We Accidentally Deleted Programming  Welcome to syntax. Welcome to a brand new episode of the frontend happy hour podcast. Welcome to this week's JS Party >> live from Ship Shape Studios. This is Whiskey Web and Whatnot with your hosts Robbie the Wagner and me, Charles William Carpenter III. >> That's right, Charles. We drink whiskey and talk about web development. I mean, it's all in the name. It's not that deep. >> This is whiskey web and whatnot. Do not adjust your set. >> As more and more code is generated, developers end up spending more time reading and reviewing code. This creates an issue on reliably shipping quality code. So even though AI is promising, a lack of proper guardrails can slow down teams. Code Rabbit is an AI code reviewer. Unlike AI code generators like Cursor, Windsurf or GitHub Copilot, it reasons and submits feedback on your codebase while you raise a pull request to merge code into your VCS. Code Rabbit is a self-improving agent and it learns from your interactions and creates learnings. Code Rabbit generates one-click commit suggestions for some review comments, making it easy for developers to address review comments. Code Rabbit is used by 100,000 open- source projects like B, Nex.js, and it's installed on 2 million repositories so far. It's also performed more than 15 million PR reviews. >> Hey, what's up everybody? Welcome to Whiskey Web and Whatnot with your hosts Robbie the Wagner and Adam Thomas Argyle the Nerd. What's going on? >> What's up y'all? I'm feeling good. How are you? >> I am good as well. It is once again Margarita Thursday, which I just call it this to make my wife upset because it is clearly either thirsty Thursday or Margarita Monday, but I like to do the dumb one that doesn't make any sense. Oh, nice. We do uh Woody Wednesdays. We go to Little Woodies and then we used to do like Fridays. Well, that was really stretchy, you know. >> Now it Pokemon. That one always stayed good. It's not a Friday, but Friday doesn't start with P. It's okay. >> Yeah. Yeah, these things are hard. All right, so I'm excited for the whiskey we have today. This is a special edition. Where's my camera at? Whistle Pig Liquid Death collaboration. I'm a big Liquid Death fan. I really like their uh Dr. Pepper flavor. I forget what they call it. Dr. Death, maybe. >> I should try that. I love Dr. Pepper. >> Yeah. So, is this is a wheat whiskey. It is 85% wheat and 15% barley. They use liquid death mountain water to proof it down, which like hopefully they got a discount on it because liquid death is not cheap. Can even they're like popping all the cans and pouring them in there. Um, yeah, it's 86 proof. I don't know what the age is, but like you get the idea. Liquid death collaboration. Looks cool. Smells sweet. Smells fruity. Yummy. Okay. Yeah. I can never tell. Like sometimes there's a little funk and I can't tell if like my glass is dirty or >> it's got baby stuff in there. Yeah. >> Yeah. >> Shouldn't have mixed similac in my uh nice norling glass. Why' I do that? >> Yeah, I do. I feel like I smell strawberry. Strawberry cinnamon perhaps. I'm going to taste it. >> Oh man, I'm getting like a berry medley. It's nice. Really light spice. Smooth and soft. That is yummy. >> Uh coating of my tongue from the front to the back. So I got like the spice and then just the the saturation as it spreads across. Yeah, this is very good. I guess maybe some of the similar notes that I like from rise maybe come through in this cuz it's all like uh grains and I guess is corn a grain? No, I guess >> it is not. >> Yeah. So like it doesn't have corn. I don't know. >> Which I feel like makes it uh tastier like taking the corn out. I'm glad there's still some spice. Always like a little spicy bite. And then uh it's pretty low on the proof. Do you do you pay for proof? Like do are you like, "Oh, I'm if I'm going to spend money, it better be high proof." Do you do you care about that? Is that what >> No. No. I just care about >> Yeah. >> Yeah. I care about interesting things and it being like if the distiller wants me to try it at 135 proof or whatever, like, okay, like I'm down, but I'm not seeking that out. I I usually like it to be a little higher than like uh 80 proof or whatever. Like whatever the standard is. So this is a little bit higher, I think, right? It was Yeah. 86. So it's got a little more burn. Um yeah, I think it's very balanced. Like doesn't burn too much. Nice spice, fruity notes. Pretty good. >> Pretty good. I'm happy. So yeah. What What's your rating? >> Well, I don't know what the scale. Do you do you remember what our scale is? >> Zero to eight. If it's zero, you're like, throw that [ \_\_ ] away. If it's an eight, you're like, "Clear the shelves. I want this forever. Just put it in my belly." Uh, and then, you know, anything in the middle. Um, yeah. What you got? >> Um, yeah. I don't have anything bad to say about it. Uh, but it also is maybe not like my favorite one ever. It's so hard. We've had so many. I'm just going to say flat seven. >> Flat seven. Solid. I was thinking six, six and a half cuz Yeah, it's uh solid. I like it. I will have it again. I do like that it's kind of soft and sweet. It has it. Maybe that's the low. It's the low proof is probably the softness. >> And the mountain water. >> And the mountain water. Yeah, that that matters. Oh. Oh, I was reading about this, too. Um, it was uh aged in a coffin. >> Wait, was it really? >> It was really So, they started out in an American oak like regular barrel, but they move it from a barrel into a coffin. >> Uh, which is why it has the uh coffin symbol on it, and it's called gravetock. So that's why it's called gravetock is there's an American oak coffin. >> Oh yeah. >> That apparent and they're like teasing that it's >> final months resting in peace in a 380 gallon charred American oak casket >> then was proofed with liquid death mountain water. Wow. Okay. >> Love it. I like that dedication to detail. >> Yeah. Money to burn. >> Oh, dude. Speaking of like death and and American wood, here, check out this banjo I bought myself for my 40th birthday here. >> It's already beat up a zombie. >> Nice. Love it. >> Covered in blood. It's covered in little zombie icons. Uh there's like a >> knife through his head there. There's someone's foot. Uh it's called the zombie killer. And it is very cool. Here, let's see. Oh, I don't ever play this way, so it's weird. I'm going to try to play Mario. Nope. I'm not going to be able to do it standing here like that. >> Hey, it sounds good. >> It does sound good. It's great. >> Having only heard you play banjo like lowfi trying to throw it up against the mic, I feel like this one sounds better than the other one you had. >> Cool. I'm hoping this is my heirloom. Like, dad's dead, but don't worry, we've got his zombie killer uh banjo, so he'll live on forever with that sick banjo. It'd be cool if they like reinforced it where the like splatters are so you could like actually hit it on stuff and it'd be fine. >> That would be cool. Did you see the saw that was around the edge, too? So, the ring, there's like a tone ring, but around the ring is a saw. Like straight up edges of a saw. It's super cool. >> That sounds dangerous. >> Yeah, I cut my finger. Ah, I should put some of my real blood on there just cuz if it's a heirloom. >> Oh, man. I used to bleed on my guitar so much when I would play shows because I would like hit hit it wrong and catch my like nail or something and just like >> Oh, yeah. >> be just pouring blood all over the guitar for like 10 minutes. >> Cool. >> Yeah. >> Yeah. Yeah. We have to find some recordings of of you uh in some bands cuz I I was saying we should do I don't know if we get Chuck or somebody else to do it, but we need people to listen to like a little bit of band I was in and a little bit of band you were in. See if they can say who was who. I think that'd be a fun game. I mean, I feel like it's going to be kind of obvious, but I don't know. Maybe people don't know the past emo me and they'll be like, "Who's this young kid singing?" >> Yeah. Yeah. I mean, if you if you don't know like which uh part of the band you were, I guess is like maybe >> Oh, so you didn't sing. Okay. Okay. That's cool. >> Yeah. Well, I did do some singing. Um but like Yeah. So, we'll see. We'll see. We got to uh finish the developer band as well sometime. >> I want to >> so many things on the stinking to-do list. It drives me nuts. My to-do list is so big. There's like stuff on there. I'm like, I would like to get to that. And I'm like, yeah, but there's stuff I haven't done for the past 3 days that I need to do. So, >> yeah, >> I'll be an adult, I guess. Um >> Yep. That's what I unfortunately pick on the daily, but uh maybe one day we won't have to pick that. I like the uh the irony that I I texted you too. It's like we need to spend some time automating AI to do some of our tasks so that we can have more time. And I'm like wait that's a chicken and egg. I'm like if I don't have the time I can't create the automations to save me the time. I'm like >> agents are great for that cuz like they might not get it for me every time but I can at least be like hey I'd like this and then come back later and if it happened to do it then I've got it. Like um I did that. Tell me about Claude Co. Yeah. Did you try that? >> I have not. But uh before we jump into that, I just wanted to say like uh I have I've automated one thing now and it's not 100% automated, but I made um this thing. I was like, "Hey, uh I think I used warp on this one, not open code because open code had the thing where you can't use your claw tokens anymore." >> Yeah. >> Um >> rugpool. Rugpull. Yeah. >> Yeah. Um but I was using uh warp and I was like, "Hey, um build me this thing. I It can be swift because like it's just going to be me that uses it. So it can be Mac only. I don't care. Um, I want to put in a link to a whiskey.fm episode and I want you to pull the title, the show notes, and the transcripts. Pull like five to 10 topics. Then take those topics and make them into like concrete things you could make an image of. Not just like React hooks like like something that's like uh a React hook doing like I don't know like make it like a thing that could be drawn then give it all back and like give me an image that I can use as the art for the episode cuz I'm trying to make like I spend too much time on those for no reason. So I'm like >> they're super cool dude. It's not no reason. I love those things. every single one of them. I'm like, "Dude, the TLC in this is awesome." >> Yeah, but like I don't have the time for that. So, they might get less good. They'll they'll probably have less people in them and be more topic based. Um cuz I used to like try to describe the person that was in the episode and have it like draw what they would look like or whatever. >> Um but now I'm just going to be like, "Hey, what do we talk about?" Like, boom, we got it. Like better than having no art, for sure. So, I'm trying that out. Working okay so far. >> Cool. I've used uh Warp to create an automation for me on my personal blog so that every time I drop an image into the image folder, it sends it through my I have like two phases of processing that I do as well as generate the front matter for me. So that was like a really cool one to give to AI because the ins and outs are so clear. Um >> and then I saw something this last week too that I was very impressed by which is someone was um making a whole website a a data driven entire website. So, let's say you've got uh let's just pick music. You want to do a website for the top 20 songs of 2026, whatever. >> Okay? >> So, you go get the data. You stick it in uh you start a markdown file and you're basically you build out this pretty large prompt. It's almost like a big PRD, but in the middle of the PRD, this person stuck a API key to Korea and to um Gemini Nano Banana and said, "Hey, I I have no media for this site. Use these keys and create images that are engaging and um you know, help with the site or whatever." And then they took this humongous prompt which uh and yeah generated images for the whole thing which was anyway I thought that was cool but they pitted it against a Ralph loop um versus not a Ralph loop and the Ralph loop um was arguably not as good although it was really splitting. Some people were like the Ralph loop one is better and I was like I kind of like the first one better. They were both really good though. The PRD was just so well written with all these little details like I'm saying about like the image uh generation and these API keys. I was very impressed. Also very intimidated. I was like, "Oh geez, this site looks awesome and it got made for this person within minutes." I'm like, "Crap." >> Yeah. Yeah. I mean, I think sites everything is doing better and better at. I think images are still kind of hit or miss. Like it especially if you tell it not to do a thing because if you've said the words, if you're like, "Do not put facial hair on this person." is like, "Well, you said the words facial hair, so I'm going to put so much facial hair." Like, you can't tell it not to. You have to describe what you want and just omit the things you don't want or it will freak out. Um, but yeah, the next thing I want to build is like >> um we need a thumbnail generator. So, there's one built in to Riverside, which I've been trying um for like YouTube thumbnails to be like like you know, like you need just the random guy with the weird face like and a title. Um, and so it's been doing those and it won't change the text. I even asked it. I was like, "Can you like make the text not just the first word highlighted in yellow and then the rest black and like give me different fonts and do different stuff?" It's like, "Oh, absolutely. I have full control over that." And I was like, "Cool. Can you like change it to this?" And it was like, "Cool. I did it." And it was still the same thing. I was like, "Clearly someone has baked this into your like base knowledge somewhere." >> Hey, that's that AI harness stuff. We've talked about that. It was like Grock. I was, "Hey, Grock, I want you to make me a meme.me. make me a spicy meme that pisses off developers, you know, like uh I know you know what makes people mad and is is spicy. Do it. And it um it was like hooks. It had some comment about hooks and then um it couldn't even spell react right uh in the in the image it made. I'm like, "Ah, you're so stupid. This is bad." >> Yeah. Yeah. Anything like that there isn't a ton of source material on it has trouble with, too. Like what I did a I had a tweet series for like I don't know if I was trying to do one a day or one a week or whatever it was. It was unsustainable. I was trying to do like dad jokes about programming and I was like all right AI I've got like 10 pretty good ones here like can you understand like how I did these and like try to use the same sort of patterns pattern matcher. Yeah. >> Yeah. And it was like Yeah. Yeah. Yeah. And it gave me like a bunch more and they were all awful. So, I was like, "Oh, god." All right. There's just not enough people making dad jokes about programming, I feel like. So, >> oh jeez. Um, I tried to give a layup to So, this is Yeah, I like to describe this as giving a layup to AI. You're like, "Hey, AI, I'm spelling out everything. You You're smart, right? Use your intelligence to learn from what I just taught you." And And it's like it never works. >> Um, so one of the things I have on the agenda here to talk about is uh AIM. Um, which is just a trendy acronym. It's not even a What's your aim handle? Um I'm trying to Well, I succeeded. I Do you know what flip is? Um from that was made by Paul Lewis. The acronym first last invert play. Have you heard this? >> I don't think so. >> It was a way for you in JavaScript to take an element that needs to it's like in line in the page and now you're going to make it fixed position like over here. Uh, so if you're listening everybody, I I'm holding my hands up in two different places and one hand is closed and one hand is open. All right. So anyway, there's like an element needs to go into two states. Uh, and they're both >> Tell me about sprites. Is one sprite entering? >> Yeah. Oh, sprites. >> No, I'm sorry to derail you. Keep going. >> Oh, it's all good. Okay, so you got element number one and you you like click it and it needs to morph to the position uh in position number two. And it's not two different elements. it's the same element needing to exist in two different layouts and you want to animate it between these things and you want it to be a CSS transition so that it's like interruptible and stuff like that. Okay, so Paul Lewis comes up with this uh awesome simple way to do it called flip. And the F in flip is for first you take the element where it is in the closed state um at the beginning and you take a it's not a picture but you get its bounding um qualities. You get it's like what's its top and its left? What's its height and its width? What are all these different properties about? What's its background color? And then you do the L, which is last. you actually put it into the new layout instantly um by just setting the properties and causing reflow. So then it goes into that end position. You take your uh measurements again and then you invert it which means you put it back to the first position and now that you know the first and then that end position you can interpolate between the two states and you get this really nice morph effect with JavaScript and a CSS transition. David Corshe who we just had on a show recently has done a ton of flip. He wrote an article on CSS tree. Flip is a very cool technique for very dynamic, meaningful morph transitions. And I last week figured out how to do it all in CSS using anchor positioning, uh, interpolate size, and oh crap, there's one more feature. Oh, starting style. >> And so I did this. >> Yeah. Well, I called it AM stands for anchor interpolated morph because the okay, so if you think about it, it's like a the first part was uh finding out the measurements of the element, its height and its width and its position. Anchor can do that. You just set that first element, you're like, "Hey, you're an anchor uh element." And so now anything else can like know the positions and bounds of that of that element. And then you can take the next element which is just in its natural resting place uh and then say starting style start your styles from where this ele other element exists. So I'm able to make these elements morph from any element in anywhere on the screen uh using CSS because I can use anchor to measure where they are. They know where they are naturally and then they can like morph from that position. Anyway, it's really cool. So I'm like figuring out how to do flip with CSS uh and anchor. So Anchor is really cool and anchor just went baseline in Firefox. So, >> I was just about to ask, is it supported everywhere yet? >> Uh, I mean, it's not like we can use it. It's so funny. All these cool things come out. I'm like, at work, I can't use it. Even even when it's baseline at work, I'm like, I it's I have to support iOS. >> Where is it? >> Oh, cuz because of older stuff. So, I don't >> Yeah, for me it's like I support it everywhere. I don't give a [ \_\_ ] about the older stuff. Like, if if the newest browser that's not like a Canary supports it, ship that. Like, >> nice. Well, so you're Yeah, big fan of baseline. My site kind of does that. Well, my site will early adopt all sorts of things. I'm willing to manage two versions or even just have a lame version and then a dope one. I'm like that's that's cool with me. But at work we uh we can't do that. It's like it can't be broken. >> Yeah. >> Uh >> yeah, work is different. I'm talking about my personal stuff. I even did like brley before it worked and like everything like Chrome was the only thing I think and I was like hey it's way smaller going to ship it. Like >> that's awesome. >> Yeah cuz like there's not that many people coming to my personal site or whatever. So, I'm like the people that are are probably developers on like the latest Chrome. Like, it doesn't doesn't matter. >> Um, >> but yeah, I'm excited for the anchor stuff because I've been wanting to use that. Um, >> like you can't do everything just yet. You're going to need like some polyfills or extra code or whatever to support like we're using a floating UI um in Shephard to do like basically what anchor stuff does. Um, >> yeah. And I would like to replace that because it is big and it is like the biggest part of the uh the library as of right now. >> It's also badass but yeah. >> Oh it's great. Yeah. No shade on them. I just like you know in the chase or not chase race to the smallest libraries. Um I'm trying to do that. And also I might replace spelt with solid because okay spelt five for whatever reason and Rich is not helping me with this. So, Rich, if you hear this, uh, maybe we can make it better, but like it it increased the bundle size by like 10 kilobytes or something. Um, >> like it went from like 17 to like 28 or like I guess >> four to five it went up. >> Yeah. Mhm. And so I think what he was saying is like the the wayfelt 5 works is like the base part like uh I don't go I can't think of words. the like uh the main part you're shipping like the little what do you call that the core the core the core is like bigger um but then each component is smaller and since we have like five components we don't benefit from that because like >> it's like if you had a like a thousand component app then you'd be like oh my god it's so much smaller because like each component is smaller but we have five components and so like it's not smaller because the core is bigger which I think is what we're hitting um we are also still using the like legacy don't use this classes thing because like >> because it's uh it won't let you send a reference to an element back out still so I need some magic from someone who knows how to do that to fix that um but I was like yo AI um this is pretty well tested you want to convert it to solid for me and it was like dope and it did and it was back to like the size of spelt 4 and I was like hm could ship like so um I'm interested but I I don't want to like I like philosophically like spelt so I don't want to like move away from spelt because just because like I want to make for sure that it's like what I have to do before I do it. >> Yeah, good call. You know the it kind of goes into one of our other topics. It's like uh you want to make the smallest library. How about your library doesn't even ship code? >> Yeah. See, did you see this post I put in the Discord as article somebody wrote? >> No. Where they How would if it doesn't ship code, what does it do? >> Well, let's let's talk about that. They ship the specs and um all the expectations. >> So, it's basically shipping instructions for an LLM. So, here's the thing. this person shipped a library that basically does relative dates, you know, like 5 seconds ago or 20 20 days ago and it's all localized. It does all it does all the things like a huge date functions library would do. However, it ships no code. It just ships the the great spec and instructions and expectations so that an LLM can do it in any language. And so this person's like, I made a library that works across Elixir and Ruby and uh Rust and whatever because it's not uh it didn't prescribe to a language. It just is a spec. And I thought this was super interesting because I'm I'm very prospect driven development. >> Yeah. >> And this was the first time I saw somebody kind of do it at a micro level where they're like, "Oh yeah, I'll just ship all this stuff as a light." And I'm like, "This is kind of awesome." Um, the only thing that's like there are downsides to it. Like you can't really log bugs against the library, which also maybe that's an upside. I'm like, "Oh, no one can log bugs against me. That [ \_\_ ] rules." >> Well, you still could. I mean, it's not a like, you know, uh, can't read whatever of undefined bug, but it's like, >> yeah, >> um, you know, the spec is maybe a little wrong in this spot. Like, I want to reward it like >> add some tests for this case that the AI keeps screwing up. Yeah. >> Yeah. So, I think I think that's where we're going because like >> um people are already like not even interviewing with code at all. They're just like, "Hey, show me how you can prompt and like let's see how productive you can be. I don't care how you do it." Um >> Jake, stop. >> Yep. Shopify is revamping and updating their interview stuff. What's up, dude? >> He's chewing up a box. He says, "I know you're up here ignoring me." I can smell the dog breath through the camera, bro. >> Yeah. Yeah. Sorry. You were saying >> uh something about >> uh Shopify is updating their their interview stuff, too. And I was asking um and and I had and I have my own answers to this, but like what if someone shows up to the interview and they just uh Ralph Wiggum Loop the whole question or what if they take it uh give it to We were talking last week about Gas Town. They just give it to Gas Town. They're like, "Hey, mayor of Gas Town. Here's a interview. Why don't you just solve it?" And then they just sit back and they twiddle their thumbs. So, hey, what do you want to talk about while uh all my agents do all this interview? You know, like, you know, do we reward that person? Do we not reward the person? So, there's all these different scenarios now that we're It all depends. Yeah. So, basically, we don't want anybody showing up that just prompts the whole time and can't explain anything that's going on. It's a bad sign. >> Or do you though? So, I mean, this this is where I think we're going is like there's going to be some percentage of people. Hold on. Okay, hold on. I'm going to shut Jake out. One sec. All right, edit that out. Somebody editing this later. Okay, so there's going to be some percentage of people who are just like innately uh sants at prompting, right? So, like if they're for some reason just really good at controlling all AI and they get really good results that work really well every time, do you care if that person understands anything about the code? If they're shipping insane amounts of really good stuff that people who do know about code go, "Wow, that's good stuff." Like, I would say no. >> You're right. I hate that. But >> it seems very fair in terms of like they're producing the thing we want. The output is desirable. Well, they can't explain it, but who cares? They can't explain why their products are good or Yeah. >> Like if you make good art, you're probably like, "Hey, I don't I don't know how I do it. I just go like and it's good." Like >> some But then again, some pe some artists who are like classically trained are like, "Oh, I know all of the different strokes and I like do it this way and like you know, like >> I think they're in their own colors, right?" Yeah. >> Yeah. There's room for both. And I think it's not as much of a science as we were led to believe. There's a lot of art to coding and making an especially in web development like in a uh you know purely backend runs on your like computer app or something there's not as much art to that because it's more just like input output but like there's a lot of gray area and stuff that has any kind of UI. >> Yeah. I mean this is all very prospec too because then the co because the co we're just moving so quickly towards the code matters less and less which also comes in another one of the talking points which I think we've briefly done before which is uh does do types even matter does having a database matter because if a database is a layer of indirection for all your data and AI doesn't give a crap about that AI can just build a site off of local files and it can write to files and then you have no dependencies like how much and where is this going to end where code is is completely irrelevant uh in favor of >> yeah prompting and spec and spec authoring. I even think the spec side is even cooler. Like Toby at Shopify is has a lot of great um like ideals and guidelines and just like things that guide him in general. And so you could kind of take a lot of that stuff, codify it into like the heart of the specs of your company and then have all these additional people go build on top of those and elaborate and then your whole kind of system is a manifestation of your ethos and your your services uh kind of combined. And then you combine that with brand and like a design system and some other things and you're really kind of reaching a new level of um just producing software where the code doesn't matter. it matters more about the the core that you were talking about earlier isn't the core of the framework. It's like the core of like why why are we doing stuff >> uh becomes more important than what are we making and that >> I've I've always liked the why over what. Um so I'm Anyway, this is Dude, it's wild. >> Can it slow down please? I'm I'm exhausted. >> Yeah, I wish it would slow down. Um, this we were talking about this at work is why I put this uh this bullet point on here because we're just now in like a a nine or 10year-old codebase adding some TypeScript in which is always hard in an old codebase with lots of Croft and uh we were basically kind of having a philosophical discussion around like what do we want to recommend as the way to do TypeScript right because you can do it a few ways you can have like one where it's like oh yeah I want everything to be super strict Like if you type a file, I don't want to ever see an any. I want everything to be perfect. I don't want to ever use a type from that. That's wrong. There's another philosophy that's like, well, I don't care about most of this cuz it's pretty trivial, but this one thing is really hairy. I want to like type that and then I want the rest to be any. I don't care. Um, and we were just kind of talking our way through like what do we want to do? How do we approach converting the codebase? And then we were kind of like, wait, so you've got, you know, thousands and thousands of files. It could take years to convert. Maybe you can do a code mod or some AI stuff to like make it pretty quick. I don't know. We're, you know, in the factf finding phase right now, but we're like, but what is the point anyway? Because if we're going to like like the as an alternative to converting anything at all to TypeScript, what if you spent a lot of time investing in your guardrails for AI and you were like, "Hey, everyone on like all these other teams who is maybe not as senior or doesn't know the full code base or whatever, like use this when you're writing code and it will keep you from doing dumb stuff that like TypeScript would also have saved you from, but like we don't care about TypeScript anymore. We just want you to like write everything through this. It'll check and make sure like you're not doing that. And then TypeScript doesn't need to exist except for if you fully typed it all too then AI could go way faster because it could be sure about being like oh I know all these types are for sure this way. So like there's still some benefits but it was kind of like we're probably never going to do that because why would we spend a year of like a developer's time when AI can just kind of save us? >> Oh man, I'm in I'm in a code base all day right now. Over a million lines of TypeScript, no Annies. Uh, everything's typed. >> I love it. Jealous. >> Um, >> well, I I mean, there's just there's just so many trade-offs. Like, okay, first off, AI is not wildly empowered by this. It's it's equally as hindered as like a as like a human is. And also, I think it might actually be adding overhead. >> Um, because it it churns. It doesn't know. All it knows is like, I'm gonna try to go do this thing. And it goes and does the thing and it's like, oh, there's a lint error. Oh, there's a type error. And then it just like bounces back and forth in there. And then it still can't see. And even if you teach it to see, um, it doesn't Oh, man. We We're trying to turn this thing into like a little human. Um, and I'm not seeing it wildly do great things inside of a big codebase. And we just Okay, so here's a good example. We just saw uh codeex 5.2 come out. Uh, chatgbt codeex, right? Yeah. That's open AI codeex anyway. And its whole thing is they're trying to pitch, hey, if you got a big ass codebase and you finding that AI models are really shitty at going through it and having longunning ideas, we just made the AI model for you. It's called Codeex 5.2 and it's specialized for longunning [ \_\_ ] And it's like >> I don't think it's actually going to do that much better, man. Like it's kind of like there's so much code and there's so many edge cases and it's like CSS. CSS constantly shows up and it's all like it's all like my job's done. You're like, "Did you look at it?" And it's like, "Yeah, I took a picture. Look at the picture. Isn't it pretty?" And you're like, "It, no, it's vomit, dude. Like, did you even know?" And it's like, >> and then the other thing, too, is like you're trying to do code mods or migrations or whatever it is. Like, it can take a picture of the beginning state and it'll take a picture of the end state happily and be like, "I took pictures." And you're like, "Yeah, but what >> do they look the same?" >> Yeah. Well, a lot of times they're not supposed to. So, it's like how do you how do you describe what's supposed to change in the picture? That's the follow-up one. If it's a layout change based on like the uh >> Oh god, my brain doesn't work. Like the visual diffusion or whatever, wherever it like, you know, makes it all noise and then the picture. >> Yeah. So, like if we can figure out how to tell that like this part of the noise is now blue, it could be like, "Ah, yes, I've I've made the noise a noisy blue right there and it is now blue." Um, I don't know. I feel like there's a way to do it. >> We got comments in here from Mr. Ray Dev. He says, "Codeex is crazy, worse than even CL." So, Mr. Ray Dev was like a long running, let me try it. And then all of a sudden, it's not. And I'm just not surprised by that cuz um I I want them to be good. And I think I talked about it last week too where I've been rolling forward a skill. So that way when I spin up agents that um you know like I have one little agent does this task and it helps evolve a skill and that way it passes information forward and I don't have to keep a database. I don't have to keep beads. I don't have to keep git. I don't have to have gas town. It's like I just have this one migration. I don't need one code mod. I'm not trying to do this to my whole and it still um it just screws up enough times and it also hiccups on types constantly. So that's why I'm looking at TypeScript going, what are you really doing right now other than perceptively slowing us down? Um, >> yeah. Yeah. I mean, I'm I think using it as a llinter and not trying to be super strongly typed and using it like sparingly, I think is the way I would recommend using it now. Like if it's a project that just I'm working on, I'm going to I'm a completionist with the type, so I'm going to get it all typed. But like I think in terms of you know leading a team just type the like one or two things that you know people are getting wrong a lot or whatever like help them there but other it's just a llinter like you could also write a eslint rule that would like check am I using this thing wrong and like you know I feel like it's like that it just has less steps. So like I think that is the way I'm going to probably be using it going forward versus trying to have a fully typed everything. Um, although I love when things are fully typed, but >> fully typed feels good. I'm not gonna lie, dude. It always That's why Rust is popular. If you can make your Rust app run, you're like, "Hell yeah." >> Per like a kitten. Uh, you know, like there's no mess ups there. I feel like AI though, like h just it once you're past like four files, it just starts to go downhill. And even when they run these things overnights and weekends, like people are just spending tokens. And I'm not seeing amazing results yet. And so here's the thing though is like I'm assuming it's going to get there. And so um I'm going to keep tinkering. I'm going to keep trying to see how these tools work. Um because they're not done growing every week. It's just wild. So we'll see where it all goes. >> Yeah, it's been pretty good for me. I've been So at work we have an exception to use co-pilot and only co-pilot basically. Um >> sucks to be you. >> Well, however, so we have a loophole now. This is new development as of today. Um, GitHub tweeted like open code is Jared tweeted >> fully supported and I was like, hm. So I was like, okay, is this a loophole where we can use open code now? And the principal on my team was like, >> yes. And I was like, hell yeah. All right, we're >> you got to go install that. Oh my open code. It's uh a really great layer on top of open code. I'm a big fan. >> So I want to come back to that and hear what's in that. But did you see they have a app now? Like it's not just the CLI. You can install an open code app. >> I saw you write that and open code black. So inform me. I am not caught up. >> So open code black is this thing. Um it's like uh the claude code like or cla whatever where they have like the $20 tier and then the max tier that's like $100. It's like their version of that, but it just like spreads the work across whatever model they want at the time. So, it's like Open Code Zen kind of except with a like more bakedin price instead of charging you per token. Um, and so people were signing up for it cuz they would open up like 20 spots at a time and they would sell out like before anyone could click it. >> So now they have just a wait list like you can go buy it and be like we'll charge you whenever we like onboard you finally. Um, >> so I signed up for for it and I I'm gonna give it a try. Um cuz open code I think is the best uh developer experience thus far. So I want to try it with a like I don't have to worry about every couple hours it going hey you got to wait a while to like do anything else. Um and also like we said you can't use it with I was using it with my anthropic like $20 code which no longer works. So you can use anthropic like charge me based on every character you send me. Um but I don't love that. So >> yeah, dude. Let's uh let's compare. Let's wait. Open code black cornite for dead. >> Nice. It's good. >> I'm trying to like So wait, what is it that you like about open code over claude over um uh warp for example? Or even like are you in cursor or are you in windfer and windsurf are dead to me. I don't want those. >> Okay. So you're terminal all day. No idea. >> Are you using work trees? >> No. >> I'm not doing anything super advanced. >> Um, and a lot of my preferences are just vibes. Like was the first one >> was to do the LSP stuff. So it like it was better at TypeScript than everyone else just cuz they connected it to a TypeScript LSP. >> And so that won me over. Like that was the first thing where I like told it to do something and it gave me a thing that I would actually ship. Like I did a lot of AI stuff before and it was like ah this was like an interesting you know debugging exercise but I'm not going to ship your code and this one I was like wow you did a good job. So I I like liked open code for that. Warp has done a good job as well but it seems to like I don't know what it is. Maybe it's because it's kind of like I have it set to autopick the model or whatever. Maybe it's picking bad models. Um but it's like very hit or miss. Like sometimes it does a terrible job. Sometimes it does a great job and it feels kind of like AI used to feel like um so yeah sometimes I've I've been using warp more especially since the claw stuff stopped working in open code um and I've gotten some good stuff out of warp. Warp actually fixed a bug for me today >> with uh in swatch there's a >> um magnifier color picker thing. I don't know if I talked to you about this bug before, but so you got this circle, right? And you move it and the middle pixel like picks that color. And on Windows, it works if you have a just a 100% DPI screen, like there's no scaling factor. But if you I have >> 72 or or what do you mean? >> Like like your your screen resolution. Let's say you have a monitor that's like 10. >> Oh, it's at it's at its natural natural resol. Yeah. Yeah. So then it worked. But if you like I have mine set to 200% because I have like a 5,000 pixel monitor and so if you don't it's like teenytiny and I can't read anything. Um, so at 200% it was not it was like a thousand pixels off like it was clear offset like wasn't factoring in the DPI and uh so I had asked Open Code a bunch of times to fix this like I'd released several versions and it kept being like it's fixed ship it and I shipped it and it didn't change anything and um so Warp was like went down the same path for a minute and was like oh yeah let me fix it like we got to see what we multiply and divide and whatever and then was like wait what if we just make the windows side DPI aware. So we don't have to do any math. We just let it like pick the pixels. I was like, >> why haven't why has no one said that before? Why is that's an option? >> Um so so we did that and it works now. So like warp saved me there. I don't know what model it was using or but like yeah it's it's like I feel like warp is right 50% of the time but does like super magical good stuff when it's right and open code is like uh stuff you could ship more often but is maybe not as magical and fancy and good. I don't know like it's it's more the baseline that I can kind of count on. Um I think there's room for both and I've I've had success with both. >> Yeah. So open code I've been playing with that for a couple weeks. Uh it's now available at Shopify uh as well as cloud code. Both of which I can kind of pick whatever model I want. And so I have a lot of experience with both of those. Oh yeah, we'll talk about oh my open code here in a sec. But I also use cursor a lot. Um cursor has some killer features. Like people are like uh I I used claude today to to spin up three agents to do some work. And I'm like you know in cursor that's just a dropown menu, right? You you write your prompt and you say how many agents do you want to do the work? You say three >> and it creates work trees, bro. So, it's even fancier than that. So, you go like three. >> I haven't used it in a while. When I used it, it was like open-ended text box. I was like, "Can you make this change?" And it was like, "I made it." And I was like, "No, you didn't." It's like, "Yeah, I did." I was like, "No, you didn't." So, I I didn't Maybe I should try it again because it was terrible when I used it last. >> It's also like pretty wildly different. Well, it has two modes. Same thing with anti-gravity. Like, they have a agent modes where you're like, I don't even look at the code. you're almost purely in like a but I don't use those as much but anyway yeah if you if you open up cursor fly out the side menu write your prompt look for the one X that's underneath your prompt change it to three or four and all that does is create agents and work trees for each of those work streams they work you get to watch their little batons come back and then they highlight green you click them and you see their diff and it changes your code as if it would have done before and then you can like either accept it or try on the other one and try on the other one. So every time you click one, it changes your code to their changes. You can see it in your browser if you've got live reload going on. Figure out the one you want, accept the one, delete the work trees, and go back to now a bunch of possibilities and pick the best one. >> Yeah. I mean, okay, wait, and I rephrase like cloud and open code can do that exact same thing. It's just you have to be very explicit and deliberate and you don't get the really rich switching kind of experience >> cuz like open code definitely uses sub agents cuz I've seen it say like I've got a bunch of sub aents doing this stuff but it doesn't let you see >> and pick like >> it's not doing them side by I think it's doing like separate things. It's like four agents doing four different things not four of the same to see which could have the better output. You know what? So I think that's a compelling >> use case because you do get wildly different stuff. It's nondeterministic. So, like if I had four things doing the same thing, I picked the best one. I like that. I might have to try that. >> Yep. There's also um there's so many this this world is crazy right now. Okay. So, the other thing is you have people uh and I've been trying this too where you pit different AI models against each other. So, no longer like right now we're in a very mono model. Oh [ \_\_ ] that was M and M and M. Uh right. You're like monogamous with your model. But a lot of there's people out there that are polyamorous with their models. That's just fun to say. Um, and they uh are basically saying, "Okay, hey Gemini, you go do this task. Hey Claude, you go do this task. Hey, Chad, you go do this task." And then, "Hey, Claude, go review this work that's that came back from Claude." Oh, or whatever. Anyway, you have them review each other. Um, and that's like one of the superpowers of open code is that you can kind of pit them and warp, right? You can open up you can open up command T and warp. You hit command D to split it vertically. Command shift D and now you got four quadrants like you're in T-m. You can have four different agents reviewing other agents work to sort of get yourself into this scenario where you're not trusting a single agent to do everything. That's wild. And so yeah, you combine that with like work trees and these other things. So here's the magic of warp that I still really like is that um I don't have to care about what is an MCP server, what is a skill, and what is a hook and what is a command. I get to tell it what I want and it sort of deres the best thing for me. And my favorite thing is still MCP discovery where I'm like, "Hey, I need this thing to to get done." Uh, like doing that, too. >> It was like I was like, "Do some git stuff." And it was like >> pulling the git kraken mcp to figure out how git works and I was like, "Oh shit." Okay. >> Yeah, that's really cool stuff cuz uh it's really annoying configuring things like it's okay. Let me let me rephrase. Configuring things is very fun. It's also a time waste. So like if something can kind of configure on the fly based on my request, that's really cool. Uh and Warp is the one that constantly does that for me where I'm like I I ask for a thing and I say, "Hey, go verify it." And then it'll be like, "Oh, sure. I'll spin up a browser and all of a sudden Chrome opens up with this special profile and I can see it taking pictures and then it starts to cost me a lot of money because it's like, I took pictures and then I took the picture and I tokenized it and I sent it to the server and I reviewed it and I'm doing it again and again and again." I'm like, "Stop taking pictures right now. You're going to eat up all my credits." >> $20 later. You should use this MCP. >> Crap. Yeah. But yeah, that's uh I was trying to just like tell someone why I like Wind Surf better than Cursor the other day and I just couldn't. It was vibes and I'm like it's annoying. I wish I wish I could tell you. >> I had both and that was the same for me too. It was like um at the base use case of like you know opening a file and just working on some stuff and hitting tab to tab through stuff. It was better. I don't know why they both do the same thing like but it was just I felt better with the like tab autocomplete and at the time the agentic stuff in both was abysmal so it didn't matter. But like the you know dayto-day I'm still writing code. I just want you to autocomplete a couple things worked way better >> in Windsurf for me for whatever reason. >> Okay. So let's talk about co-work and then I want to talk about agency. What do you know about I saw co-workers like >> making files for people and has file system access and I was like nope I am not ready for this thing to like just be reading all of my [ \_\_ ] No thank you. I want to give it to it with a spoon. I'm not ready for automated. I haven't used it, but I just saw somebody tweeted like, "Thank God anthropic is coming for everyone else's jobs now." And I was like, "Yes, yes, that is uh" cuz like I do find myself being like, "All right, I have a, you know, I can't think of an example, but some task I want to do in life. Like, I'd like to uh plan this thing, schedule this thing, uh, you know, research getting tickets to something, doing whatever." Like I would like to be able to do approach that in the same way of just being like hello agentic thing like I want to do this stuff tell me the best way like find it for me. So I think it's compelling to like apply that workflow to the rest of your life but like yeah I I do think you're going to hit a token problem and like if we're using it for everything uh the cost is going to catch up at some point like you can't get used to using it for everything. Um so I don't know. I'm cautiously optimistic. I think it's cool. I want to try it out. But I just I just wanted to mention that it exists. I don't have much else to say about it as of now. >> Me neither. I kind of got scared by it. I was like, I'm not ready for that. You know, the thing that scared me the most though wasn't co-work kind of like taking those jobs. It's that I saw uh AI is now growing crops. It's growing food. Uh and I was like, "Oh, so farmers, maybe we won't have farmers anymore." I was like, "What if AI disrupts the farming industry?" Yikes. >> Yeah. Well, the farming industry is a whole another episode we could talk about. Like I think that whole thing is broken. We should not have big giant farms where like if one company has like a bad growing season, we don't have food. Like the local farm model is what we should be doing, but I'm not I'm going to get off my soap box cuz we're going to run out of time. So, um yeah. What? Tell me about agency or whatever you were saying. >> Okay. Okay. So agency um in in an age uh where u uh intellectual stuff is on tap uh what you need now are people that are adventurers and tinkerers and the people that you can trust to go figure it out. Um, so you need to we're seeing a rise of agency, meaning that we want to give people less of a specific role and trust people to just go around and do good like work for the company. You can focus on this area. Uh, and we're not going to tell you very much what to do. What we want to do is trust you that you're equipped with AI and that you are just constantly doing good. You're you're deploying agents over here to do that. that's improving whatever you're over here improving internal processes. You're making an internal tool like you are free now. You're you're now liberated. Your chains have been broken from your front end role or even your design role. We want to give you agency and and trust you to to just go do good. And I'm like that's really cool. >> So it's cool. But like in that world, and I guess this is the world we're in everywhere, even if you are specialized in one specific area, like how do you ever get a raise or a new position or like if you're just kind of like you're in charge of the vibes of this area of work? It's like, well, how do I show that I did good at the vibes? Like >> you used your agency and you actually changed. It's like >> Yeah. >> Okay. Well, yeah. So, I mean, I guess there's that. Like if you're if you're your own manager in your own department and you're setting your own goals and you're writing RFC's that you comment on yourself and you're like I don't know like I feel like there has to be someone at some level that's deciding like is the criteria you came up with good? Do we agree? Did you do well? Should you get more money like or I don't know. Um maybe all of that is done. Maybe there's no more promotions for anyone ever. I mean, I feel like we're pretty quickly heading towards uh we don't need developers, designers, or any of us anymore. So, that's depressing. >> Yeah. Please give me a severance of the rest of the money that I would have made for the rest of my life, and I'll be happy. Oh, yeah. My my 401k is not not ready to have no job. It's going to be wild if uh if it really does displace as many things as it's kind of positioned to. >> Yeah. No, I'm I'm trying to get in the physical goods game. Like I was talking to my friend who is a local farmer. He has popcorn on the cob that he just like he doesn't take it off the cob. You just get the whole cob. He gives you a bag and you put it in the microwave. I thought you had to take it off the cob to pop it, but you don't. >> And so I was like this is so cool. Like people would love to take a [ \_\_ ] Cobb and put it in their microwave. Yeah. >> Why does no one sell it in this form? So I'm like, "We should do that." Like why are we do that online? >> Yeah. Um and then he starts scheming. He's like, "Oh yeah, you just need 40 acres and you can grow 1 million cobs a year." And like I was like, "Okay." Like I'm I'm not ready for that. But like yeah, I want to get into something like that. Like experiment. Like I think AI can help you iterate on your ideas and you're like have a good website with basically no effort and like throw up some Instagram ads, see where it where it lands. Like I don't know. I I want to get into more of that kind of stuff because I feel like, you know, the developer stuff can't can't last forever. Like I think we'll outpace most hopefully like before we're stay. But >> yeah, I don't know. It's bleak. It is oddly bleak. Um I mean last week was bleak, right? Now we're bleak still. I don't know where >> we always will be. We We still bring the joy though. We're having whiskey. It's okay. >> Hey, I showed a zombie killer. Oh, wait. I want to talk about something that brought me joy this week. It's uh want my MTV.verell.app. >> What? >> It's you can watch MTV. So, do you remember um >> MTV just stopped running? Right. I saw that. Like >> Well, that's what I'm saying. It It doesn't have to stop. You can watch someone's curated everything off of YouTube, all the music videos, put them organized by like era and commercials and put it into a thing. So it's like, did you ever watch Nickreboot.tv? I think it was Nick Reboot.tv. >> You could just go to this website and it was playing stuff and you just tuned in like it was old school TV. There was commercials, you know, skip it, skip it or whatever. Um, and you'd watch Hey Arnold and you'd catch half the episode because you tuned in late, you know, and it was awesome. >> Yeah. uh got shut down for probably reasons. But >> uh this MTV one is awesome in the same way. You can be like, I want to listen to 2000's um music videos and you just go there, you hit a button, and there's like 3,000 >> music videos from the 2000s that'll just randomly play for you and play commercials in between every It's been >> awesome, dude. And I'm finding new music. I'm like, how did I not know that song? That metal song was badass. >> I was always a Fuse listener. Did you ever listen watch Fuse? I didn't, but my spouse was huge in a fuse. Yeah. >> Oh, yeah. Yeah. It was uh like as soon as that came out, I don't know what it was like the alternative MTV, like more rock based. And I just had it playing like all day. Like I thought it was so cool. >> Um but yeah, I'm I mean music videos I feel like are a lost art. Like people are still making them, but it's like I'm I personally don't like to go consume YouTube. Um I'm probably in the minority there, but like I like a thing that just picks it for me, curate it. I don't I don't want to go like click on it. I just want it to be a a feed with commercials and like Yeah, like traditional TV sounds good. >> You're going to love this. It's been really nice for while I'm working. So, I just put it on. Although, it's been it's okay. So, let me rephrase. It's been really nice. That's still the phrase, but also sometimes really distracting cuz I'm like, "That music video is really sexy." I'm like, "Holy [ \_\_ ] that was that I watched that as a 14-year-old. I'm like, goodness." I'm like, "That [ \_\_ ] was hot as hell." Jeez Louise. So anyway, I just from like watching these music videos like that was okay. Holy jeez. Wow. Um but also some of them are just hilariously like simple and awesome. So >> yeah, >> enjoy everybody. We'll put that link in the show notes. >> Yeah, we always end up uh watching music videos at the end of like anytime we have people over and we're hanging out and it's like 1:00 a.m. That's like just throw on music videos. So that'd be cool. Just throw that on. You don't have to even pick the next one. Yeah. Yeah. Good call. So yeah, I uh got sixish minutes left. Um I do want to mention the topic that we didn't get to. Did we get to it? And then I was like, should we talk about it more next time? I don't know. Anyway, the fact that booze and nicotine and like, you know, any sort of like thing that are fun or historically thought of as fun um are like way down. Um >> yeah. >> And it's like people probably still like fun. So, I don't know if it's like a one one thing could be like a health thing. Like maybe people I know there's a lot of different like I get so many ads for like this cocktail alternative it makes you feel good. I'm like feel good how what drugs are in it? Nothing. Okay. It probably doesn't do anything. Like >> so I'm like um I don't know. I'm skeptical about those. I don't know if you've tried any of those. >> The NA party beer. Oh, I keep getting Yeah, I keep getting those like, "Oh, you still want to party but don't want to drink? Here, have this NA beer." >> Beer. Beer is awful. I don't want to drink beer if it doesn't have alcohol in it. That's stupid. Um, so no, but like the like kin euphorics, have you seen those? Like no. >> Um, >> it's it's just like a herbbally I don't know, neutropic, like supposed to be like good for your mind, make you more social, etc. Like you know, the good parts of drinking, but like not the bad, but like I haven't tried them, so I guess I can't [ \_\_ ] on it until I try it. But I am very skeptical. Um, but anyway, people could be doing some of that, could be doing whatever. How could that succeed though? If if actual booze is down, how could the NA alternatives be up? >> I don't know. But yeah, sorry. Anyway, keep going. >> My hypothesis is that everything is down. I think >> everyone is losing their jobs. The entire economy is down. Inflation is up. Recession big time. >> Minnesota's throwing bombs into people's cars as they're trying to exit ICE. Yep. It's nuts. >> Yeah. Like um I saw on the Today show the other day that pizza is down. like 30%. >> Like what's >> Yeah. Yeah. >> That's what I'm saying. It's like if even pizza can't do well, like pretty much an American staple. Like >> um I think everything is down because everyone is like I can't afford to live much less go get a [ \_\_ ] pizza. So like I'm going to make a pizza at home if I want a pizza. Like I think that's what's happening. So I don't think people like I think there could be increased demand for like all this stuff. Again, I'm not sure. I'm just like, you know, spouting whatever. But I that that's how I feel. I I don't think that it's like doom and gloom around the alcohol industry and it's going to fully go away because like you got to replace it with something. People need something to numb their minds. Um >> tea, man. Just a just, you know, an herbal tea. Um I keep wondering about supply and demand. So as soon as I'm like, "Oh, these things are down. So supply is there, demand is down." I'm like, "What? So is everything going to get cheaper?" That would kind of >> Well, that would be a welcome thing. Yeah, cuz the whiskey industry has gotten nuts. This stuff is way too expensive. Like Chuck was telling me about it and uh I don't know the full story, but the the essential facts are basically like uh you know they sell like this whistle pig, let's say they'd sell it for 20 bucks, 30 bucks, whatever, like the original cheap price. And then they were noticing people would buy all of them and resell them on like, you know, I bought thisl liquor.com for like 60 bucks. And they're like, wait, you guys are buying all this cuz you think it's good and just like making it more expensive. We'll just make it more expensive. We'll just we'll cut out the middleman. And so that's why everything has gone up like to whiskey like the cheapest of a like not anything that's not like a Jack Daniels is like if it's a nice whiskey, it's going to be 50 bucks plus like because >> that's what's happened. And it's So, is a dinner for two anywhere? >> It's ridiculous. >> Okay. Yeah, that's a Yeah. So, we've had this debate with one of our friends. It's like he's like, "Okay, if you go to like, I don't know, fast food, McDonald's, Taco Bell, whatever, and you're like, "All right, I've got two adults, couple kids, whatever. The amount of that you'll spend on that, he's saying, is actually sometimes more than you'll spend going to like a sit down restaurant." because like if you go to he has certain ones you go to that like if you get a adult entree your kids eat free or something like that and so it's like the kids aren't a factor in the cost so you got two adult entre at like 20 bucks or whatever and then like um yeah I'm like well what about tip or whatever and he's like no he insists like it doesn't matter what he does it's cheaper to sit at the like sitown restaurant I was like that's [ \_\_ ] up man like that's not the way it should be like fast food has gotten super expensive >> super expensive I've noticed the same thing the Uh there's so we do Woody Wednesdays and one of the reasons we do Little Woodies is because it's um locally it's like a local famous person. It's wildly cheap, really good quality. Their French fries are amazing and I walk out of there and I can feed the whole family for 45 bucks. I pretty much can't do that anywhere >> anywhere. Um >> yeah, >> it's crazy. So, yeah. Well, Seattle is its own beast cuz when I was there, I got Taco Bell delivered and I got uh let's see, like a burrito and two tacos or something like not a lot. And it was like $67 or something like with the all the like cuz they're like, "Hey, you're in Seattle, bro. Like, this is going to be at least $30 just for us to even open the Door Dash app." And then like >> like there's so many extra fees and I was just like, "Holy shit." So, yeah. I mean, stuff is probably just more expensive in general. And in Europe, >> if you buy if you're buying food for just yourself, the order stuff is just stupid. You're going to pay more in fees than the food is. Is >> that's why I always order like two or three meals whenever I get delivery and then you just microwave it. >> Oo, good idea. Dang, that's a it's a life hack right there. >> Yep. Yep. Yeah. And also Costco has Uber Eats and uh Door Dash gift cards. You can get $100 of gift cards for like 75 or 80 bucks something. So, >> how does that make any sense? >> I buy them every time I go. They limit you. You can only get like four per month or something because they don't want you just having infinite like percent off at Door Dash, I guess. I don't know. >> This is so weird. This is so weird. This world is so weird right now. >> Yep. Yep. All right, we are at time. Is there anything that we missed that you want to mention before we end? >> No, man. That was awesome. >> Cool. All right. Thanks everyone for listening. If you liked it, please subscribe. Leave us some ratings. Please actually do. I'm not going to say the same spiel that I always do. Go to Spotify, click our show, click rate, give it five stars. Do the same on Apple. It is very quick. Those of you that are listening that haven't done it, shame on you. I've asked you so many times. All right, we'll catch you guys next time. Yeah, like you've been watching Whiskey Web and Whatnot recorded in front of a live studio audience. What the [ \_\_ ] are you talking about, Chuck? Enjoyed the show? Subscribe. You know people don't pay attention to these, right? Head to whiskey.fun for merch and to join our Discord server. I'm serious. It's like 2% of people who actually click these links. And don't forget to leave us a fivestar review and tell your friends about the show. All right, dude. I'm out of here. Still got it., We Accidentally Deleted Programming, YouTube, 01kh207nwrktmavgyqsv2yqn0r.