1020: Do You Read The Code?

Auto-accept on, brain off? We’re getting into whether anyone actually reads the code they ship anymore — then it’s HTTP’s first new method in 16 years (meet QUERY) and OpenAI’s compute-hungry “Sol Ultra” tier landing in Codex. Bring your hot takes and hang out with us live. Show Notes 00:00 Intro 00:31 Welcome to Syntax! 01:44 World Cup Chat 03:21 Do you read the code? Theo’s take Primeagen’s take Dax’s take notch’s take David Fowler’s take David Cramer’s take 16:50 Query: NEW HTTP Method RFC 10008: The HTTP QUERY Method The new HTTP QUERY method explained 22:39 Github Repos on a CD-ROM Github’s post 29:59 Brought to you by Sentry.io 30:35 Gea - compiler-first UI framework Gea React vs Gea 41:23 Fable First Impressions Anthropic announces Fable comeback 49:32 Codex Gets Sol Ultra 52:59 Update: Scott’s Robot 56:36 Junior Slack Agent Junior 01:03:55 Supply Chain Security Updates Min release age now in all tools npm now adds a temporary, preventive safeguard for high-impact accounts 01:07:40 HTML element Chrome introduces element 01:10:16 Claude Code Auto Accepted Answers after 60 seconds "No response after 60s — continued without an answer” 01:13:21 Claude Code Fingerprints Requests to China Claude Code Is Steganographically Marking Requests 01:16:24 Responding to Chat 01:21:06 Organic Maps Organic Maps for all platforms 01:24:21 Starring the Computer Computers in movies and tv 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

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  This is another argument I have about React is we've kind of stagnated. There's not a lot of innovation. AI is just spitting out the same old React. And so in a sense, that kind of feels like we're stuck. Like even if something like this comes along and it does it better, it doesn't matter because it doesn't have the critical mass and AI is just going to poop out like bad React code with a bunch of use states and use effects in it anyways. But something better might come along.

• 0:31Scott Tolinski?

  What's up, everybody? It's Syntax Live, another wonderful Monday as we get into some stuff here. We're going to be talking about reading code. Is that a thing you should be doing? What are people talking about? Because it seems like there has been a lot of conversation on if you read the code and a ton of really interesting people have chimed in on this. I have a feeling like I'll know your answer a bit, CJ.

• 1:01Speaker 00

  Yes, I have thoughts.

• 1:03Scott Tolinski?

  Yes. What's up, everyone? It's kind of shocking that we are just slightly late to this, given that this is the first episode without Wes, who is kind of a... Is Wes ever late? I guess Wes isn't late. Maybe it's me.

• 1:18Speaker 00

  He's an on time guy, yeah.

• 1:20Scott Tolinski?

  Yeah, maybe it's my fault. Sorry, folks. I'll take, I'll take responsibility for that. But what's up? Welcome to Syntax Live. Every single week, we get into the topics that are going on the web. We're getting into new libraries. We're getting into new projects and we're getting into what people are talking about. My name is Scott Tolinski and with me today is CJ Reynolds. What's up, CJ?

• 1:41Speaker 00

  Hello. Not much. So I'm not a soccer watcher, football watcher, but I did watch the Mexico versus England game. Are you watching the World Cup at all? Are you watching?

• 1:53Scott Tolinski?

  I, you know what? I'm not a big soccer guy either. But I did watch that game and that was an exciting ass game. That was crazy.

• 2:01Speaker 00

  It was spectacle, like red cards, penalty kicks. Like, yeah, it was, I was sad Mexico didn't win, but it was an incredible game. Yeah.

• 2:10Scott Tolinski?

  I'm the mind that soccer would be better if, one, the players obviously didn't flop around on the ground. That's table stakes for me. Like, as a hockey watcher.

• 2:20Speaker 00

  Yeah.

• 2:21Scott Tolinski?

  As a hockey watcher, it's like, come on, guys. You know, the hockey players will take a puck to the face, lose a bunch of teeth and keep playing. Like, you can get up. It actually really bugs me on defense when a defenseman like goes down and lays down. Like, brother, you need to be standing up and helping your team. What are you doing?

• 2:39Speaker 00

  Exactly. But I also like the opposite, though. So, near the end of the game, someone got really hurt, like rolled his ankle. They just dragged him off the field and kept playing because there was like two minutes left. So, I thought that was hilarious, that they don't pause the game. They just like, we gotta get this corner kick.

• 2:53Scott Tolinski?

  Yeah.

• 2:54Speaker 00

  Yeah. But it would be fun if there was like actual, we need a penalty box in soccer. We need like actual fights. Put them in the box for 10 minutes. Let them cool off.

• 3:02Scott Tolinski?

  Let them cool off. Yeah. I know. I'll make football people mad about my takes on soccer. But another one for me is no offsides. Get rid of offside. Let them cherry pick. Let them just kick the ball down, run and score a goal. Let's get some goals going. That's my caveman approach to soccer right there. That's how I fix it.

• 3:21Speaker 00

  Nice. I'm excited to get into it. I think, so I personally, I've been staying off of X, but Scott did bring this to my attention and I've been reading some of the replies. So, Scott, do you want to get into this first story? Do you still read the code?

• 3:35Scott Tolinski?

  Yes. CJ, I gotta say, how could you stay off of X, the everything app? The app that has everything. The everything app. Yeah.

• 3:43Speaker 00

  I check it once a week before we go live. It's good for my mental health, but yeah.

• 3:47Scott Tolinski?

  Yeah, I believe that. Yeah. Man, I gotta say, one thing that X has been doing a terrible job lately is serving me a bunch of crazy reaction videos. Like, I don't care. Like, this is not TikTok. I don't want to see somebody else watching a video. Either way, the tweet is, "How much better do the models have to get before you stop reading the code?" And that led to responses from all kinds of people. So, even just this first one here, man, it's crazy the amount of people that chimed in on this. So, like David, Kate Piano. Justin Schroeder says, "Negative 20%. Reading the code is meaningless now. It always was. You have to verify it." Interesting.

• 4:28Scott Tolinski?

  Cheng Lao, "Reading code, I barely even look at the folder structure now." That's Mr. PreText himself. Reason ML. Glaber from Terso, "For database or OS code? I will expect that never. A lot of us still routinely read assembly code." So Glaber is big on reading the code, which, I mean, Terso, great product. So.

• 4:53Speaker 00

  I would think, like with Glaber, so he maintains a framework. He maintains a library that people use. And I actually watched a video, it was an interview with Rich Harris, the creator of Svelte. And he talked about, like, if you're creating a library or a framework, you kind of still need to be reading the code because you're on a bit of a frontier, in a sense. Like, the thing you're making doesn't exist. So I would expect that, and I would hope that people creating the libraries that we use are still reviewing it with a human in some way.

• 5:20Scott Tolinski?

  Yeah. Otherwise, you're going to get OpenClaw, which, OpenClaw, for all of its pop- you could argue that because of its popularity, that doesn't matter. But every single time you update OpenClaw, it breaks. Where, like, Hermes, they do read the code, and it is more curated, and Hermes never breaks. So, yeah, that's just anecdotal right there. Prime Engine says, "I think this 'read code slash don't read code' really comes down to one thing. Do devs need something to fight about? This is our new React." And Prime nails this here.

• 5:51Speaker 00

  Yes, absolutely.

• 5:52Scott Tolinski?

  Because before this, is HTML a programming language?

• 5:58Scott Tolinski?

  There's just an endless amount of bike shedding on this kind of stuff that happens. So, yes, devs are just going to fight about anything, reading the code or not. That's not what matters.

• 6:09Speaker 00

  And I think the answer, just like with everything else, has nuance and lies somewhere in the middle. Because, I mean, personally, if you've seen my takes on this channel, you would expect that, yes, I read the code. But I will say, for some personal projects, for some one-off scripts-

• 6:25Speaker 00

  I don't. I do actually turn on auto-accept edits. And that's because a lot of times, those aren't projects that are going to have a long life. It's something that I'm just testing out. I want to get it up and going. So I'm more than happy to just let the AI write the code because I'm there to test the functionality or just like test the thing itself. But when it comes to production code, when it comes to like code that you're shipping for your company or that you're contributing to, you probably should read the code just as an engineer. If you want to pull up the tweet from Kramer, he responded that says, "No good engineer will tell you it's useless to read the code, by the way. That's the easiest signal." I think like it's part of our jobs. Like we have to care. I think like you're too lost in the sauce if you just think that just prompt, prompt, prompt, prompt, prompt.

• 7:09Speaker 00

  Because that also gets into like you're not necessarily caring about token usage or bill or efficiency or moving too fast. And these are all things that we actually need to care about in real production code.

• 7:20Scott Tolinski?

  Yeah. Kramer is always dead on in my experience. And it's so funny, the people, the people that respond to him and like he like doesn't know what he's talking about. Like, bro, Kramer is like, one, he's a cracked engineer, but he's also built something insanely large. So it's like, yeah, he runs a, I mean, he's built this business into a billion-dollar company. Like, that's crazy to me. So.

• 7:52Speaker 00

  Yeah. If you don't know, like Kramer is one of the co-founders, former CTO, former COO of Sentry, who we work for as well.

• 8:02Scott Tolinski?

  Yes. I'm not just big-upping my boss either. I'll just say that.

• 8:06Speaker 00

  No. And you can see it from the proof. Like, I mean, we have a little more insight into this because we can see all his messages in Slack and like we hear him from all hands. But like, this guy, he still is, even though he's not an engineer or CTO anymore.

• 8:22Speaker 00

  He's still writing code, he's still using agents, he's still working on workflows, he knows what matters, so.

• 8:28Scott Tolinski?

  Man, he's built a bunch of stuff that I'm using right now, whether that's the Dex thing.

• 8:33Speaker 00

  I use Dex every day, yeah.

• 8:34Scott Tolinski?

  Yeah, for my issue tracking and stuff. We also had David Fowler says, "I see a lot of influencers talking about reading the code and building software factories, but I don't see much discourse about what software engineering teams of the future will look like. Software is a multiplayer game as much as everyone will try to convince you that all you need is a team of agents." Yes, I so much agree. David nailed this, and I think there are some people who are working on this problem. Back when we were at JS Nation, there was a panel that I was a part of, and afterwards a discussion room, and a big portion of the

• 9:14Scott Tolinski?

  room really lined up on the idea of pair prompting and what that looks like in working on agents together. And I think there's some really neat tools coming in this space that I'm not quite sure what I can share yet, but that's a little teaser for maybe future episodes when I can share some stuff. But I do think that there is some really interesting work being done in that space. What does teamwork look like in this world of AI agents?

• 9:44Speaker 00

  I think that's the biggest question, because a lot of devs have, even on one individual, one big team, every dev is going to have their own workflow, and they haven't necessarily figured out how to collaborate on that, like how to share skill files or workflows. And personally, even though I'm on the syntax team, I'm still kind of a solo dev in the sense that I'm working on a lot of side projects, and all the workflows that I've created are just for me and my ability to write code and review it. But that, I haven't even thought about how that works on a team. And I think that's going to be the answer moving forward. I mean, I feel like it is. I feel like there will be some people that say, "Oh, we don't need teams going forward." And I did see some tweets

• 10:26Speaker 00

  from, maybe they were rage baiting, but I saw tweets from people that were like, "We don't need managers anymore, just hire a few devs and that's all you need, like one or two devs and that's all you need." But, I mean, like, real software is built by multiple engineers collaborating, and like, that's...

• 10:43Speaker 00

  We're still not there in terms of how we work with agents. So I think it's something to think about as well.

• 10:48Scott Tolinski?

  Yeah. Folks, if you're just joining us, we're talking about, does reading the code even matter? Twitter has been a buzz. All a lot of like really interesting people have chimed in on this. This actually was one of my favorite tweets because CJ says, some guy named Notch says, how much better do the poll requests have to get before you stop reading the code? And I said, do you really not know who Notch is? And he was like, no.

• 11:12Speaker 00

  Yeah, I didn't.

• 11:14Scott Tolinski?

  People don't know Notch created Minecraft also just Minecraft, something that has been going on every single day in my household for the past billion years. So not just for me, my son is absolutely obsessed with Minecraft, like many other third graders. So yeah, before you stop reading the poll requests, yes, poll requests have to get better. That's funny.

• 11:37Speaker 00

  Yeah. There's one more tweet from Dex I want to highlight, and he always has the most level headed takes. He says, guys, stop trying to argue with the code doesn't matter posts and just ask them if they're so ahead of the game, why aren't they richer? Like he has multiple posts like this, like if, if, if really we don't have to read the code anymore, code anymore, like show me the receipts. Like, have you built a million dollar business with this? Are you like actually making money from it? Are you actually more productive? Cause it does seem like, especially in the Twitter space, there's a lot of people just spinning wheels and burning tokens with, with not much to show for it. So.

• 12:10Scott Tolinski?

  I feel like I, I'm crazy sometimes when I see the amount of people who like, what are you actually making? And is anybody using it? No, you know, so I, I mean, there's so much of that. Somebody, Patrick says, I don't really take this, this famous software engineer says X, so it must be correct argument seriously. Successful people are wrong all the time. Yes, Patrick, you, no, Patrick, you don't have to apologize. You wrote this just fine. I can't read. So, no, I think this is, I think it's dead on Patrick.

• 12:45Speaker 00

  Yeah.

• 12:46Scott Tolinski?

  Another one.

• 12:47Speaker 00

  No, sorry. That's where the nuance comes from. It's just multiple takes, right? Because not every engineer is going to have it perfectly right. That's why we're seeing all sides of it.

• 12:54Scott Tolinski?

  Yeah, yeah, totally. The next tweet was from Zach Corman. Zach's a good follow. And Zach does cybersecurity stuff. I think he's a real good follow. He says, "I find Theo's takes on this exhausting. No one knows your process, your role, or the way your brain works, but you. It reminds me of all the people who tell me that I'd stop writing and reading code as a manager without understanding why I did it." Man, this is such a thing where people become a manager. They feel like, "Oh, now I have a team of agents to go off and do the work," and then they lose sight on everything, where, again, it's the same thing of like, you kind of have a bigger role than ever to be able to control things.

• 13:40Scott Tolinski?

  So Zach is a good follow. Again, cybersecurity stuff, if you are interested in that.

• 13:47Speaker 00

  I think it also plays into one thought I had, like, in reading this initial thread, was the fact that, like, before agents, people already didn't read the code in pull requests. Like, if you're ever working on a team and you made a really big PR, you'd get multiple comments, "LGTM, like, looks good to me." So this is an old problem, right? Just like, when there's a lot of code to review, it often doesn't get reviewed to begin with. And now you couple that with the fact that we're producing code at 10x speed, people are already set in their ways and are already not reviewing. So, I don't know. I think we're here to say, like, take a step back. There was a post from a few months ago, just, it was like, "Slow the F down." Like, we can produce code at 10x, but should we?

• 14:30Scott Tolinski?

  From Mario.

• 14:31Speaker 00

  From Mario, exactly.

• 14:32Scott Tolinski?

  Mario Zetchner, the creator of Pi, says that he slows down and he reads the code. Now, me, to give my answer to this, whether or not I read the code entirely depends on what project it is. Is it a project I care about? If it's a project I care about, I read the code. If it's not... You know, whatever, just slop it together. So I, for me, it's an entirely situational dependent thing.

• 14:56Speaker 00

  I completely agree. And yeah, that's where I'm at as well. It's, and there was a blog, I need to, I'll link it in the comments if I can find it. There was a blog post that I found on Hacker News that, very similar to Mario's post about slowing down. It just talked about like, if you actually want to be successful with agents, you need to turn off auto accept and you need to review the changes. Because we're also getting into the realm of like, when we were writing code by hand, we're building up this mental model of our code base, right? As we code, we, we learn more about where this code lives, how it relates to the whole system. We're kind of like building up this mental model as we code it. But when you let AI do everything, you're not building up that mental model. So in a sense, that is moving into the realm of like, what is my job anymore? Like, why, why am I even here?

• 15:41Speaker 00

  So I think you should be slowing down. I think you should turn off, especially on features that you care about, features that need to be engineered well, need to have good code. You should turn off auto accept. You should just slow down, take your time.

• 15:54Scott Tolinski?

  There are some things I'm going to say that I don't read. If it's something that I don't understand. For instance, I had an agent port a shader for me to another language. And I don't even know, uh, I don't know shader code. I mean, I, I've like tried to learn 800 times to write shaders myself and I'm not good at it. So when it worked and it was done, I said, fine, that's fine. That's AOK with me. So I think the most exhausting answer and the correct answer to this is it depends. If it's something you care about, it's absolutely worth still reading the code. And if the question that, well, if it works, does it matter? Well, just because something works doesn't mean it's good. Lots of shit works, that's no good. So works is not the bar and works has never been the bar.

• 16:40Scott Tolinski?

  But again, devs are going to be arguing about stuff when they should be talking about what their users need and want.

• 16:46Speaker 00

  Definitely.

• 16:47Scott Tolinski?

  Yes. Okay, I think we can get off this one. CJ, let's keep it moving here. Syntax Live Weekly, every single week. Wes is at the cottage, probably just hanging out on the dock. Actually, I think he might be in New York. He's doing something with his family. So he's not here today. He's on vacation. As CJ and myself, we're talking all things new on the web here. CJ, what else you got for us today?

• 17:08Speaker 00

  I have for you something very exciting. It is a brand-new HTTP method. So if you're not familiar, the HTTP methods are like GET, POST, PUT, DELETE. Give me some other ones, Scott.

• 17:21Scott Tolinski?

  I only use GET and POST. Those are the only two, because the other ones just kind of are there, and I never even consider them. I don't know why.

• 17:32Speaker 00

  Well, yeah, so they exist. And the thing is, they've existed for a very long time. So the idea that there would be a new HTTP method is kind of interesting. But this is RFC 10008. It's currently a proposed standard. It hasn't made it into any web browsers yet. But Node.js has actually implemented this in their server-side HTTP module. So if you're implementing a server, you could actually listen for it. But the idea is, so it's called query. So now in the long list of methods, you've got GET, PUT, POST, PATCH, DELETE, and now query would be a new one. But the idea is, sometimes you might have a request that you're cramming data into the query string

• 18:16Speaker 00

  because you want it to be a GET request. So in the example here in the introduction, the common query pattern is like slash FEED, which is some endpoint, and then question marks, you have the query string, and then a bunch of query parameters, like Q equals foo, limit equals 10, sort equals minus published. And this is essentially how you craft a request that has some data inside of it. Basically, the data is in the URL. And then the server can act on that data. But this is and could be a, I think the right term is item potent? Item potent request? Basically, it's a request that doesn't modify anything on the server side. It's just a request for data, the server returns something. And we do this a lot. And the reason we do this is because get, the get method, does not officially support putting

• 19:01Speaker 00

  data inside of the body of the request. So typically, if you do have some more complex data that you want to send, but you still have an item potent request, you would use a post request. And we actually see this in like GraphQL. If you've ever built a GraphQL API, you do like slash, I think it's slash query. It's always a post request and then it has a big massive payload of all the things you're requesting or reading. But that's the reason they use a post request is because they need more than just the URL state, they need actual data in the body. And so HTTP query is here to solve that. Essentially, it's an HTTP get, but it can have a body. So you can put those query params in the body as like form URL encoded params, or

• 19:45Speaker 00

  you could have JSON inside of the body. You could set the content type to whatever you want. But the idea is this will still be classified as an item potent request. Like it doesn't modify anything on the server, it's really just requesting data. I have this other blog post that shows some more concrete examples. So yeah, in practice, essentially like GraphQL would be one of the things that could adopt this. So instead of a post to slash API slash GraphQL, they do a query and then that query includes the GraphQL query in the body. Similarly, like they show an example here where instead of using a form URL encoded body, they actually use a JSON body.

• 20:25Speaker 00

  So you have a query request to an endpoint, include JSON, and it's essentially just a JSONed version of the query string, but it can get more readable because you can have like array types inside of there. Typically, like with query strings, if you want an array type, you'll have like multiple of the same query parameter. So query, it's coming, it's exciting. And like I said, it's only supported in Node.js HTTP module. So far, but the idea with an RSV RFC is it's a request for comment. So implementers of browsers and everyone else can kind of chime in to decide, should this thing exist? Should we kind of change the spec a little bit? And eventually, this will make it across the web. I don't know how long it's going to take, but eventually,

• 21:10Speaker 00

  if everyone buys in, it will make it across the web and you'll be able to use query instead of post or get.

• 21:16Scott Tolinski?

  Interesting.

• 21:17Speaker 00

  Yeah.

• 21:17Scott Tolinski?

  Yeah. I mean, this is the type of thing where I think, and you know, what's this going to do for us? Devs on a daily basis, but I think it's neat and it feels modern. It feels nice.

• 21:31Speaker 00

  Yeah.

• 21:31Scott Tolinski?

  I don't really have too much other than it's interesting, but yeah.

• 21:35Speaker 00

  I have some comments from Hacker News I can show.

• 21:38Scott Tolinski?

  Oh, good. Good. That's what we need.

• 21:42Speaker 00

  The main take that I think a lot of people have is query is just get using get with a bodywork. So this is one take that's like, you may not know this, but you actually can put a body inside of a get request. The issue is not every HTTP server can handle that. Not every client can do that. And so that's the whole idea is like, you're not supposed to use get with a body. It's a hack. So having an explicit method makes sense because it, yeah. And their commentary is just because it works doesn't mean it's the right way. But I think that's the general idea is most people will be like, why do I even care? Why couldn't I, why don't we just use post request? But I think having something explicit for something we do all the time is the reason why stuff like this exists. So.

• 22:25Scott Tolinski?

  Yeah. And honestly, being additive, you know, it's not some game breaking. It's not going to destroy anything. It's just additive. You can use it, but yeah.

• 22:37Speaker 00

  Exactly. Yeah. So Scott, what do you got for us next?

• 22:41Scott Tolinski?

  Yeah, I got something very fun. In terms of very fun, where the internet decided it was going to be giant babies for one day, GitHub, they decided to have a fun little promotion just because the Sony PlayStation announced that they are going. Discless, I had to make sure I focused really hard when I was saying that word just now. Discless. And that they will no longer be shipping discs in packages. And so GitHub said, "Oh, we heard you and we agree in light of recent developments in physical media, GitHub is proud to announce that you can obtain your public repo on a CD-ROM." And they even had a little link here where you can go ahead and get one.

• 23:24Scott Tolinski?

  And the internet had a very normal response to this, where everyone just started complaining about it. "What in the world? Why would we ever want that? What? Why don't you fix your infrastructure?" As if the marketing team at GitHub, who's probably just a couple of people putting this promotion on, are the ones going to be single-handedly fixing your infrastructure. Like, "Oh, let's, let's re-divert that energy and money." Like, this is costing any kind of money that would then cause GitHub to somehow not be able to fix their infrastructure. It is crazy. And I had people arguing with me left and right about this thing.

• 24:08Scott Tolinski?

  I said, "This is just a fun promotion." They're like, "You gotta think of the optics." It's like, brother, like, people are just in such a dang grumpy complaint about anything that like, I cannot comprehend being mad at this. They're like, "You gotta think of the timing and the optics." I, those two takes, like, fine. I can understand the optics of it, that GitHub hasn't had the best uptime lately. But the people who are saying, like, "It's diverting funds from infrastructure," or "This is why GitHub is down all the time," like, you people are so out to lunch. And, and, I don't know, I got somebody grumpy at me. I'm gonna, I'm gonna put him on blast, actually.

• 24:52Scott Tolinski?

  Let me even, let me, let me do that. Uh, let me go through my, oh my gosh. Uh, yeah, let's find this. Let's find this, because this makes for good content. Yes, this guy here, "You YouTubers are so out of this world that you can't understand the importance or ways to deliver a better product by all means. You got numb to people's real problems. The performative marketing is the only thing you see and appreciate." Theater people, YouTube people, influencers. Brother, I'm not a theater person. Like, what do you, what do you think? I spent my whole career as a developer, building, and

• 25:32Scott Tolinski?

  you're gonna say that all I understand is performative marketing, because I understand that the three people at GitHub that it takes to produce a promotion like this absolutely influence their infrastructure. Zero percent. Zero. Big old zero. So, you're on my, uh, you're on my list.

• 25:53Speaker 00

  Nice.

• 25:54Scott Tolinski?

  Yes, and I, to be clear, I mean my, my shit list, not, not any other kind of list.

• 25:59Speaker 00

  I will say, Scott, for, I, I saw this news story in my RSS feed, so I'm subscribed to the GitHub, uh, like, blog posts and stuff. So I saw it, I clicked the link and I signed up. Done. Because it sounded cool. It's like, let's, let's get the, the syntax repo on a CD. That'd be really cool. Um, so I was not there for all the discourse, but I, I think people always find something to complain about. Um, I, I could see it, but yeah, you have to understand, just like, it's not, and also, we're like, I'm not backing big, big corpo here, but you do have to understand the nuances of a large corporation. Like, there's so many internal organizations, so many things happening, that like, I don't think the marketing team talked to the infrastructure team to be like, "Hey, do you think people

• 26:40Speaker 00

  will be mad if we, uh, release repos on a CD?" Like, nobody would even care to have that cross-organization communication, so.

• 26:49Speaker 00

  I don't know. I think it's a cool idea. Like, physical media is back, too. I think the marketing team was on the right track, but it's potentially, maybe the tweet itself, like, one of the chats just said, "I'm just a bit confused about, 'We heard you, we agree,' which is the first sentence, the first sentence of the tweet, which, you know, maybe people didn't necessarily say they wanted GitHub on a CD, but there was, they're trying to tie in, like you were saying, trying to tie in to all the news about PlayStation and no more physical media.

• 27:18Scott Tolinski?

  Yeah, it's a direct call to the PlayStation stuff, which had blown up. And they assumed that everybody would, would know that because that makes sense, because it was all over the place. And not only that, other folks were doing the same meme. This had been memed several times, so they weren't the first ones to respond to this and do something like this. Even other tech companies had made jokes about giving you stuff on Dats or, or tape or anything like that. So I, I thought they, they expected people to understand that and not make a joke about their uptime. I do want to shout out Frankie for the, "I understand the optics for a CD-ROM is on point." Yes, Frankie, absolutely. Optics,

• 28:00Scott Tolinski?

  CD-ROMs, my god. Dude, I, I just thought this was too ridiculous. It's like, GitHub is always known for doing fun and funny stuff like this. That's like, always been part of their brand. The Octo Cat even as itself is a fun and funny thing. And I'm sorry that you are immune to fun, and you don't get the joke, but I just don't, I don't get getting grumpy about this. I don't, I'm grumpy about getting grumpy about this, I'll say that.

• 28:29Speaker 00

  Yeah. But I did see that, because you were backing up a large corporation, they did say, one of the tweets said they were gonna give you two CDs, Scott, so I think that's, that's good for you as well. I don't know.

• 28:40Scott Tolinski?

  Oh, yes.

• 28:41Speaker 00

  I don't know if the person in charge of it, but they're like, "You get two."

• 28:45Scott Tolinski?

  Do, do you know who that is?

• 28:46Speaker 00

  It was like a reply, I think, to your, to your quote tweet.

• 28:50Scott Tolinski?

  Yeah.

• 28:50Speaker 00

  It was somebody that works at GitHub.

• 28:52Scott Tolinski?

  Somebody that works at GitHub. Let me, let me tell you who that is really quick, because it is not just somebody that works at GitHub. It's Kyle Dagel. Kyle Dagel is the COO at GitHub. So, yes, and a long time. He's, he's quite the dude.

• 29:12Speaker 00

  Nice.

• 29:13Scott Tolinski?

  Yeah.

• 29:13Speaker 00

  You get two, which is cool. So, but I think the, and a part of the other news as well was only the first 1,000 people to sign up were able to get one. I don't know how many signups they actually accepted. But like, that's the other funny thing about all of this is like, literally only 1,000 people are gonna get a CD, but the uproar triggered like, tens of thousands of people, it seemed like, so.

• 29:32Scott Tolinski?

  Yeah, the first 1,000, it maybe seems like the first 998, cuz I'm getting two. That's right.

• 29:39Speaker 00

  Yeah, exactly.

• 29:40Scott Tolinski?

  Yes. Sick. Okay. All right, that's all I got here. Just a little, little, uh, little grump list.

• 29:47Speaker 00

  A little rant. Scott's thoughts. Nice. Let's keep moving. So, uh, if you're just joining us, this is Syntax Weekly. Uh, we're talking about news stories, new libraries, new news in the world of web development. And, uh, I wanna let you know that this is brought to you by Sentry. They are, uh, error reporting that is considered not bad by millions of developers. Essentially, throw it on your app running in production. If any errors happen, if anything goes wrong, you can check the Sentry dashboard. It'll tell you exactly what went wrong. You can see a stack trace from the front end to the back end. It also has profiling, so you can nail down like, slow database queries. We use it on the Syntax app. It's fantastic. Check 'em out.

• 30:26Speaker 00

  Syntax, Sentry.io forward slash Syntax, you can get two months free on the Team Plan. And now, I'm gonna tell you about a brand new JavaScript framework. If, if you don't already have AI fatigue, here comes the JS fatigue. Uh, but this framework is called GEA. It's spelled G-E-A. How would you say that? Gia? GEA?

• 30:50Scott Tolinski?

  Gia?

• 30:51Speaker 00

  Gia? Gia's a good middle ground. I like that.

• 30:53Scott Tolinski?

  Gaia?

• 30:54Speaker 00

  Gaia? I'm gonna go with Gia. Um, but their tagline is "Write JavaScript, Get Reactivity for Free." Now, I, this actually, I saw this a few weeks ago, but didn't pay too much attention to it. But it, it got on the front page of Hacker News. I started looking into it. It looks very interesting. It says they're the, a compiler-first UI framework that turns ordinary classes and functions into surgical DOM updates. No virtual DOM, no hooks, no signals, just your code wired at build time. And you know what this tagline sounds like? Sounds like Svelte. Yeah, it sounds like Svelte. It's like what Svelte tried to do. And so, like, looking through the docs and like, looking at their examples, it very much seems like Gia is kind of like a middle ground between Solid JS and React.

• 31:40Speaker 00

  Because even Solid JS tries to do a similar thing that Svelte does, which is like, compile ahead of time. But you can look at their, their charts. And they show you that a simple hello world of Gia is only 214 bytes. Whereas a hello world in Solid is 10,000 bytes, Svelte is 23,000 bytes. So they're really trying to harp on the idea of like, we have a really optimized compiler that only ships what's needed, no extra stuff. Because I think even Svelte now ships like a tiny runtime for like reactive variables using proxies. And there's a little, there's a little bit of runtime code, but in this case,

• 32:21Speaker 00

  Gia is thousands of bytes smaller than what you would ship with Svelte. And they also have a chart that's not just a hello world, like a to-do app, because that's where the real stuff comes in. But this also shows that it's a lot smaller. So Gia with Minified JS is 15,000 bytes, Solid is 16,000 bytes, and Svelte is 38,000 bytes. And so that's what I saw as interesting to me is like, you can see how close it is to Solid. And I'll show you the code and talk about the code in a second, because it uses JSX. But this was kind of the idea behind Solid as well, like, write code that looks a little bit like React, but a lot of it's going to get stripped away. There's no virtual DOM, it uses signals. So it's very interesting. Any thoughts so far, Scott?

• 33:06Scott Tolinski?

  My thoughts are a little bit of, I hate the, the, the text treatment on this website. I'm sorry, Gia folks. It makes it very difficult for me to read this page. Now, I- I think it's cool. I like this, obviously, because I like Svelte. But in the same regard, I don't think this is moving the needle enough for me. That said, when I looked at the code, I did like the way that it looked. And it felt very organized in the way that I like to write code. I don't really love JSX, so for me, that's a bummer.

• 33:47Scott Tolinski?

  But how they handle stores just being classes, how they handle components just being classes. I like all that. And the fact that you're really shipping nothing, that's all good. Do I see this going anywhere? You know, I don't know. But I do think it's a good, a cool thing to try. And I'm down for people trying this kind of stuff. One thing I saw was interesting was there is a... We're all just gonna be saying this differently, I guess. Gia UI, which is based on zag.js, which I didn't know about anything like that. So one thing I gotta love is a framework taking UI components seriously.

• 34:27Speaker 00

  Yeah.

• 34:28Scott Tolinski?

  And that zag.js, which I'd never heard of before you showed me this, is a UI components powered by finite state machines. So it's a state machine-based UI components library. Now granted, when I went to look at zag, I didn't really like how the components looked. This felt used, like, use language from React. It used writable stores instead of hooks, or not hooks, runes. And in general, like, it's doing a lot of spreading. So zag, I didn't really love, but I do love that this framework took UI components as being an actual part of the framework.

• 35:12Speaker 00

  I love it, because that's one of the things that I don't like about the React ecosystem, is there's just so much choice around it. I mean, you could say that nowadays, people have kind of settled on using something like shadcn or, and like, that's built on top of base UI. But a framework launching and thinking about that is a good thing. So they have things like a button component, a card component, input component. And then in terms of interactive components that use zag.js, you've got like dialogues, tabs, accordions, so...

• 35:40Speaker 00

  I think it would make sense, like if you're shipping a new framework in this day and age, like you have to have, it's kind of gotta be a bit batteries included if you want to even slightly compete with some of the other frameworks out there.

• 35:50Scott Tolinski?

  Yeah. Henri's asking about the Bratley versus Gia bit, Bratley being compression algorithm in the browser. And I, I think what that's saying on their website is that it's saying that here is the size of Gia pre Bratley, here is the size of Gia without Bratley, and here's how much compression greatly improves Gia.

• 36:17Speaker 00

  That's how I read it as well. I think that's one of the aspects is it's also potentially more compressible for whatever reason, but I would, I would read this third column as Gia with Bratley is 30 times smaller than Solid, and it's 70 times smaller than Svelte.

• 36:37Scott Tolinski?

  That's what I'm thinking as well, yes.

• 36:40Speaker 00

  Yeah, it's a little confusing because the column title says Bratley versus Gia, but I think they mean comparing Bratley Gia size to given framework size.

• 36:50Scott Tolinski?

  Yes.

• 36:51Speaker 00

  Yeah. Let's talk about the code though, because that's, that's the thing that a lot of people care about. And I will say, like, this is one of the other things that I like about it as well. I will say, I've, I've been a recent critic, a loud critic of React, and I will give you more details because one of the things I don't like about React is the entire hooks system. The, every hook that you write inside of a component is run on every single render, not necessarily invoked, but it's recreated on every render. Basically, everything you put inside of a React function component has, that code has to run on every re-render, and I don't like that. I think it's a very silly model, and SolidJS solved this, where essentially you write your SolidJS component, it runs once, and then you have effects that can potentially run later, but you don't have

• 37:36Speaker 00

  this whole idea of this component runs over and over again. And yeah, Gia does that as well. So they don't run over and over again. And the idea is if you're going to build a component that needs internal state, you actually use a class component. So it kind of goes back to like the old React way of a class extends component. And then instead of a render function, you have a template function. But what that means is you can have class methods. So you can have like state update methods or effects, but you just call them as instances on the class. And I actually liked that. It's a better mental model for me in terms of what the code is actually doing versus how I write it. Because in React, you have all these use effect, use memo, use callback. And it's very hard to reason about like, what order does it get called in? Where does it live? That kind of thing.

• 38:21Speaker 00

  So you essentially, if you want a component that has state, you have to create a class that extends component. But they also support functional components, but purely for presentational UI. So the a functional component looks exactly like a function component would in React, except there are no hooks. You can't have any use effects. You can't have any variables or state inside of here. It's purely presentational. And so I like that split where like, if you want interactivity, if you want state, you have to put it into a class component. And then you have to create methods on it. So I'm a big fan of that. They have an example, like there's a a doc on their site called React versus Gia. I would say, take a look at that if you're a React developer to kind of see the differences. But there's a to do app here in React.

• 39:06Speaker 00

  And this is kind of what I'm talking about. Everything is defined above the return. You've got your use state. You've got your methods. Whereas if you contrast contrast that with Gia, your state are just properties on the class. And then your methods to affect that state are are just methods on the class. And I guess they actually separated that out into a store. So they you can have this stuff inside of a class component, but you could also have a separate object store that has those methods on it and then use that store inside of your component. So.

• 39:36Scott Tolinski?

  This feels, yeah, this feels like old React mixed with Svelte. And I think in a nice way, in the UI stuff, like it being very CSS variable focused is great. They're having some nice little examples inside of here. It's interesting and I'm happy that someone's taking an approach like this. So I think it's fascinating. Am I gonna reach for it? I mean, I'm so firmly, my tooling and everything is so firmly embedded in the Svelte world that it's like, I'm gonna have a hard time using anything else at this point. But it's still cool. I think it's cool. And I think this is the type of lesson I think other frameworks should pick up.

• 40:19Scott Tolinski?

  Or people who are looking for something to build in an ecosystem could think of like, all right, what is stuff like this doing that my framework is either missing? Or I would like to have in a bundled up all nice kind of way. I think it's cool. Yeah.

• 40:33Speaker 00

  I think, yeah, I think just having people thinking about how to do this better is great. Because this is another like argument I have about React is we've kind of stagnated. There's not a lot of innovation. AI is just spitting out the same old React. And so in a sense, that kind of feels like we're stuck. Like even if something like this comes along and it does it better, it doesn't matter because it doesn't have the critical mass. And AI is just gonna poop out like bad React code with a bunch of use states and use effects in it anyways. But something better might come along. I like that people are at least taking the time to think about it. And I do, I like the mental model here. I think going back to class components is a good play, especially for being able to structure your code in this way. I like it.

• 41:19Scott Tolinski?

  Word. Yes. Cool. Alrighty, well, let's keep it moving, folks. Syntax Weekly, we're bringing to you all the cool stuff. I wanted to talk a little bit about Fable. Let me even just show this. So last week, we got Fable.

• 41:38Scott Tolinski?

  And we received notice that the Department of Commerce has lifted export controls on Fable 5 and Mythos 5. We will begin restoring access. And then again, it took towards the end of that day, I know this because I was checking, and Fable was restored. And I gotta say, I have used all of my Fable credits as of this morning. So I'm officially booted from Fable because that takes me up until past the deadline at which they will be removing it, which I believe is the 7th, so tomorrow, even though my my usage limit said it would reset Thursday. Either way, I no longer have access to Fable.

• 42:19Scott Tolinski?

  I burned through all my Fable credits. I'm here to tell you that on a vibes level, it's definitely the best model that I've used. It's done a whole lot of stuff. And one thing that I really liked was the ability for it to execute large things really reliably. But again, one thing that I have not used it on, CJ, is I have not used it on any projects where I actually care about the code. I've been using on my Vibe Coding Slop Fest stuff, and I had to go back and forth with it less. It was able to do more complex things, and it did so better and with less handholding. So again, I, as somebody who's used, feels like I've used everything that's come out,

• 43:04Scott Tolinski?

  I think the hype around Fable was real. In fact, I can even show you some of the stuff it built for me. That's it. Yeah, it, so Papers made shaders, I think it was Paper shaders. They released open source shaders. It, one thing that it did for me, where is the water? Water effect. One thing that it did for me was port it to my WebGL project here, and

• 43:32Scott Tolinski?

  Actually, this is not, it's WebGPU, not WebGL. I get those messed up here. And so I'm building a programmatic, what do you call this? You'd call it like a programmatic video interface to generate video. So whether that is, whether that is titles-

• 43:51Speaker 00

  Animations, like think After Effects style things as well.

• 43:54Scott Tolinski?

  Yes, yes, think After Effects with sound effects and all this kind of stuff. So I'm building it to be programmatic that AI agents can create these things, or you can create, every single one of these examples can be done in JSON. So these all exist as JSON. And they all have various different things. And it one-shotted to me, which, this one actually, I've tweaked recently, but it one-shotted for me this kind of stuff, where I said, "Take me Mermaid diagrams and turn it into a timeline, and then have several different themes for it," or whatever. But it one-shotted the ability to translate this application into Mermaid diagrams and have them look nice.

• 44:34Scott Tolinski?

  Which, again, Opus to me, or Opus, Claude models to me remain goaded on having a better aesthetic, just a better aesthetic barometer. But it did quite a bit of restructuring this large repo, which this repo, very large, very interesting, and it did a lot to take every single thing that I had working as just animations and turn it into something that could be manipulated by the UI, something that allowed for keyframes. Oh yeah, built a whole keyframing system for me. And these demos right now, CJ, I will say, these demos are just kind of like tossed together,

• 45:17Scott Tolinski?

  demos to have a demo. Let me do, let me do for keyframes. It built this whole keyframing system that is like-

• 45:27Scott Tolinski?

  Let's see, like a DaVinci Resolve style keyframing, where you can modify various properties on the keyframes and have them... I mean, it's DaVinci Resolve-esque, you know? It's pretty crazy that it was able to do this. So, I personally, I mourn the loss of my Fable credits, because I think I was really enjoying my time with it. Did you get a chance to use it?

• 45:54Speaker 00

  I can talk a bit about my experience with it, and I'm sharing my screen now to show my Copilot usage, because I'm all out of Copilot credits. But over the weekend, I was building a physical media management app. So I've been ripping, I've been finding CDs at thrift stores and record shops and ripping them. I'm trying to create my own, basically, streaming server of music. And I was planning out that project, and I thought a good test would be to use all of these four models to plan that project. Essentially, I have a folder of artists, albums, files. I needed to scan that and then also sync with MusicBrain's API to show me metadata about the actual albums and everything.

• 46:37Speaker 00

  So I wrote out a big prompt, copy-pasted it into all three. One aspect of this is KIMI K2.7 is now available in GitHub Copilot, so I was able to test with that. But I basically pasted that prompt in. I described what I wanted to build. I described my tech stack. And this chart isn't exactly right, because both OPUS 4.8 and GPT-5.5 were used for building it. But I will say, you see this bar graph here for Fable? That's how much usage it took to do the same amount of planning. Now, I will say, like, the one piece of it that made it feel better was it went really deep. So for the plans that were generated by GPT-5.5, OPUS 4.8, and KIMI 2.7,

• 47:21Speaker 00

  it basically took the prompt that I wrote, maybe did a few documentation lookups, and was like, "Okay, this is the plan that I'm going to present to you." But Fable just went, just started researching. It was like, okay, this is what you want to build, but I see you also want to talk to the Amazon API. And so it started reverse engineering the Amazon Music API. And like, it was just tool call after tool call. So it definitely is more agentic. Basically, the way that they've trained it is it will try and try and try more before it's done. And so that's really the only thing that I could see before I completely ran out of usage credits on GitHub Copilot. But I will say, comparing, comparing the four models, that was the main difference. Like, it took the prompt and also rolled with it.

• 48:04Speaker 00

  Like, it basically did things that I had to later prompt the other models for, but it just went ahead and started doing it. So I could, I could see myself planning with Fable 5 and then maybe implementing with GPT-5.5 or OPUS-4.8.

• 48:20Speaker 00

  But yeah, that was my experience with it. And like I said, I ran completely out of credits. If we're talking about AI usage, I actually found that it's cheaper cost-wise to pay $100 a month for Codex, which gives you access to GPT-5.5. Because I've been running some, some agentic flows, building out that music app over the weekend, and it barely touches that, that usage for Codex. Whereas I was able to code for like, maybe two hours, and then it completely ate through all the usage of my GitHub Copilot.

• 48:51Scott Tolinski?

  Yeah, Copilot, man, that usage even without, even without using Fable is not that much. So that's not that surprising. I, I do, like I said, I, yeah, I agree with you. I think it does a really good job of having a wide view of projects, just like it said it does. In my experience, it's, it does way better when you have a big project and you want to plan and make big changes across that project. And I feel like when I asked it to do something that was more likely to affect many things other than one localized spot, I found it personally to be much more reliable in the end result.

• 49:30Scott Tolinski?

  So that, that said, the next big news on the clock is that Soul is going to have Ultra version available in the Codex, in your Codex app. So. Soul, the latest, biggest model from OpenAI that is not yet released, will be available in its most juiced up form, its most roided out form in the Codex subscription. So for me, I'm kind of stoked about that because what is it? GPT-5 Pro isn't available and I like to use the biggest, baddest model at any given point to do some testing. So I'm going to be excited to see whenever this drops.

• 50:13Scott Tolinski?

  I get the personal feeling that it's going to be available very soon here, like this week soon. That's based on vibes.

• 50:23Speaker 00

  And I saw news about 5.6, is that grouped in with the news with Sol Ultra? Or is that a separate release, GPT-5.6?

• 50:31Scott Tolinski?

  Isn't 5.6 Soul? 5.6 is Soul.

• 50:33Speaker 00

  Oh, it is Soul. Okay, okay.

• 50:34Scott Tolinski?

  Well, Soul is, sorry, GPT-5.6 is in three different models. It's Soul, Terra and Luna, yes. I believe Luna being the smallest, yes. Terra, Earth, yes, got it.

• 50:47Speaker 00

  I see, I see. That makes sense then. And I've, like you mentioned earlier, I do prefer the anthropic Claude models. But now when it's coming down to like token usage and pricing, it's like I'm settling for GPT-5.5 because it's like I don't want to pay for the Claude, the Claude usage tax. So it'll be nice to see 5.6, to see, is it on par with Opus? But again, it also comes down to vibes, like the vibes of Opus output is just so much, it's like I can't describe it because it's just vibes versus the output of GPT-5.5. Like GPT-5.5 feels rigid, it feels less creative.

• 51:27Scott Tolinski?

  Feels exact and that can be good, but it also feels like it wants to write his record over and over again and have one function call another function that calls another function that calls another function that eventually does the work. Like, that's the reason why I have a hard time with 5.5. Because to me, I feel like I have to babysit its results more. But maybe that's just the tools and how that I have mine set up. The thing I do love about GPT models is that I can use them in my dearest open code and I do like my open code setup significantly more than I like Claude code. So for me, I will be stoked to try it out because I do, right now I use GPT-5.5 for some things and I use, like for instance,

• 52:13Scott Tolinski?

  My robot, I find the robot programming to be way better with GPT 5.5. And I don't know if it's because there's a lot of Python involved or what. But I just have not found Claude to do a good job at this dang thing. So, um, yeah, I use them both. I think they're different. Co-Capsy.

• 52:33Speaker 00

  I think that's kind of what you got to do. Like, that's why I like having a Copilot subscription is because I can try both the OpenAI and Anthropic models. But like I said, now I'm kind of leaning into just the Codex models specifically because I can use it with OpenCode. And that's where I've figured out all my like agentic flows is with OpenCode. And you can't use Claude code with it. You have to use an API key.

• 52:57Scott Tolinski?

  Yeah, yeah. Do we want to do a robot update? Somebody said my robot looks good.

• 53:02Speaker 00

  Yeah, let's talk about it. Yeah, so for those listening in, Scott has this cute little handheld robot. It's about, was it like a foot tall? How tall is that thing?

• 53:10Scott Tolinski?

  Yeah, he's a couple, yeah, a foot and a half maybe with the antenna, a foot tall without the antenna. So Randy, if you want to show my screen, I, we named him Marshmallow. I let the kids name him. And Marshmallow, I have now built in this, you can tell I've been using GPT for this design. This is a GPT asked design right here. And I've built this whole thing where you can, let's see if I can make him dance. And let's see if this actually works. Oh, is he, is he not recovered? Every single time I show the robot on stream, it doesn't work. Okay, it's my fault here.

• 53:43Speaker 00

  It's the demo.

• 53:44Scott Tolinski?

  Oh, there he goes.

• 53:45Speaker 00

  Oh, wow, yeah, he's moving his head.

• 53:48Scott Tolinski?

  Yes, he's moving his little head.

• 53:51Speaker 00

  Oh, so this is the, you have emotes. This is the curious emote.

• 53:54Scott Tolinski?

  This is the curious emote, yes. Yes, ooh displeased. I wonder what displeased looks like. Either way, this is the Richie Mini. He's built by Pollen and Hugging Face. We talked about him before. And inside of him is Wi-Fi and a camera and a speaker and all kinds of stuff. And I built in like wake listening so you can say, hey, Marshmallow, and he'll respond to you. You can do start conversation. Hey, Marshmallow, how are you? Oh, well, some, I found the timing to be the hardest part about this because you're trying to do, gosh darn it, every single time.

• 54:36Scott Tolinski?

  Every single time, folks, it like it works perfectly fine. Hey, Marshmallow, what's up?

• 54:43Scott Tolinski?

  You thinking now? Oh, there we go.

• 54:47Speaker 02

  Yay, I'm up and ready.

• 54:49Scott Tolinski?

  Yes, Marshmallow. Yes. That's so great. We're so happy you're up and running. I gotta fix this timing.

• 54:56Speaker 02

  Got it, Ricky Marshmallow, yes.

• 54:58Scott Tolinski?

  Yes, got it, Ricky Marshmallow, yes. Okay, we're gonna stop the conversation. Uh, Marshmallow is up and running, folks, and, uh, man, this has been a fun project, but I gotta say the, um, yes, you are. You are up and running. Uh, it's it's been very fun to work on a project, but the, um, back and forth with like this type of thing has been tough because it's like, all right, uh, how do you how do you debug the like latency between how fast he responds? I'm using GPT something mini, um, I think I think 5.5 mini even. Um, but it's it's only whatever is available on Hermes, because this bad boy is hooked up to Hermes. That means he's got a memory system. He's got access to a whole bunch of tools.

• 55:43Scott Tolinski?

  That does blow up the context a little bit. It is going to increase the latency. But if I ask him like, hey, do you know, uh, who my kids are? He's gonna know who my kids are, and then he can remember things about them. He's got a favorite color that he came up with himself. And as a fun little thing for my kids, this has been a blast. They were doing math with Marshmallow. They were doing math with Marshmallow, uh, yesterday, and they were just having the dang most fun with it. And I didn't even tell them to start doing math with Marshmallow. They just started doing it. So, reteam any, folks. This is a cool project. You have to build it, though. And then, again, it like, uh, you have to build it. There's no internal thing, so it has to be connected to Wi-Fi to a computer to run the daemon, but,

• 56:28Scott Tolinski?

  uh, yeah. It's a robot update. Robot update for y'all.

• 56:34Scott Tolinski?

  Yes, getting a little bit better every single day. Another thing I wanted to talk about was Junior. I have long been a hater of Slack and agents. And I checked out Junior just because this is another Kramer project over at Sentry. This is from Sentry Labs, junior.centry.dev. And I definitely have never been one to really want any kind of agent system inside of Slack. And then I talked to Dax, and when we interviewed Dax, he mentioned that their team, the way they collaborate talking to agents, and I thought it was really interesting. So I said, oh, you know what?

• 57:15Scott Tolinski?

  I know Kramer's working on a Slack bot, so I made one for us using Junior. And I gotta say, it's been awesome. So Junior connects to all of your stuff, the plugins that come with it. Again, this is more, you could think of this as more like a work tool for developers inside of your Slack channels. The plugins that they have are CloudFlare, Datadog, GitHub, Hex, Linear. There's a memory plugin, Notion, scheduler, Sentry, Versale. And I have connected this thing to our YouTube analytics, so we can ask it questions about YouTube. I've connected it to our Linear and our Notion, our email marketing stuff. And you can have all these things.

• 57:57Scott Tolinski?

  It's got a database, so it can store things in a database. And man, I gotta say, it's been kind of cool. Like, what was it? Today, we were like, we need to be reminded of when we have holidays off. So I said, by the way, I named our agent Tasty. I said, Tasty, give us a reminder every Monday for any upcoming holidays in the Netherlands, Canada, and the US. And Tasty's just like, no problem. We'll remind you on Monday. Based on our own company's internal stuff.

• 58:26Speaker 00

  Yeah, and so we're actually working on a holiday. You're all welcome for that. But so does it also like modify its own code? So when it said, yes, I will do that, did it create a PR with that skill? Or did that get stored as a memory somewhere? Like, how is that gonna actually?

• 58:42Scott Tolinski?

  It's a cron job.

• 58:43Speaker 00

  Okay, and so when we talked to it to do that, it literally just talked to this plugin to schedule that thing to happen?

• 58:49Scott Tolinski?

  Correct.

• 58:50Speaker 00

  I see, I see. Okay.

• 58:50Scott Tolinski?

  Yes. I have been working on it via GPT-5 and with Claude models to add in the YouTube Analytics plugin. I'm adding in a podcast analytics plugin itself. And because they have all these plugins already here, it was pretty trivial to get this up and running. But I gotta say, I'm kind of into the idea of having agents in Slack on your work team. I, yeah, I mean, it created a PR for the syntax repo that was really nice and simple. And it did so first try, gave you a preview URL, checked it out. And like, yes, that's not that much different than using an agent.

• 59:32Scott Tolinski?

  But like the fact that as a team, we could talk about it in chat and somebody could say, oh, I wonder we should do this. And just right there be like, hey, Tasty, go and do it. And Tasty's like, no problem.

• 59:42Speaker 00

  I like it. I like the, just the fact that it integrates with all these things. Cause so many times we're jumping between services because we use one thing for our email scheduling. We publish videos to YouTube. We want to also see podcast analytics. We use Linear for task tracking. We use Notion for documents. Like all of the things we use to work as a team are spread across all these places. And so what an agent is really good at is getting like API access to all of these things. And then kind of mashing them together based on what we're talking about in the channel. So I really like that aspect of it. I will say.

• 1:00:14Scott Tolinski?

  Yeah. I hate Linear, sorry, or Notion.

• 1:00:17Speaker 00

  Yeah, okay. Notion, yeah.

• 1:00:19Scott Tolinski?

  Yeah.

• 1:00:20Speaker 00

  But I, I'm very security minded. So the fact that we've given this bot access to all of these things. I want to figure out how to break it. I probably shouldn't have said anything because I'm going to make a video about it. But.

• 1:00:32Scott Tolinski?

  CJ, you're fired. If you do that, you're fired.

• 1:00:35Speaker 00

  Don't change your prompts. No, because the whole thing is like, I'm going to break it, but also make a PR that will fix it. Because when you give an agent access to lots of systems like this, like if it has write access to certain things, it could potentially be prompt injected. Like if there, if it, like right now, I already looked at it. It had, it only has read access to YouTube. It has read access to YouTube Analytics and YouTube information. But if it had write access, there could potentially be a prompt injection in one of the comments left on a YouTube video. So somebody could prompt inject.

• 1:01:05Scott Tolinski?

  Yeah. If it had write access.

• 1:01:08Speaker 00

  If it had write access and it doesn't. So like, that's a good security posture is like least amount of access as possible. But. I will say, this is also why I'm hesitant to like make my own personal AI agent. Like I haven't gotten in the, in the world of Hermes or Claude Bot or whatever it's called now. Because of all this potential for injections or exfiltration of data.

• 1:01:33Scott Tolinski?

  Yeah, yeah, I know. For me, it's all about what it can read and what it can access. Yeah, I, I think there's definitely threats and, and whatever. But if I'm like, if I'm not having it just automatically read all my email or letting other people essentially get access to things that it can read. Yeah, sure, you could prompt inject it by visiting a site or something like that. But I think Hermes is doing a good job of being very cautious and cognizant of all that stuff. Again, I think it's all about how much you want to give it access to. Yeah, the, the junior stuff is interesting. And again, I, I thought, I even, I think I even said, I will never want to, want to code from Slack. I think I said that directly on this show.

• 1:02:13Scott Tolinski?

  And this Friday, I, or do, even today, I coded from Slack on, with Tasty. And I feel, I feel okay about it. That's crazy. Yeah.

• 1:02:24Speaker 00

  Yeah, I mean, especially like getting back to the first thing we talked about in this video that like, if it opens a PR and we're happy with it, great. Like we, that was literally a time saver because otherwise we would have had to spin up an agent locally or like directly open a PR. So the fact that Junior has access to our repos, has access to the, the history in our Slack channel. So like, you can basically have a discussion about what new feature you want or what thing is missing. Junior has all of that context and then it can go off and start building. So, yeah, I'm, I like it.

• 1:02:54Scott Tolinski?

  Yeah, I will say, uh, it would be really cool for like producer Randy. Sometimes we get in a situation where the site's not building and the site won't build because maybe there's a syntax error in a markdown file or something. And now Randy can just say, Because it has access to the Vercel logs, you can just say, hey, why did the latest build break? Can you fix it? And then it can go and create that PR directly there. And now CJ doesn't have to stop doing what he's doing to go and fix it. So that's awesome.

• 1:03:23Speaker 00

  This is the exact type of workflow we were promised when AI was super hyped up. Because that literally is something that I used to do. Like Randy would be like, hey, the PR won't merge. There's a failure, a syntax error, if you will.

• 1:03:35Scott Tolinski?

  Yes.

• 1:03:36Speaker 00

  And then I would go in and be like, oh, it's this. We can ignore it. Or, oh, I just pushed to fix. But those are like little one-off things that I actually would trust an agent to do. Because a lot of times it is a one-line fix, or it is just a simple look at some Vercel logs. So this is the future we were promised.

• 1:03:50Scott Tolinski?

  Yes. Cool. Folks, we are Syntax Live. We're talking about all the new things on the web here. CJ, do you have something else for us today?

• 1:03:58Speaker 00

  Well, I've got several things. The first thing is good news for the JavaScript ecosystem. So this was a tweet from Colin Hacks, the creator of Zod. But they show that in all of the tools you use in the JS ecosystem to manage your dependencies, whether it's NPM, PNP, Bun, or Yarn, now every single tool supports the minimum release age property. And it's spelled differently across every tool, but every tool now has that as an option. Where essentially you can say, install, let's say you do a PNP install for a new package. If there is a recent publish of that package, with this setting, that package, that new version,

• 1:04:40Speaker 00

  would have had to have been published for a specific amount of time based on this setting. And this is to prevent certain supply chain attacks, right? Because a lot of times when a package is compromised, it's actually caught within the first few hours. And if you have this setting set, that means your code would not have even tried to install or update to that version, because it basically waits a certain number of days or hours after a certain release has been out in the wild. And so this is one thing that your teams should adopt to prevent the spread of supply chain attacks.

• 1:05:14Scott Tolinski?

  Only you can prevent the spread of supply chain attacks. Giant worms.

• 1:05:20Speaker 00

  Yes. Throw this in your config. And so that's one piece of good news. But I'm actually linking this up with a post from GitHub, and the title is "NPM Adds Preventative Account Protection for High-Impact Accounts."

• 1:05:33Speaker 00

  So the idea is if you are the owner and maintainer of a package on NPM that gets millions of downloads, I don't know what their exact classification of high-impact account is, but basically just think some of the most downloaded packages on NPM. Anybody that is a publisher for those accounts, if the accounts running those packages uses one of their two-factor authentication recovery codes, those are the codes that you like write down on paper or save to a hard drive when you first sign up to a service, not the ones generated by your authenticator app. And the idea is if you're using a recovery code, that means you lost access to your authenticator app. So if you use one of those codes, if you change, or if you change your email,

• 1:06:18Speaker 00

  that account is placed into a 72-hour read-only state. And again, this is to help prevent the spread of supply chain attacks because a lot of times people lose access to their accounts in this way, whether they were phished, or maybe they got hacked and got access to those recovery codes. And this will now prevent whoever potentially got access to those things from publishing malicious packages. Now, it's a little bit of an inconvenience for the person, because let's say you do legitimately want to change your email. Now you just have to plan for that and know that for the next 72 hours, you won't be able to publish new versions. So another, another good win for the world of, of NPM and preventing supply chain attacks. I think these are some much-needed things. We've been waiting for like, what is Microsoft going to do? What is NPM going to do?

• 1:07:02Speaker 00

  And I think this is a good, a good step in the right direction.

• 1:07:05Scott Tolinski?

  This is a good step. I just wish everybody would have adopted the same stinking property name for this. Come on.

• 1:07:12Speaker 00

  Yeah. So for our podcast listeners, for NPM, it's min-release-age. For PNPM 10, it's minimum-release-age. For PNPM 11, it's minimum-release-age in camel case or Pascal case. And then bun is also Pascal case. And then yarn is NPM minimal-age-gate in Pascal case. So they're all, they're all different, different properties for doing the same thing. Yeah.

• 1:07:38Scott Tolinski?

  Yes. I love it. Yeah. All right. What else do you got, CJ?

• 1:07:42Speaker 00

  Yes, this will be a quick one, but Chrome has introduced the user media HTML element. So if you're not familiar, if you're writing a JavaScript app and you want to get access to the user's camera or microphone, there is a get user media JavaScript API. But this will allow us to request access

• 1:08:03Speaker 00

  To a user's webcam or microphone with simply having an HTML element. It's just the user media element. It's a very similar to the geolocation element that got added. But essentially, you can add this to your page. And instead of having to call JavaScript to request access to the user's camera or microphone, you can just use the HTML element itself. And so this is much less, I guess, intrusive. And there's also, they talk about it in this article, but a lot of times that API for get user media is a little bit buggy, right? You might land on a website and say, no, block this request. But then the site breaks because it expected you to have access to it. The idea with this element is that there'll also be like

• 1:08:48Speaker 00

  a more centralized way of managing that access on the page. So your scripts are less likely to error out if like it doesn't have access or it's easier to like re-trigger that that request for access if needed. So it's pretty cool. Had you seen this, Scott?

• 1:09:02Scott Tolinski?

  I had not seen this. And this is something that I feel like, yeah, it makes perfect sense. Why don't we have this? So I'll take a user media element all day. Yes, absolutely.

• 1:09:14Speaker 00

  And right now, this only exists in Chrome. I don't even know if it's in Canary or which one. But basically, this is how new features get shipped to the web. It's going to be tested out quite a bit in Chrome. And then other browsers will potentially adopt it once they settle on the API and behavior. So it's another fun one.

• 1:09:34Scott Tolinski?

  Another fun one. User media is a fun, fun stuff to play with. It's a, yeah, a big fan. I've been working on my screen recorder app for a long time. So I've been deep into that world.

• 1:09:45Speaker 00

  Yeah, because that basically lets you record camera microphone to separate files, but you're using Electron. Is that accurate?

• 1:09:51Scott Tolinski?

  I am using Electron, yes. And you can record cameras, mics. You can record infinite amount of, maybe not infinite amount of sources, but as many sources as you want. It'll kick them out of separate files. And then also, there's like, I added a simple mode recently where you can just have it all wrapped up into one file automatically for you if you want, just because I need that sometimes.

• 1:10:14Speaker 00

  Nice.

• 1:10:16Scott Tolinski?

  Cool. What else you got?

• 1:10:17Speaker 00

  Yes, I have two things related to Claude code.

• 1:10:21Scott Tolinski?

  Yeah.

• 1:10:21Speaker 00

  And I use Claude code at work. I don't use it for personal stuff.

• 1:10:25Scott Tolinski?

  Interesting.

• 1:10:26Speaker 00

  But this was, this was a bug that was filed five days ago. And it says there was no response after 60 seconds, continued without an answer. So if you're not familiar, Claude has the ability, and a lot of these other coding, agent coding tools have the ability as well to ask you a question to get clarification on something. So maybe you're writing a plan, it'll ask you, oh, do you want to use this versus that? Or how do we, do we want to approach it this way or that way? And you can answer that question with a prompt or by choosing from multiple choice. But the Claude team secretly added this feature where if the user does not answer within 60 seconds, it just automatically chooses an answer and keeps going.

• 1:11:08Scott Tolinski?

  Wow, the recommended one?

• 1:11:10Speaker 00

  Yeah, well, yeah, it chooses like whatever, yeah. Because a lot of times they'll be like, I recommend you choose this answer, but here are some other options. And so, yeah, it'll choose the recommended option or whichever option it thought. And just keep on coding. And there's a massive thread on GitHub that filed this as a bug that was like, this should not be on by default.

• 1:11:29Scott Tolinski?

  I agree with that, yeah.

• 1:11:30Speaker 00

  There are multiple stories in the, in the GitHub comments of like, this happened to me, like I was running three different sessions and I don't always have every session in focus. So it just kept on coding and then like did some stuff in production I didn't want it to. And I think like, that's the other aspect of this. They actually had detection to see, okay, if the terminal is not focused, meaning like you're using the web browser or some other app, that's when this will kick in. So it's never if you are, if you directly have your, your, your terminal focused. But yeah, people went wild in the comments. Yeah.

• 1:12:03Scott Tolinski?

  I never have my terminal focused when it's cooking on something. I just let it go and then look at something else. Oh, I'm looking at something else now, yes.

• 1:12:10Speaker 00

  And on macOS at least, even Claude code running in the terminal has notifications. So like, if it does ask me a question, I can see, oh, a notification pops up. I'm going to go take a look at that in a second when I'm done with whatever else I'm working on. I would never want it to just keep on coding if I don't respond for 60 seconds. So I do believe they, they, and here's the other issue. They fixed it, but they, they left it on by default and then people kept complaining more and more. And finally, it's off by default, I do believe.

• 1:12:41Scott Tolinski?

  All they had to do was prompt, prompt their model to fix it. And you know, if nobody takes a look at it, it will fix itself. Right? That's the, the feature? Yeah.

• 1:12:52Speaker 00

  Exactly. So yeah, this is, we've talked about this before, but the fact that Anthropic owns the harness, right? They own Claude code. They can kind of decide how it works. In a sense, auto continue after 60 seconds makes them more money because it just starts consuming tokens. You would expect that, I mean, and this also gets into what we were talking about at the beginning of the video of like, if you don't review it, or if you don't have the chance to review it, it's just going to keep on coding. So it's kind of you pushing you in that direction anyways. And then I have this related story, which it's something else that Claude code was doing in their codebase. And the title is Claude code is steganographically marking requests.

• 1:13:34Scott Tolinski?

  Okay. That's a word. What's that word mean?

• 1:13:36Speaker 00

  So steganography is when you hide information inside of something. Typically, you see it with images. So like steganography can like hide a message inside of an image or something like that. But they use that word to say that the way that Claude code, the CLI, is writing its prompts is allowing them to do certain things on the backend. And they basically reverse engineered the Claude code locally. But the idea is when you send a prompt with Claude code, it has a piece of text at the top that says today's date is year dash month dash day. And that's the standard format. But there is some code that will detect, like if you're running Claude code from the

• 1:14:19Speaker 00

  Asia Shanghai time zone or the Asia, you're, you're, you're, you're, you're, I'm, you are, you're, you're, you're, I'm, you're, you're, you're, I'm, you're, I'm, you're, I'm, you're, I'm, I don't know where that is. Is that like Russia or? I don't know.

• 1:14:31Scott Tolinski?

  I'm gonna go ahead and say I do not know.

• 1:14:33Speaker 00

  Okay, I'll look it up. I'll let that run in the background.

• 1:14:37Speaker 00

  Northwest, oh, it's northwestern China. So they're actually just two different Chinese time zones. So basically, if it detects that you're in either of these two Chinese time zones, it changes the prompt to have slashes instead of dashes. And then it changes the apostrophe to use a different Unicode value instead. So if you were working on code in China and you were even, let's say you were just watching the request that Claude Code was making, you would kind of have no idea that it basically just fingerprinted you as someone using Claude Code in China. But the way it fingerprints is it uses a different apostrophe character and uses slashes instead of dashes. The other thing is it detects, I think, like what domain you're running from.

• 1:15:23Speaker 00

  And this is essentially to help prevent distillation attacks where some of these Chinese labs like DeepSeek and some of the others like ZAI are potentially prompting Claude, saving the responses and then using that as training for their own models. So it detects where you're running from in terms of your domain. And there's a whole list of like DeepSeek, Moonshot, Minimax. And all of that is passed in the prompt as well. So all of this to say that the Claude Code harness that you're using locally is able to do things like this, like auto accept your code after 60 seconds if you don't respond, or let their API know on the backend that you're running this code from a specific

• 1:16:08Speaker 00

  part of the world or from a specific company, which is fascinating. I mean, like, they built the tool. They have every right to do that. But the fact that they're doing it is interesting because they could be doing it for other things as well.

• 1:16:19Scott Tolinski?

  For sure, yeah. And I'm sure they are. Yeah. Shout out to Mitchell Wilson, by the way, in the chat. "I made a chat in your terminal that is literally a single bash script. I explained it on my blog." All right. It's called Ask: Stupidly Simple and Useful. Ask, dude, how do you feel about the Ask UIs in these chat tools? Do you prefer it to be a quick, do you prefer the 1, 2, 3, 4, pick an option? Or do you prefer just a normal chat box to write in your response?

• 1:16:56Speaker 00

  I like both and Claude Code lately has been giving me that. Basically, it'll give you like three options. Do you want to do it this way, that way, or that way? Or write your own response? And so a lot of times I'll choose that. Like, the options you gave me aren't exactly what I want, but let's talk about this a bit more. Yeah.

• 1:17:13Scott Tolinski?

  Yeah, I like the UI. In fact, I get grumpy when it doesn't give me the UI. It's like, especially when it gives me like, pick one of these four things just in the text, but it doesn't give you the UI to click them.

• 1:17:24Speaker 00

  Ah, I see. You have to type 1, 2, 3, or 4. Yeah.

• 1:17:27Scott Tolinski?

  Yeah. It's so stupid. I prefer the UI. Ah, yes.

• 1:17:33Speaker 00

  Nice.

• 1:17:34Scott Tolinski?

  Cool.

• 1:17:35Speaker 00

  So I think...

• 1:17:36Scott Tolinski?

  Yeah.

• 1:17:37Speaker 00

  Were they sharing a site that they built?

• 1:17:39Scott Tolinski?

  It's on CS... We should probably make sure of what the site is before...

• 1:17:44Speaker 00

  But we'll double check it and we'll come back to it. We can show... But yeah, thanks again to Mitchell for the $10 donation. We appreciate that.

• 1:17:50Scott Tolinski?

  Yes. Okay, I verified the site is just fine. Or at least, on the surface, it looks fine. I've not... Which is good enough for me. Let's share it. Okay. You got your screen up, CJ?

• 1:18:05Speaker 00

  Yeah, so I was just...

• 1:18:07Scott Tolinski?

  Oh, I posted it in chat. C-S-I-M-W.com. Cool stuff and interest of Mitch Wilson.

• 1:18:17Speaker 00

  Can you share your screen?

• 1:18:20Scott Tolinski?

  You don't trust Mitch Wilson?

• 1:18:23Speaker 00

  Here's the thing that Mitch Wilson could do. He could grab my IP address, and he already has yours, so just share your screen instead, so he doesn't have to get mine, too.

• 1:18:30Scott Tolinski?

  Let's check out his network request here. Mitch, I'm trusting you, Mitch.

• 1:18:36Speaker 00

  Yeah, but it more so comes down to like the server side, because the requests originated from your IP address. I don't know, I'm having internet issues. I don't want to have more internet issues.

• 1:18:46Scott Tolinski?

  I'm on a VPN, Mitch, so don't try anything.

• 1:18:48Speaker 00

  Amazing.

• 1:18:48Scott Tolinski?

  I'm behind six firewalls.

• 1:18:50Speaker 00

  And I'm not on one, so Scott, go ahead and...

• 1:18:53Scott Tolinski?

  Yeah, I can't. I'm sorry, I lost the dang screen, the share screen. In classic me fashion, I got too many dang links open. Yes. Hold on one second. My god, video ninja. I got so... Maca West like grouped them all together now, so when you do like the exposada show, all of them, I like cannot find anything I'm looking for. So sorry, Mitch. Ask, chat with your AI tool. Oh, you are tracking me, Mitch, but I blocked it. Blocked that. Blocked.

• 1:19:25Speaker 00

  Was it just like a Google Analytics script or something?

• 1:19:27Scott Tolinski?

  I'm 100% certain. Oh, Azure. Okay. Application Insights. Yes. Okay. Okay. You know how many times you're in the terminal grinding way and suddenly you need to know some random thing? Like, how do I grip for that specific pattern? What's the flag for that tar command again? Yes. So it's basically a quick little ask CLI tool that prompts through the Open Code Zen API and prints the response directly in your terminal. That's it. Easy.

• 1:19:58Speaker 00

  Yeah, I've thought about building something like this before, but like with a local model. Cause yeah, like I don't know the number of times I've gone to Claude or ChatGPT, just like, remind me the syntax for this command.

• 1:20:10Scott Tolinski?

  Yeah.

• 1:20:11Speaker 00

  That's pretty cool. I thought of like also pairing this with like natural language as well. So you like pre-pend your prompt with the name of whatever CLI tool you build or whatever it is. And then plain text, like, I want to list all files, including hidden folders and show the size in human readable format. And then it instantly figures out the right command and runs it.

• 1:20:32Scott Tolinski?

  See, that's kind of like, I mean, without the commands, that's kind of what I want the robot to do. I want the robot to just be like, hey, what's this? Hey, what's that? And, and just have him on my desk in like voice prompt at any point. I actually might hook him up to my foot pedal.

• 1:20:47Scott Tolinski?

  I've been loving my foot pedal to prompt and stuff. Maybe I could have a specific trigger for marshmallow. That way I don't have to rely on the wake word detection and I could just say, hey, what is this terminal command? And do it Mitch Wilson style.

• 1:21:00Speaker 00

  Nice.

• 1:21:01Scott Tolinski?

  Yes. Cool. Well, thanks Mitch. Thanks for sharing that.

• 1:21:05Speaker 00

  Yeah. All right. I got two, last two things to show you. These ones are a little more fun and interesting. And the first one is an app called Organic Maps. So in the realm of being privacy conscious and not letting Google own all my data, I've been searching for a new Maps app. And if you use Android, you typically are just using Maps that's built in, Google Maps. If you're using iOS, you're probably using Apple Maps, but.

• 1:21:32Scott Tolinski?

  No, you aren't. No, who's using Apple Maps? That's.

• 1:21:34Speaker 00

  Do you use Google Maps on iOS?

• 1:21:36Scott Tolinski?

  Absolutely, yes.

• 1:21:37Speaker 00

  Oh, wow. Okay. Okay, but as you could assume, Google is tracking absolutely everything about you when you're in the Maps app. It knows your exact location, it knows the places you've visited, like the number of times, I know you can turn this on or off, but I've got notifications of like, "Oh, I saw you visited this restaurant. Would you like to leave a review?" There's also, like, there are settings to turn that stuff off, but Organic Maps is completely disconnected from all cloud services and allows you to download maps offline. That's the other thing, is you typically might want to use this if you're like hiking, or biking, or maybe you're going to an area that doesn't have cell service. Like technically, Google

• 1:22:18Speaker 00

  Maps allows you to download areas of the maps, but the nice thing about this app is it's all based on Wikimaps data. You can select a city, or an entire country, or a state, and have all of those maps downloaded. And that also includes points of interest.

• 1:22:35Scott Tolinski?

  Yeah.

• 1:22:36Speaker 00

  The offline maps in Google Maps, I believe, is kind of just like streets and stuff like that. But once you've downloaded the maps, you can also search for places. And one of the killer features that I've yet to see in a lot of these free map apps is navigation. Now, I haven't used it yet, the navigation piece, and that's one of the most finicky parts of these apps that are Google Maps. Like, Google Maps has navigation tuned. Like, it, like, the, I remember using, like, the old, like, physical GPS devices and trying to figure out when to turn. Like, is my turn now, or is my turn 10 feet from? Google Maps, in my experience, has that dialed. Like, I've, I can now gauge where I need to turn. So that's the one thing I haven't, I haven't tried yet. But just putting

• 1:23:21Speaker 00

  this on your radar because I also like to go hiking or camping. Sometimes I don't have service, and so this app is really great. And I've been using it on Android. It's also supported on iOS and all the other places, so.

• 1:23:34Scott Tolinski?

  Yeah. You know, I used offline maps in Google Maps quite a bit when I was in Greece, and they did have places of interest, as far as I know. At least, maybe I was connected or something without knowing it, but there were times where I was getting at least that bit of stuff still working. So offline maps, man, it's a low-key feature that I feel like that's really great that that, like, if I were to use something like this, I would Upset if it didn't have offline map support or navigation support.

• 1:24:05Speaker 00

  Yeah.

• 1:24:05Scott Tolinski?

  Oh, my gosh. Do I have a sandwich delivery?

• 1:24:07Speaker 00

  Sandwich delivery.

• 1:24:08Scott Tolinski?

  Thank you, Court. I got some peaches, too.

• 1:24:11Speaker 00

  Oh, peaches. Some nice wholesome bread. It's got like oats and wow.

• 1:24:16Scott Tolinski?

  Oh, yes. It's butter lettuce. We love some butter lettuce.

• 1:24:20Speaker 00

  Nice. All right. I only got one more thing to share and then we can let Scott eat a sandwich. And this one is a site called Starring the Computer. So it's at starringthecomputer.com. And it's a index of a bunch of different types of computers. So like Acer, Alienware, Mac. And it has screenshots of when those computers have appeared in movies. So I thought this was fascinating. Like, for instance, if we go to the MacBook, you can see that even going back to like 2009, there are movies that featured a MacBook. So you see screenshots from the movies using that tech.

• 1:25:00Speaker 00

  So this is a fun little browse. Also, like a trip down memory lane because like some of the older movies you might remember.

• 1:25:08Speaker 00

  And you can see people using that old tech. I do believe it might be movies and TV shows.

• 1:25:13Scott Tolinski?

  Yeah, it says movie and TV. I did see Young Sheldon on.

• 1:25:18Speaker 00

  Do you like that show or what's it?

• 1:25:19Scott Tolinski?

  No, I don't. No, I couldn't catch me dead watching that show. I'm sorry. If you're like Young Sheldon, folks, that's not for me. But I love this. You know what you could even do is you could hook this up to something to pull clips of these. Like, ooh, I want a Honeywell H200. Let me pull that clip from Casino Royale 1967. And then it can find that clip and pull the clip for it. I always just think about clipping like, because I'm deep into this like programmatic video space right now. So not like building programmatic videos, but like, ooh, what could I do to like make finding clips easier?

• 1:25:57Speaker 00

  And I think like maybe even Randy could use this because we've done in the past like reacting to hacker clips. But this site is like a massive index of computers being used. So I'm sure it has. Hacking movies that we could react to.

• 1:26:10Scott Tolinski?

  Yeah, Jackie Chan has a GE 635 inside of his Subaru in Cannonball Run 1981. I did not know that.

• 1:26:18Speaker 00

  Nice.

• 1:26:18Scott Tolinski?

  I would have just guessed that was a fake ass computer, so shout out to Jackie Chan and Cannonball Run. This is cool. This is very old web, CJ.

• 1:26:27Speaker 00

  Yeah, yeah, but I found it on Hacker News. I love it. I'll be perusing that quite a bit.

• 1:26:32Scott Tolinski?

  Yeah, I feel like I just went back in time.

• 1:26:35Speaker 00

  Farinda, who just donated 500 Kenyan shillings. So, thank you so much. Thanks for tuning in. Thanks for the donation.

• 1:26:43Scott Tolinski?

  Yeah, thank you so much.

• 1:26:44Speaker 00

  I'm guessing you're in Africa and Kenya. Amazing.

• 1:26:47Scott Tolinski?

  Sick. Cool. I don't really have anything else, CJ.

• 1:26:52Speaker 00

  That's all I got.

• 1:26:53Scott Tolinski?

  And I'm not just saying that because I got a sandwich here.

• 1:26:57Scott Tolinski?

  Henri asked if I was a teenager in 1981. No, I was not a teenager in 1981. I was a teenager in the 90s. I was a 90s kid, so, yes.

• 1:27:08Speaker 00

  Nice. Yeah, Donnie says every time CJ mentions Hacker News, we drink. I did mention it quite a bit today.

• 1:27:13Scott Tolinski?

  CJ. See, we gotta have some digital, I don't know, digital frowny faces anytime Hacker News gets mentioned.

• 1:27:25Scott Tolinski?

  Yes. All right. Well, I don't have too much else. Ready to get out of here, then?

• 1:27:30Speaker 00

  Let's do it.

• 1:27:31Scott Tolinski?

  All right. Well, thank you so much, everybody, for checking us out. We will catch you next week for another Syntax Weekly. And who knows, maybe we'll have a giant sun-sized AI model to talk about. Who knows? Fingers crossed.

• 1:27:46Speaker 00

  Cool. Peace.

• 1:27:48Scott Tolinski?

  Peace.