Park?

[01:00:20]>> I have seen bits of it.

[01:00:23]>> I remember at a pool party, maybe like five or 10 minutes of Jurassic Park. I understand. It's got the dinosaurs and the stupid jeep >> and things like that, >> but it also has computers, right?

[01:00:34]There's Yeah, there's lots and lots of computers in Jurassic Park, servers, server racks, uh CPUs, and this article goes into the fact that everything they showed in the movie was a real computer.

[01:00:50]Like the they kind of go into the fact that there there would they knew that nerds would be watching this movie and so they didn't want to show computers that were fake or wouldn't be what somebody in those positions would really be using. So they shot the screens. I I would So do those screens not have like refresh rates that couldn't be filmed because I know that was such a major concern for filming TV for so long.

[01:01:13]>> I believe you can kind of tune your um on the camera itself. It's like the shutter rate. If you get it exact, I don't know if they did that, but if you get the shutter rate exact, then it won't show the lines.

[01:01:25]>> That's exactly what I had today. My I went into I'm at my cottage right now and the LEDs were strobing at a different rate than my office at home and my camera was just was was rolling.

[01:01:38]>> Yeah.

[01:01:39]>> So I had to change the shutter.

[01:01:41]>> You had to change it on your camera.

[01:01:42]>> Yeah. Pushed a bunch of buttons and it started working. I I need to pee for my camera, man. I >> stupid little Sony things. I don't know what any of these things do.

[01:01:54]>> The Sony menus are terrible in those cameras. And to make matters worse, Wes, the I have two Sony cameras and they're both the menus are so different and and one one non-menu based thing that drives me nuts about them. They both have two SD card slots on them. On one of the cameras, the uh number one slot is on the top and two is on the bottom. On the other one, it's reversed as well as the cards flip front to back differently.

[01:02:19]So, every time I open the flap on my camera, I'm like, which one is this got to go in and in which direction? I got to like try. It's like worse than the USB problem. Um, >> and you you know they'll never fix it because like the professionals know those menus cold and it's just such an awful interface for everybody else.

[01:02:35]>> There's people that make custom uh firmware, but ain't no way in hell I'm doing that on my expensive camera. Uh, I will say though, Wes, MCPs for me have like really leveled up in the past 48 hours or so, maybe longer than that. I I've been using uh Figma MCP to So, one thing that I don't want to do is I don't want to do AI thumbnails on this channel, right? We don't want to have AI creating the thumbnails because AI makes all design look like that. But what I wanted was I have a design direction and I wanted AI to be able to uh shift these components around. And so I had the Figma building blocks in place and I told the Figma MCP to give me several different directions using these assets.

[01:03:23]And it did a pretty freaking good job at least uh moving things around to the point where I could then and there's no generation happening. It's essentially just shifting things around and changing copy for me. Um just cuz I was stuck creatively and that got me out of a little bit of a creative rut and then I finished it myself. Uh, two, Da Vinci MCP. Uh, the Da Vinci Resolve MCP is so good. It's so so good. I wrote a little Python I wrote AI wrote a little Python script that could tell who was talking, me or Courtney, based on the frequency of our voices and then did cutting based on that. And I was just like, this rips.

[01:04:01]I love this. And if you watch my Reachi mini video on this channel, every single graphic I made in that video was done with browser tech using Dainci MCP to uh generate the well it would first generate the motion graphics that I have in an app that I'm building. turned that into a video using Da Vinci MCP, insert it in the timeline with the markers at the exact timings I wanted to and could go back and forth that way for all computer graphics.

[01:04:31]>> Man, MCP rules. Uh, big fan. Yes, >> the Da Vinci one in my learning. I learned Da Vinci over the last what, like two months or so, and just simply understanding what the heck is going on and being able to type into a box and ask it these questions. So, good.

[01:04:50]Yes, questions in box. Good. All right, CJ, what else you got for us?

[01:04:54]>> Got one last thing for you, and this is just a fun one.

[01:04:58]>> So, this is a 3D Voxil environment where you can ride some of the trains in Tokyo live.

[01:05:06]>> So, they're pulling in >> Oh, I have I have to mute it because it's talking to me.

[01:05:12]>> Denja. Um, so you choose a line and this is real time because the the Tokyo um >> scrolls. Okay.

[01:05:21]>> The Tokyo Transit has real-time data about like where that train is, how far away it is from the next stop, when it's stopping, how long it's stopping, and so they created this 3D voxal environment that looks like >> uh Tokyo, and then you get to just choose a train and ride along with it.

[01:05:38]So, right now, um, we're on the outer loop for, uh, Osaki, but then you can switch trains, um, and ride that train.

[01:05:49]>> Awesome. I'm going to Tokyo in March, and I'm going to be riding the hell out of these trains. I've ridden the the Tokyo uh, train quite a bit, but um, >> this is sick. Where does this data come from? Is this like Google Maps or Apple Maps? the actual Tokyo Transit provides the real-time data.

[01:06:12]>> Man, that is >> I'm going to need some Tokyo Rex. I already do have quite the itinerary planned out for the 10 days we're going to be there. We're also going to uh head up head over to Kyoto for a couple days.

[01:06:23]But, uh I'm very very excited about planning some stuff out. So, if there's anything unique that we have to be doing in Tokyo, let me know because I'm going to be there.

[01:06:33]>> Yeah, I'll DM you some Rex. It was a blur. I was only in Tokyo for 4 days last year, so that was a bit of a but >> Oh, thanks, Wes. Yes, I know. I'm just dying cuz last time I had good Aonomiyaki was in Tokyo and I've never found a place here that does it well.

[01:06:48]So, I'm stoked. And uh I get to use my language skills, which I am also stoked for. I've been doing some practicing.

[01:06:58]>> Nice. That's all I got.

[01:06:59]>> Just silence. Just silence. Nobody says a gosh darn.

[01:07:03]story. Say, just do give me that. Plate me there.

[01:07:07]>> You've been practicing your Japanese.

[01:07:09]Wow.

[01:07:10]>> Yes.

[01:07:10]>> You going to have a sword?

[01:07:13]>> Yes. Can I have a sword? Uh, no. Wes, what are you talking about? Oh my god.

[01:07:21]>> Oh my gosh.

[01:07:21]>> Oh my gosh.

[01:07:22]>> CJ's even a bigger nerd. Okay, so let's stop being mean to Scott and move that over to CJ here.

[01:07:28]>> Yes.

[01:07:29]>> All right. What else have we got? Is that it?

[01:07:31]>> That's all I got. You guys got anything?

[01:07:33]>> That's it.

[01:07:34]>> Sweet.

[01:07:35]>> I don't have anything else. I'm going to I'm going to leave now. Okay, guys.

[01:07:40]>> Slow exit. Yeah.

[01:07:47]All right. This is Where is he going? He's gonna do more feet. Let's see the feet.

[01:07:54]>> Doing more feet. I'm not doing more feet.

[01:07:56]>> Yeah. So, that's all we That's that's all we got for you today. If you want to see Scott's feet, go watch the VOD. Um, and uh, we do this every week. So, if you're just now tuning in, be sure to watch the VOD. But every Monday, same time, come hang out with us. And we're going to try to do a raid on Twitch. I don't know if we can we'll be able to find anybody, but that's basically where we take all the viewers that are currently in our stream and send them over to somebody else's stream. So, stick around on Twitch. Twitch.t Twitch.tvaxfm if you want to join in on that. But, this has been awesome. Thanks for joining me, guys, and we'll see you all next week.

[01:08:28]>> Peace. S.