Pod Insights Syntax - Tasty Web Development Treats 1010: No one cares anymore? About this episode On this episode, Scott and Wes dig into the messy reality of modern front-end work, from struggling to find skilled devs and navigating team chaos to questioning code quality, testing, and even whether AI is stealing the joy of programming.
Show Notes
00:00 Welcome to Syntax
01:06 The Challenge of Finding Skilled Front-End Developers
05:11 Understanding Design Mode and Its Applications
10:33 Navigating Team Dynamics and Code Quality
12:37 The Importance of Testing Strategies
13:39 Learning and Growing as a Developer
18:11 Consolidating Multiple Animation Libraries
21:16 Draw UI with Code Only
22:38 Avoiding Interview Scams
26:40 Embracing Change in Tech Careers
32:21 Why People Don’t Do Software Updates
41:04 AI Kills my Joy of Programming
49:18 Sick Picks + Shameless Plugs
Sick Picks
Scott:
USB LED Light
Wireless Carplay Adapter
Wes:
Shameless Plugs
Scott:
Phases Podcast
Wes:
Hit us up on Socials!
Syntax: X Instagram Tiktok LinkedIn Threads
Wes: X Instagram Tiktok LinkedIn Threads
Scott: X Instagram Tiktok LinkedIn Threads
Randy: X Instagram YouTube Threads
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Wes Bos? Welcome to Syntax Day. We got a potluck episode. That's where you bring the questions, we bring the answers. Some awesome stuff we're talking about today. What is document. Design mode even for? Did you even know that this was a browser standard? Feeling anxious about the future of your career. Feeling sad that I care so much about the code. Does that even matter anymore? Right? Should I update? As we got a weiner wrote in and said, I'm not updating anything anymore. You know, no more updating. I'm sticking with what I have.
0:31 Wes Bos? Refactoring. Someone's got motion and GSAP. What do you do in that that situation? Right? How should you refactor into one? And finally, I just want my team members to care. My team members genuinely do not care about integration versus end to end and all these things. How do I get them to care so that we can write better code at the end of the day? If you got a question, go to syntax.fm and click the potluck button, submit it to us. If you got a question or you got anything you want us to cover, type it in the box. We'll answer on an upcoming episode.
1:06 Scott Tolinski? All right. So let's get into the first question here. It's from Wei. Why is it so hard to find good front end devs? My company is hiring, but everyone I interview doesn't know the difference between PNG and JPEG? Or what the purpose of a bundler is? It feels like everyone out there just knows a little bit of React and has very little in the way of web fundamentals. My gosh, yeah. Front end is hard. Oftentimes, I feel like front end is looked down upon in a weird way. Whether that is because people don't like CSS or HTML, or they don't feel like it's
1:49 Scott Tolinski? difficult. Well, arguably, I think, I mean, front end has gotten like crazy in terms of breadth of things you have to understand, but also you have UX considerations. You have to have accessibility considerations. There's just so much there.
2:02 Wes Bos? Speed, performance, all kinds of stuff.
2:06 Scott Tolinski? And anybody who has been part of this industry, kind of pre the React boom and stuff like that, would know that, you know, you used to work on front-ends and all this stuff, and then, all of a sudden, developers became really in-demand. And what happened? A thousand and one boot camps sprung up, where all they teach you is, here's how to make site with React, okay? We do a create React app, we make things click, we make things open, here's maybe a little bit of CSS, here's to install things. That's the front-end to some of these folks. So, there was a huge boom of boot camp graduates who really didn't ever have that full experience of working on real projects that really matter with
2:48 Scott Tolinski? the team over the course of time, when web fundamentals weren't just something that you, something that would be nice to have. It was like something that you had to have to make anything that worked across browsers and actually was shippable, right? So, Wei, I think it's really just that you need to keep looking. Front-end developers who are skilled are out there, and they want jobs. But at the same time, there's a lot of folks who did just do the boot camp thing and never really got the breadth of experience that you would need to be considered a good, well-rounded, full-stack, front-end dev.
3:24 Wes Bos? Yeah. I also think he's probably not offering enough money. If you're, if the type of candidates you're, you're finding don't know the difference between a PNG and a JPEG, or know what a bundler is, right? That's, there's a wide gamut of, of people in that regard. So, you're either looking in the wrong place, or not offering enough money. Because there are tons of very good devs out there, but...
3:48 Scott Tolinski? Could also be a location thing. I remember the very first front-end job I got, the, my boss, who had interviewed me, was like, "Oh, you were the best app..." It was my first job. And it was like, "You were the best applicant, because you knew about CSS tricks, and you knew about this, and you knew about that. But everybody else I interviewed, like, didn't know any of that stuff, or didn't care, or wasn't looking at blogs, or weren't staying on top of the industry, or whatever." And that was way back in, you know, 2011, so...
4:22 Wes Bos? Real quick, Amsterdam. We are going to be at Amsterdam for a Syntax meetup on June 10. This is going to be a part of the opening party for JS Nation and React Summit. These are awesome conferences that are happening on June 11 and June 12. And you don't have to get a ticket to these conferences to come to the Syntax meetup. You can simply just go to syntax.fm/meetup and grab a free ticket for that. But we highly, highly recommend that you also come to the conferences. Some of the brightest minds in the industry are going to be at this thing, so check it out. JSNation.com, React Summit, you can get a combo ticket. Use coupon code SYNTEX for 15% off. Again, that's syntax.fm/meetup for tickets to the Syntax meetup.
5:05 Wes Bos? And then also you go to JSNation.com and ReactSummit.com to grab tickets to the conference. We'll see you there. Peace. Next question from Humongous, good name. Design mode, it's a weird property I never see used, but seems like it could be useful. Why was it added to the spec? What's the problem? What problem was it solving? Does it serve a purpose today, or is it some vestigial artifact of the old web? So, Design Mode is essentially if you type document dot design mode equals true, it essentially just makes the whole DOM content editable. And every, every six months, somebody figures this out. It's a really good tweet if you want a tweet to go viral. Because it's, it's like, what? The browser has this thing where you can just turn it on?
5:50 Wes Bos? And then when the, the whole document is content editable, literally every text node, you can just click on it and type and delete and whatever. And then under the hood, what it's actually doing is it's, it's modifying, it's adding and removing the HTML and the text and the text nodes. So it's no different than just doing a content editable on the entire HTML document. So like, like, why is it there? The content editable was-
6:17 Scott Tolinski? Are you eating that? I'm sorry, I have to stop you. Are you eating the web here?
6:21 Wes Bos? Content-
6:22 Scott Tolinski? Content editable?
6:23 Wes Bos? Content edit-able.
6:27 Wes Bos? It was added, I think it was added to the browser with the hopes of like, we would have this like GUI, where you can just like, make this really rich text interface, and you can just click into it and type in what you want. And the reality is that like, WYSIWYGs are extremely hard. We've seen many, many WYSIWYGs over the years of like, how do you make something where people can simply just click where they want, and type, and drag and drop, and resize, and all of this stuff? How do you make that, but like, make it like manageable and like, still nice on the backend? And we've seen many people try and fail, and it's not something I ever wish on myself.
7:11 Wes Bos? So I think the idea with adding it was that we would have this beautiful GUI, and I don't think we've really seen that pop up. Anybody who's doing it has to implement a whole extra layer on top of it. But why do we have it?
7:27 Wes Bos? I don't know, it's useful in many cases where if you just want to simply modify a webpage, if you're taking a screenshot, if you want to do something, just pop content editable on the same thing, and it'll rip. Another thing is like, I have like invoices and receipts when people buy a course from me. And what always happens is somebody says, can you add this VAT number to it? Can you change the name of it? I put the old address in it, like, I'm not gonna get it approved. It's all, it had so much just busy work of people saying, my invoice doesn't have the right information on it. So you know what I did? I slapped the content editable on that thing and said, change it yourself.
8:09 Scott Tolinski? Yes.
8:10 Wes Bos? Add whatever you need to do to get your boss to be happy with your invoice, go ahead and add it. And that stopped all those emails from coming in. So there's certainly-
8:20 Scott Tolinski? That was such a good call.
8:22 Wes Bos? -use cases for it, yeah.
8:22 Scott Tolinski? Yeah, I ended up doing that too, especially just because like you said, that was like such a huge, such a huge customer service thing. Like, I need this on my invoice, I need this on my-
8:32 Wes Bos? It's Europeans, man. They love putting numbers on things, and yeah, it has to be exactly-
8:37 Scott Tolinski? Yeah. Yeah, content editable, this is, this is an embarrassing thing for me to admit, but I didn't realize that with like a universal API, I just figured it was like a browser feature or something, not like a API that every browser has. For some reason, I just thought it was like, oh, it's a Chrome feature.
8:57 Wes Bos? Oh, no, no.
8:57 Scott Tolinski? I don't know why I thought that.
8:58 Wes Bos? This is a big part of, of HTML.
9:01 Scott Tolinski? Or design mode, not content editable. I knew content editable.
9:04 Wes Bos? Oh, yeah. It's kind of cool because you can use like, like Command B and Command I to bold and italic. So if you want like a quick and easy way to edit your, some text and save that HTML out, it's great. It's, it's when it starts getting a little bit more complicated and you have like new, like if you hit enter, it adds like spans and then you, oh, there's too much space in there. And before you know it, you've made Notion and you can't move.
9:30 Scott Tolinski? Yes.
9:31 Wes Bos? You can put a line in between two images.
9:36 Scott Tolinski? It's like Microsoft Word. And Notion does feel like Microsoft Word when you're working in it.
9:41 Wes Bos? It is. And like, I don't, I don't envy the people that have to build on, have to build like the, the editing experience inside of Notion because like,
9:50 Scott Tolinski? Yeah.
9:50 Wes Bos? I've never seen anybody do it good. I've never, ever. And like the, the probably like Obsidian is pretty good, but that's so much more stripped down. I, I don't know that there is any good interfaces. As soon as stuff gets complicated and as soon as you have all these different features and all these little edge cases, that must be a nightmare to work on the, the WYSIWYG engine behind Notion.
10:14 Scott Tolinski? Yeah, I know. It's so funny. I'm currently trying to teach my wife Courtney how to use Obsidian and she's getting it, but it is, yeah, it's, it's not as friendly as Notion, even though it feels like it is to me as a developer who's used to working in those types of interfaces forever. Yeah. Yeah. Sick. All right. Next one here is, I want to work with people who genuinely care. My current organization is full of developers who don't care about the platform or code quality. They make no architectural decisions and just keep pushing code that isn't well thought out, well tested, or even looks good. Hey, you, your coworkers might just be an AI because this is sounding a whole hell of a lot like an AI to me.
10:57 Scott Tolinski? Management talks about customer complaints and the need for better code, but the senior engineers don't care either. They write poor code themselves. For context, we're a startup with about 50 developers. Fifty, that's, yes. That's not the worst part. Another senior developer and I actually care about engineering quality, and we've tried to drive a lot of improvements, but they keep getting pushed back. The team always finds faults with our suggestions or comes up with vague reasons not to support us. A lot of opinions I hear don't hold up. For example, they insist on writing end-to-end tests for every small thing we ship. I've explained why integration tests would be more effective
11:39 Scott Tolinski? and that end-to-end tests are slow, costly, flaky, and a lot of the times won't scale. I even shared articles from Kentsey Dodd's, but they won't listen to feedback. The hard part is that the product is gaining real traction and growing fast. I don't want to walk away from a high-growth company, but this people problem is genuinely stressing me out and holding back my growth. I want you to consider something here. I want to work with people who genuinely care. I want you to consider something. You might not be right on everything. I'm just gonna, I'm just gonna throw that out. I'm not saying you aren't, but you might not be right about everything as much as you think you are. Personally, I prefer end-to-end tests for everything.
12:22 Scott Tolinski? Me, I prefer end-to-end tests for everything, and other people might too. Sure, costly, timely, those types of things. Potentially flaky, yes. All of those things, all of the above. But man, I got a lot more value out of end-to-end tests than I do anything else.
12:37 Wes Bos? Can you explain the difference for the audience? The difference between an integration test and an end-to-end test?
12:43 Scott Tolinski? Yeah, an integration test tests exactly what the name suggests. It tests how things integrate with one another. So, systems and how they integrate. Where a unit test tests one single unit, like a small unit, like a function. An integration test is more or less testing the pieces and how they fit together. Where end-to-end tests is testing as a real user using the app. So, many times, especially on the web, end-to-end tests are done with something like Playwright or Cypress, which I haven't heard of Cypress in a while now. I know people probably use Cypress still. Most people have landed on Playwright, but what those are doing is it's writing the test code is...
13:23 Scott Tolinski? Running your app in a browser and clicking around and using it like a user would. So therefore, it's kind of the ultimate integration test. Because it's a whole system integrated, in my mind.
13:35 Wes Bos? It's one step away from having like a guy that just checks it for you.
13:39 Scott Tolinski? Yeah, so that was the red flag for me that was like, ah, maybe, maybe they do have good reasons for things. Now, I genuinely know what it's like to work with coworkers who don't care as much as you, or have pushing poor projects out, or, or in my case, it was oftentimes they don't have the pixel attention to detail that I had. So I would see coworkers push out a client's site, and I would just be like, oh, you didn't even do this? Like, ah, you didn't even check this? And that was like frustrating for me to be in that project, and those were like agency projects. So it's like, okay, those, those, those things are gone the next week, or whatever, I don't have to think about it.
14:23 Scott Tolinski? It's not a big old project that the whole, everyone's going to have to work in. I think the biggest thing that your team needs to get on the page with is a set of standards in hard enforcement. Things like, uh, things that like robots can do to tell you what the problems are, like linting stuff, or, or hard guidelines, things that like block code of a certain type from making it into the repo. But this problem will not ever fix without the management, the engineering management,
15:04 Scott Tolinski? taking control, or even believing that it is a problem. So, one, I want you to consider that other people might have valid opinions. Two, I want you to talk in a non-aggressive way, a non-judgmental way.
15:23 Wes Bos? Yeah.
15:24 Scott Tolinski? To your engineering manager and express some of your concerns. That's tough for a lot of people, especially people with coding egos. I know best, I'm doing it this way, they're not doing it the right way. This isn't a, they're doing it wrong, you're doing it wrong. It's a, hey, I've noticed these things in our project and I have some concerns for these things. Uh, do you think it's worthwhile for us to establish some guideline practices and enforce them?
15:53 Wes Bos? Yeah.
15:54 Scott Tolinski? In some kind of concrete way?
15:56 Wes Bos? Yeah, I think what's happening here is you're, you're probably stomping in being like, we should do things X, Y, and Z way. And they're having a hard time caring. And if you ask yourself, why don't they care? It's probably because fast-growing startup, everybody's got a lot on their plate. Everybody's always busy and they don't want some little weiner coming in and telling them, we should be doing things differently. We have to change how it is because you have to just say, okay, well, now we have to figure out this new way of doing things. And I'm, I'm now I can't ship as much code because I'm trying to figure this out. Now you're breaking things and oh, it doesn't, maybe it doesn't work as well. Um, that's extremely frustrating.
16:39 Wes Bos? So what you need to do is to just simply, like, this is so much easier with AI. It's just like, go spend some time figuring out how you can fix it and then just bring it to them. Bring the problems that you have, be like, hey, customers are complaining about X, Y, and Z. Here are three times when we dropped the ball and it made us lose money or customers or whatever. And then here, here's my, here's my, here's my solution. And this is why it's going to be easier, right? It's gonna, it's not going to be. Harder. You're not gonna have to learn a whole bunch of new stuff. It's simply just going to be easier. It's gonna be faster. It's gonna be better in every regard.
17:19 Wes Bos? And if you can do that, rather than complaining and rather than trying to be a drag on their energy, but you can sort of energize and move them along, I think that that's gonna be the move.
17:30 Scott Tolinski? Yeah, the biggest times that I've ever been annoyed with somebody giving suggestions, valid or not, is when they're giving big wiener, not big wiener energy, I would say. Their energy is weiner-esque.
17:47 Wes Bos? Yeah.
17:48 Scott Tolinski? Yes. That to me is a, is always like, okay, I get it, whatever, you know? Like, it's just a personality thing at this point.
17:56 Wes Bos? That's what it boils down to. Yeah. You're probably not approaching it in a great way. And that is just as much of a skill that you have to learn as figuring out where your semicolons go or whatever other developer skills that you have.
18:09 Scott Tolinski? Yeah. Yes.
18:11 Wes Bos? Laura says, "I inherited a codebase that historically has used motion for encode animations." Motion is really cool, motion.dev.
18:21 Wes Bos? "But recently, it looks like they also added Gsap. Is it reasonable to keep both, or should I make an attempt to consolidate to just one? What should I look at to determine which one may be better?" There are some recent animations that also use 3JS as well, if that's a factor. The 3JS stuff probably is a totally separate bag of worms because...
18:45 Scott Tolinski? Bag of worms, yeah.
18:46 Wes Bos? You know, like, that's a different bag of worms. I would probably leave that out of it. That's probably like more 3D stuff.
18:53 Wes Bos? Although, I don't know, it depends on what kind of animations they are. Maybe they could be converted. I certainly would move them into a single library. It depends on how many and if it's gonna be worth it. But if you've got pages where you're loading both libraries on there...
19:11 Wes Bos? The animations probably don't feel very consistent either. I'm a big fan of both of these. I would probably look at the motion has been like cranking lately and I'm very, I love everything that they've been putting out, both in terms of like their defaults, but also the library, the UX behind all of that. And I don't think there's any better time to do a refactor with all of the AI we have right now. Being like, like if you're on two weird things, you can just ask the little robot to go through your codebase and find a list of every single thing that is written in GSAP and figure out a plan to move each of those over to motion.
19:55 Wes Bos? I think that's a very easy thing to do and that will significantly simplify your codebase.
20:00 Scott Tolinski? Yeah, yeah, this is, uh, this is another one of these tough ones. Cause it's like, yeah, ideally you wouldn't want to have multiple of these libraries in here doing the same thing and they're both very capable. So it's like, yeah, which one should win, whatever. Is, I wonder like how many people are using, you've inherited a codebase. So if this is the, if you are the only person working on this codebase, then like, just pick which one you like more and they're both good libraries. And if you're not the only person, have a powwow about it. I do, I'm the type of person who makes an effort to just drop what they're doing and clean this stuff up.
20:45 Scott Tolinski? Even if that's not the most reasonable answer.
20:49 Wes Bos? Yeah, yeah, maybe see else, don't be a weiner. You may be, uh, there's maybe a good reason why people have done both of those things. Also, maybe not, then maybe somebody just vibed a bunch of stuff and it just added GSAP by default.
21:02 Scott Tolinski? You're right, there might be a good reason, but I don't, I can't think of too many things that motion and GSAP like don't have overlap in necessarily. But that's me, I haven't used them both extensively. Okay, next one from Bill Baboo Zee Zee. Bill Baboo Zee Zee.
21:25 Scott Tolinski? How would you say that was? How would you say that?
21:27 Wes Bos? Baboozy.
21:29 Scott Tolinski? Bababoozyz, yes. Please do a YouTube video where you are given a piece of CSS code, but are not allowed to run it. Instead, you must use a visual editor like Figma or a drawing tool to recreate what the final design will look like purely from reading the code. Then see who has made the best version closest to the original. We did it.
21:51 Wes Bos? We did it.
21:51 Scott Tolinski? We will link that video.
21:52 Wes Bos? Thank you for submitting this because I had forgotten where I had seen this suggestion. I thought it was in a YouTube comment, but I could not find it. And then we found it in the Potluck submission. That was super fun to do.
22:04 Scott Tolinski? Super fun, super challenging, especially because the drawing tool that we used was terrible and hard to use. So that just made it even more fun. And then we also have done a video where we are coding a design and don't get to look at the rendered output of the code while we're doing it. And that's going to be releasing on a Friday video as well. Probably by the time you're hearing this, it will be on the Syntax YouTube channel. If you want to see that, it was, it was a blast. And CJ set us up for some really fun moments.
22:38 Wes Bos? Charles says, not a question, just a thanks. In a recent Potluck, you talked about interview threats. Opened a coding challenge this Sunday evening and immediately saw the red flag. So let me pause for a second. We were talking about how Adib Hana, he went through some interview and they sent him like a thing to NPM install. And he paused for a second and ran it through Claude and realized there was like a post-install malicious hook that would have like totally nuked his computer and stolen all of his crypto and did all these awful things. And, and we mentioned it a couple of episodes ago. And this person is, Charles is saying, thanks because I opened up a coding challenge
23:23 Wes Bos? this Sunday evening, immediately saw the red flags. Google Drive, zip link, API key, weird API key, follow-up emails, strange spelling and formatting. I hope I would have noticed before your episodes, but I never know a thing. So man, you gotta think like the scammers. Can get the like weird spelling figured out by now, you know, with all the technology we have. If the fact that your your English is broken and not very good, like that that's the dead giveaway. Like even at Sentry, we do these like phishing email things where we have to like we had training to like learn how to detect the phishing email. And it's often like if it's written in poor English and like I it's almost the opposite now.
24:07 Wes Bos? And now if something is way too polished, that sets off my red flags and thinking, oh, someone someone ran this through an LLM. So good.
24:17 Scott Tolinski? And people still don't see that the M dashes are dead giveaway.
24:22 Wes Bos? Oh, the M dash is the the least of the worst. I just see it in absolutely everything now, you know?
24:27 Scott Tolinski? Oh, I know. I'm just saying like AI tells. It is crazy that everyone knows that it's an AI tell, yet I still get things with M dashes 24/7.
24:37 Wes Bos? I don't know if everybody does. Like, I think we're we're in it. I went to a a fair the other day in like our cottage, small town, you know? And they have all these different vendors and they're selling stuff or there's signs and stuff. And man, oh, man, have they they found out about ChatGPT, man? All of them, all the signs, all the logos of all these things.
25:04 Wes Bos? All of the write ups of these, it's it's unbelievable how much it has like shook. Like, small town, not not necessarily small town, but it's like it certainly has reached the the outer non-technical group. And it's the greatest thing to them because remember when we first figured out about it? And you go, wow, it's outputting like, you know how exciting that was initially? And then you get jaded and realize all these things.
25:32 Scott Tolinski? Yes, I know.
25:33 Wes Bos? But good. Jay, if you guys are interviewing, certainly before you NPM install or run any code in like a test
25:42 Wes Bos? that they give you for an interview. Just give it a quick scan, run it through Claude, something like that, because it seems like this is going around.
25:51 Scott Tolinski? Stay safe out there because things are getting really real. By the way, I want to call attention to this before I read the next question. We're doing live streaming over on the Syntax YouTube. Now typically it's going to be on Mondays, but some of them have been on Thursdays due to holidays, etc. But we are going to be doing a weekly livestream over on Syntax YouTube where we're sharing with you interesting things that we found over the past week. So if you want the same kind of style that we're talking about in the podcast, but in a lightning up-to-date, like we're talking about stuff that just happened that morning on there, check out our live streams because we're going to continue to do those. And we'll take a look at any projects and stuff you want us to look at.
26:35 Scott Tolinski? And even CJ Reynolds joins us that whole time as well. All right, next one here, "Just trying to provide bytes for my family. Hey guys, thanks for the awesome show. You really helped me stay up-to-date on the latest and greatest trends in JavaScript and LLM technology. Currently, I am a mid-level software engineer at a large enterprise company, and I'd like some advice on leveling up. Most days, I'm trying to be efficient and productive with my time. Deadlines make it difficult to focus on learning while working. All too often, I feel like a code monkey." Ooh, I remember that term. The code, I used to work at a place that had a code monkey room. That was like, you could just go and work free from, form out of the code monkey room.
27:21 Scott Tolinski? Thankfully, I'm not using LLMs as a crutch. I try to spend at least 30 minutes learning something new every day. How can I learn the deep or core knowledge of software or web tech that will make me indispensable in my career? What would you recommend? Building, courses, reading docs, all of the above? Now, this one is tough because there are different answers here. So if you're focusing primarily on what will make you indispensable in the future, then your best bet is to stay on top of all of the latest. AI stuff, honestly. The techniques, the skills, the stuff that other people might not be getting deep into.
28:02 Scott Tolinski? You're going to be competing with people who are just straight up using, you know, chat GPT or cloud code and whatever. So having like a good handle on all of those things is always going to be helpful in the future at this very moment.
28:15 Wes Bos? Yeah.
28:15 Scott Tolinski? That said, that might not be the stuff that gets you excited to learn. I think that's a huge gap in the way people try to learn things as they try to learn things because they feel like they should be learning blank instead of because they're excited to learn blank. So you need to figure out what excites you and then use your time to dive into those exciting things. You'll pick things up faster. Maybe you can sprinkle in some more boring things into the exciting things. The number one tip that I have for learning things is to pick projects of stuff that you would like in your life, stuff that you would get excited to be able to use or to have,
29:00 Scott Tolinski? like even if it's just like a little robot lighting up when you click a button on a website, right? If that's exciting to you, those are the things you're going to want to spend your time on because they can really push you forward into creating, into exploring new avenues and exploring new ideas. And oftentimes that process is like what opens our brain up to picking up big concepts and things that we might not have dove into before.
29:26 Wes Bos? Yeah, I think about this a lot because I love, love learning things and I find my learning happens in two ways. First, I do what's called like surface area learning, meaning that like you're on Twitter, you're listening to podcasts, you're reading blogs, you're, you're just like kind of scanning everything out there just to kind of understand what's out there. What is the surface area of this? Reading documentation, understanding what the different APIs are, what the different texts are, what the different frameworks, languages, all of that stuff, right? That's very surface level, just understanding what's out there. But. That only gets you so far. Like, there's a lot of people out there that simply just know about
30:10 Wes Bos? everything, but they don't know like the nitty gritty of like why you might want to do something. And it's, it's only once you go and build something in those several specific areas, do you realize, oh, this is why people are, are, are so fed up with that. Like, you might be like, looking at like, the LLM landscape right now and be like, why are there so many harnesses right now? And like, why are there so many different ways to do X, Y, and Z? Why are people trying to reinvent memory?
30:44 Wes Bos? In, in LLMs? And it's because those people have gone deep with some of the other solutions and have hit the pain points. They've, they think that they can do a better job because they understand it deeply. And I feel, I feel like a lot of people aren't going deep anymore. And, and then realize, like, learning things and actually hitting edges and having to like, think and solve real problems. People are just very high surface level. I understand what's out there. I can use the LLMs to boop, boop, boop, do whatever I want. So I think it sounds like you're doing a pretty good job at like, the first level, which is like, understanding what's going on. You're listening to this podcast. I think the second level is picking a couple of things that you are interested in.
31:27 Wes Bos? And that may be, like, I think it's probably a very good point to become the custom harness, Claude code, skills, cursor, whatever, at your company. If you're a mid-level at a huge enterprise company, and you become the guy that has figured it out and has the sickest setup, and you can like, show everybody, you can have a lunch and learn, that'll start to spread. You know, like, that's probably the thing right now to, to focus on and to go deep on. So figure it out. Go, go download Hermes, or go get your, get Pi, get a custom thing set up. Not because that's what you want to use, but because I think by implementing several of them,
32:11 Wes Bos? going deep, you're gonna truly understand it, and you're gonna be able to bring those solutions to your company.
32:17 Scott Tolinski? Yeah, hit edges and try to smooth those edges on repeat.
32:22 Wes Bos? Column says, "Developers won't like hearing this. Buckle up, guys. You're not gonna like hearing this, but I generally don't update software unless I'm forced to." Column, "Seemingly harmless updates have ruined enough native apps for me that I'd rather avoid them entirely." I have a few reasons. Yeah, there's certainly enough people on Adobe Photoshop CS2. They don't want, "Don't move my cheese. I know where all my buttons are." "Apps are getting absolutely massive. They add invasive features that take away from the experience and can't be toggled off." Reels, YouTube shows. Yeah, can we talk about how awful the YouTube homepage is recently? There's like three good video and then games for some reason? Who's playing games on YouTube? I don't understand that.
33:05 Scott Tolinski? I don't understand that.
33:06 Wes Bos? Do you get games on your, you have YouTube Premium, so maybe yours is different, but mine is a Hellscape.
33:12 Scott Tolinski? Yeah, I don't, yeah, but even if you look at like the Explorer section of the sidebar on YouTube is just like a list full of things people don't care about. Explore, shopping, Playables, like, what are we doing? Yeah.
33:25 Wes Bos? So he says, "Example, I got a new phone, the latest ad-written version of a simple guitar tuner that I had used for years and was installed, and I had to find a new app. Unfortunately, where I live, local communities communicate via Facebook groups." I know, I know this all too well. Everyone's always anti-Facebook. That's where everything goes down, is Facebook groups. "There is no way to see local road conditions without seeing a shrimp Jesus or a deer on a trampoline." He's talking about like AI memes or whatever. "I need OpenClaw to check Facebook for me so I can avoid seeing shrimp Jesus." "For a long time, I used web versions of Facebook and Instagram because the apps were much more pared down, but they eventually added all of that invasive features."
34:09 Wes Bos? So true. Man, this drives me nuts. Instagram? Like, you can't pause a video on Instagram.
34:14 Scott Tolinski? Yeah.
34:15 Wes Bos? You can't scrub.
34:15 Scott Tolinski? Not easily.
34:16 Wes Bos? Sometimes, it's, you have to like hold the, you have to like swipe sideways and like, "Just let me pause a video. It's unbelievable." Anyways, this is not the rant, the rant thing here. However, since I'm a developer, I'm also on the other side of it. I want users to update immediately when I ship something. How do you reconcile the tension? Am I rational to avoid updates? Is that a normal behavior from the public, the, the general public? I've been making more of an effort recently. I keep my browser updated and I sometimes scan changelog for important security updates or features that I might want. For the most part, I still prefer early versions. What's the problem? Is it me?
34:58 Wes Bos? Well, Gollum, I think he might be a little bit of a weiner. But I, I certainly understand where you're coming from, where sometimes you don't want to update because it's, it's getting worse. Or like, like I always complain about the warp update. They put this huge update button on top of my warp and it ruins my videos all the time. And it drives me nuts because we've been using the terminal for what 30 years, 40 years now, however long terminals have existed. I don't need to update until tomorrow. It can wait. I'm doing ls -l, I don't need to update my terminal immediately.
35:40 Wes Bos? And, and I understand from their point of view, because there's bug fixes. And now people are complaining because something is broken. Like a lot of times, these updates are simply just fixing bugs that people are hitting, right? But then also there's, they're changing stuff and who moved my cheese? I think that you're probably being a bit of a weiner here. I think in many cases, the updates are trying to make the app better. In many cases, the, the, the updates are fixing bugs. And if, if I'm, I often try to catch myself being like, I don't want to update. I don't want to learn something new. Like I was, we were talking about before this thing. I learned DaVinci finally, the video editor.
36:25 Wes Bos? I was on Screenflow forever. Every single update. It got a little bit more bug ridden. It got a little bit more broken, had a couple more things that were annoying. And then I was like, I just need to bite the bullet and like finally learn this DaVinci and move on over to it. So you have to like, kind of catch yourself being like, do I not just want to update? Cause it's like, this is now more work for me to learn. Sometimes it's worth, worth that. I don't know. What do you think?
36:52 Scott Tolinski? Yeah. Well, I mean, you are asking the guy who is currently running beta versions of Mac OS, iOS, iPadOS, who chose to build a tool that we actually need to rely on in alpha software. You are asking the wrong person because, bro, I update things to a fault. I update constantly. It's a absolute miracle that I have not gotten pwned by any NPM, anything, because I'm be checking updates every day. I'm just, just constantly checking updates and running updates.
37:34 Scott Tolinski? And I will tell you, it makes my life so much harder. Uh, when I had an Android phone, I used to flash all kinds of third-party OSs on there all the time. And Courtney would be like, your phone is never working. Why don't you just leave it on the one that's coming from Google or whoever? And I, my gosh, I just cannot do that. So I sadly call him, call him, I cannot get behind the behavior of not updating your things. I like the new stuff. Even if the new stuff is broken or hardly working, I like the new stuff.
38:15 Scott Tolinski? And I'm usually a UI defender of when UI changes happen, it's almost always for good reasons. That said, this latest Mac OS is just great. Garbage in terms of the UI stuff. It is nasty work. No good. And I'm not usually a hater on that kind of stuff. So I don't, I don't, I don't know what to tell you. I think you're being probably a bit of a weiner. I think there's a good, healthy boundary of just updating your stuff when the update comes in. And I think that's probably fine, but it is, it is so funny. I think there is a type of person who is not uncommon. Who is just like, this is it. This is all I want. I'm here. I'm happy. Don't move my cheese, et cetera. Yeah.
39:00 Wes Bos? I get it. I get it.
39:01 Scott Tolinski? Move my cheese.
39:02 Wes Bos? That's unfortunately not the, the world that we live in. Things are moving. Things need to be updated. Uh, new features need to be added. It's, it is frustrating when your, your cheese is moved, but uh, like one, how, all right. Um, this is the Wes Rants Show. Um, TweetDeck or X Pro, do you use, use, you, do you use that to, to surf? I have been like a TweetDeck user for, since it, since it was an Adobe Air app, like probably 15 years. Um, and I'm like, I would consider myself like a power user of this application. And one thing they did is they moved, they took DMs out of it.
39:43 Wes Bos? And now you gotta go to X.com and click on the chats and put the stupid little code in. Uh, and I just realized I had, I think I missed like 200 DMs from people. Um, probably 90% of them were spam, but there was still quite a few good ones in there that I had totally missed. And it's because they, they moved my cheese and that wasn't my normal workflow of checking my DM requests. Cause I used to just have a column of DM requests that I would just scroll over to. And I totally missed it. So that's, I, I do understand when it's like, why did you make this thing worse? You know? Why did you take away a feature? Um, but if you're loud enough, they, they, they'll take it out. Like Twitter removed the ability to have.
40:26 Wes Bos? Like dark mode, be able to select dark mode. They just, they went entirely based on your, your OS and all, people raged because they're like, no, I want my phone to be light mode, but my Twitter to be dark mode. And, and they reinstated it. So I was happy about that.
40:42 Scott Tolinski? Yeah. If I don't, if my cheese isn't moved frequently enough, I feel like,
40:47 Wes Bos? Yeah.
40:48 Scott Tolinski? They're not doing anything. There, nobody's doing anything. This thing is dead. It's, it's cooked.
40:52 Wes Bos? Yeah. That sounds, feels good. We just got a new, uh, we use Riverside. They just rolled out a new design today. And there's a couple of bugs in it, but it feels good, you know? I'm happy that they did this. Yeah.
41:02 Scott Tolinski? I'm happy. Yes. I am happy. All right. Next one from Patrick. The rapid evolution of AI is killing my spirit as a developer. Crafting code and solving problems has been the best part of this job for 15 years for me. Now, the more I adopt different AI workflows, the more anxious I get about the future, both for the dying love for this job and the fear of becoming irrelevant. I know that this is a quite common feeling that many devs share right now, but I'm curious how you guys feel about this as people seem to really love the craft. Patrick, I really love the craft, but more than the craft, I love solving problems.
41:47 Scott Tolinski? And the thing about AI workflows that make me feel bad is that the non-determinism in solving problems makes me really stressed out. I really love that. In coding, I, I know exactly what to type or where to look to solve problems. I solve them. They're done. With AI, it often feels like, all right, I have this problem in my AI setup, not just with my code, but with my AI setup. It's not doing the things that it's supposed to be doing, or that it's been told to do, or it's trying to use, uh, different tools when it has a read and write tool built into it.
42:31 Scott Tolinski? Like those types of things make me actually angry. Like I get frustrated in like, that's what I, when I'm thinking about like my,
42:42 Scott Tolinski? Joy being massacred in this job. It's strictly because the my ability to solve problems has decreased because of things not respecting the determinism that I like. That said, I think you need to take an approach for this that there are just new problems to solve now. There are new problems to solve. There's a whole host of new problems to solve and it's your job to solve those problems if that's what you want to do. And with that, again, there's crazy new browser APIs to explore. There's crazy new workflows to explore.
43:26 Scott Tolinski? And yes, AI is different. It's very different than how we used to work. But in the same regard, you're still solving problems and you're still taking on challenges and you're still using technology, advanced technology to make and do cool things on the web. So in those aspects, I think you need to look at it holistically as like, we are able to do a whole lot of cool things with the advent of LLMs encoding. And we are able to solve problems like never before. How can I use these tools to fuel that excitement rather than to let it kill my spirit?
44:10 Scott Tolinski? I think that's where you need to be, Patrick.
44:12 Wes Bos? Yeah, Patrick, you're not being a weiner here. You're probably the first person today that's not being a weiner and
44:17 Scott Tolinski? You're not. I have a lot of empathy for you, Patrick, because I feel you.
44:21 Wes Bos? I would be lying if I said I wasn't bummed about the like dying of lots of the stuff I love. You know, like understanding closures or just like the like deep tricky parts of programming that I feel like I understand deeply are now, many parts of them are just now irrelevant. It doesn't matter that I understand how like the closures and things work as much anymore. Certainly, they still matters to a point, but I don't think it matters as much, not nearly as much as it. And even just like things like Spot the Syntax Era, that's a game that we do on our live shows.
45:06 Wes Bos? It doesn't matter anymore that we're good at spotting the syntax error. I don't, I don't think it does that you're really good at understanding where the curly bracket goes.
45:16 Scott Tolinski? But does it matter if you are good at, uh, what's a good game? It doesn't matter if you're good at Settlers of Catan or something like that. It doesn't matter. It's still fun though.
45:25 Wes Bos? It's true. We will continue to do it because it still is fun. And I think like what I've seen over my career of like teaching people to code is that you kind of had two people. You had the people that were just learning to code because they had an end goal. I want to learn JavaScript because I want to build something. I want to make something. I have this idea. I simply, I, I want a job because coding pays well. You know, like I saw all of it of people that simply just, I want to build something. I want to get that job and I'm going to learn code because that is it. And then you have the other people as they, they're using code because they, they just love writing code. They love solving problems. They love the learning new things and the intricacies of it.
46:05 Wes Bos? And the people who use code to simply get to that end goal are stoked right now because
46:12 Scott Tolinski? Yeah.
46:13 Wes Bos? they didn't necessarily love coding and, and that's, that's totally fine. You, you, maybe you loved a business or you loved a project and learning code was simply the means to the end there. And now that has, has changed. And then the people who, who simply just loved coding are feeling a little bit bummed out right now because it's like a lot of their identity and self worth, self worth is kind of diminished. So I think that's, that's totally fine to feel that. And I think a lot of people are feeling that right now. But like Scott says, there are so many more new, exciting things to sort of tackle.
46:54 Wes Bos? If you are simply looking at AI and being like, my job before was to build this thing. And now instead of building the thing, I type into the box and then it builds the thing. And that's, that's the way it's going to be forever. I, I think you're, you're wrong there, right? It's, it's totally what we are doing is, is changing. And you have to sort of buckle up and realize that what your career is. It's like what you're going to be doing day to day is is going to look dramatically different in the next couple of years.
47:27 Scott Tolinski? Yes.
47:28 Wes Bos? Solving problems, still going to be the same thing. But the areas which you dip into and the kind of the stuff you dip into is going to going to look a lot different. So that's kind of my thoughts on that. And in some regards, it sucks because it's everything is changing and you're worried for your job. And in some regards, it's really exciting because now you can start like there's so many areas of my life where I'm like, oh, now I can. Like, I was always into like microcontrollers and whatnot, but like I, I can get way more into it now, knowing that I'm a technical guy who can solve problems and I'm pretty good at thinking, but now I have this new superpower. That's a really exciting way to do it. You know, video editing is another aspect of it as well.
48:11 Wes Bos? You know, I'm using Claude Code to build DaVinci Resolve plugins right now. And I don't think that that's something that I would have been able to do before that.
48:19 Scott Tolinski? Yeah. Yeah, I feel that. I think that was very well stated. You know, there's so much there. We, those of us who really care about this craft have spent a lot of time and energy into this craft. So. Yeah.
48:35 Wes Bos? That sucks.
48:36 Scott Tolinski? I get it.
48:36 Wes Bos? Happens to a lot of people, you know? Plumbers, lots of plumbers are switching from, what was it? Switching from like copper to like pecs and like in crippling them. And it doesn't mean that copper is gone. And in fact, there's lots of people who are, I would totally want copper. But like that, like amazing skill of being able to sweat a joint and perfectly seal it up. And you get these guys coming in with their little Milwaukee.
49:03 Scott Tolinski? Tubes.
49:03 Wes Bos? Yep. You know, like this stuff changes all the time. Like the only constant in life is change and unfortunately you have to embrace that.
49:13 Scott Tolinski? Adapt or die. Yeah. All right. Well, what a positive note to end on. Wes, let's talk sick picks and shameless plugs. Wes, you got any sick picks for me today?
49:27 Wes Bos? All right, I got a sick pick. I don't know if I have sick picked this in the past, but I'm going to repick it again. And this is what is referred to as a plastic welder. So if you have ever.
49:40 Wes Bos? Have something that's plastic that breaks. So namely, the thing that always breaks for us is the bins in our fridge. You know, the kids grab on them, they crack and then they start to sag. Or like, we had like a wheelbarrow that had a crack in it and it was like starting to open up.
49:56 Scott Tolinski? I got one of those, yeah.
49:56 Wes Bos? Yeah, it happens all the time where plastic stuff breaks and you can't like super glue it together because that doesn't work. So what this does, this plastic welder is you have these like staples that, let's see if I can show the camera if you're watching the video. You have these like squiggly staples and you simply just press the button on the, on this gun here and it will heat it up like immediately. And then you just sink the staple into the plastic and then you let go and the plastic rehardens again. And it works so, so well. It's unbelievable how good this is. And it's, it's one of these tools that's like, I don't know, 30 bucks, 40 bucks to buy.
50:38 Wes Bos? And it's, it's saved us like hundreds of dollars in buying like bins and wheelbarrows and just stuff around the house that you can't just simply glue. You need to like physically, it needs to be mechanically put back together.
50:52 Scott Tolinski? Melted, yes. I love this. Yeah, I, I mean, my wheelbarrow is currently cracked and I was like, oh, where do I get a new wheelbarrow? I've had this thing forever.
51:02 Wes Bos? Well, I, I literally picked up one off the side of the road, something I do all the time.
51:07 Scott Tolinski? This is so you, yes.
51:07 Wes Bos? Is I, I will drive and I see stuff on the side of the road that people are throwing out and then I stop and look at it and go, I could fix that. Cause you, you look at it and go, oh, it's, it has a crack in it, right? People, you crack it, you throw it in the garbage, right? But I, I love fixing things. So this is a very, very good tool for that.
51:25 Scott Tolinski? Yes, word. I'm going to sick pick a couple of things, one of which is USB LEDs. They are great for your car, your car, like there's parts of my car that get really dark or either in the backseat that it'd be nice to have a little LED shine in there. And these things are just little USB with an LED on the bottom. They light up little spaces. You can put one in your server rack or something.
51:50 Wes Bos? Is it, is it just like a little, like a little puck or is it like a strip?
51:54 Scott Tolinski? It's simply just, it looks like, like a dongle, like a really thin, a really small dongle with an LED that you can change colors or have it automatically change colors or whatever on there. Um, just really simple, just pop it in there and lights up spaces. So like dark spaces that have USB but no light. That works really well. I'm, I put them in my car, put one in my wife's car and her, her glove box and stuff. And they're light sensitive, so they only come on when it's dark out.
52:21 Wes Bos? Oh, that's great.
52:22 Scott Tolinski? Yes, they're really nice little things. Um, and then another car thing, Wes, is I got a wireless CarPlay adapter. And I would have thought, I don't know why I would have thought these would have been more expensive, but it's 29 bucks. And you just pop it in the USB. My USB in my Ioniq has, for the CarPlay USB adapter, is like in the weirdest spot, a place you would never, ever want to keep your phone. So because of that, I just never used the CarPlay for it. And so I just got a wireless CarPlay mini dongle, plugged it in. Now I got wireless CarPlay.
52:58 Wes Bos? That's great. I always wondered if those are good or not. Is.
53:01 Scott Tolinski? It works.
53:02 Wes Bos? I installed like a head unit in our van that had wired and wireless. And you're right, the, I put the wired one in the glove box. But it had a wireless. Wireless is the best. You just simply just get in your car and it connects.
53:15 Scott Tolinski? Yeah. And I, I even like the Hyundai UI in my head unit and everything. But the, I would much rather have Google Maps support. Cause like the Maps is fine. It has traffic. It has those types of things, but I would much rather use Google.
53:30 Wes Bos? So good. Especially iMessage. We have Tesla Model Y and it has like SMS integration and it's garbage. It doesn't work half the time. It doesn't like, somebody reacts and it's like, Caitlyn, send a message. And it doesn't say what it is, if it's like just an emoji and like, come on, it's awful. Um, and I just want iMessage integration or like Waze, you know, tell me where the cops are so I can slow down.
53:58 Scott Tolinski? Tell me where the cops are.
54:00 Wes Bos? Yeah.
54:01 Scott Tolinski? So I can hide my drugs. No, whatever. Um. Okay, I got a shameless plug for you, Wes.
54:08 Wes Bos? Yes.
54:09 Scott Tolinski? My wife has been doing a podcast. She's a doctor of psychology. She taught in schools. She ran a testing center for kids and she's taught master's level human development courses. And now she has a podcast about parenting. So if you want to hear a podcast where it's scientific parenting tips, done in a way that's fun. In fact, our latest episode, as the time of recording this, was on board games and like which board games are actually educational for your kids, different board games for different age groups and how the idea of like a family game night, of regular family game night can build all kinds of awesome skills within your kids. So check that episode out. It was a lot of fun.
54:53 Wes Bos? Don't, don't, people listening, don't be a weiner and let your kids play board games. That's.
54:59 Scott Tolinski? Don't be, don't, Wes, you don't even know what the benefits are. Builds resiliency. It's building ritual amongst your family, family time.
55:07 Wes Bos? My kids love playing story and board games and whatnot. I just, I can't do it. They, they play with, with mom. Yeah.
55:16 Scott Tolinski? Do it for the kids.
55:18 Wes Bos? Scott, this, your phases.fm says 2025 in the footer.
55:23 Scott Tolinski? Blame that on transistor.
55:26 Wes Bos? Oh, I got a, I got a text from Justin Jackson just this morning. I'll, I'll send him a message.
55:34 Scott Tolinski? Well, I'm updating it. There is a way for me to update it, but I never even changed that. So I never set that in the first place.
55:40 Wes Bos? Yeah.
55:41 Scott Tolinski? That podcast is Phases. Check it out at Phases FM on YouTube or phases.fm.
55:47 Wes Bos? All right. Please submit us your questions. Go to syntax.fm and click on the potluck button. Submit your questions. We'll answer it on a future show. Peace. Peace.
Previous 1009: 54% AI-Generated and Climbing — State of AI Next 1011: tmux + Terminal Maxxing with Ben Vinegar Page updated October 30, 2026 · transcript and summary are machine-generated