WEBVTT 00:00:12.680 --> 00:00:16.620 Scott: Welcome to the Bikeshed Podcast, where we talk about all things software engineering, 00:00:17.250 --> 00:00:23.320 Scott: current events, and the game of life. I'm your co-host, UI simp, Alacritus Pollyanna, 00:00:23.350 --> 00:00:27.860 Scott: and Coke Zero addict, Scott Kaye. And along my side with me is my co-host, 00:00:28.520 --> 00:00:35.280 Scott: Web Framework connoisseur, the most feared PR reviewer, and GitHub Copilot's own copilot, Matt Hamlin. 00:00:36.960 --> 00:00:38.840 Scott: Matt, what's our agenda today? 00:00:40.100 --> 00:00:41.280 Matt: What a wonderful intro. 00:00:42.700 --> 00:00:53.220 Matt: Yeah, I think today we wanted to dig into a little bit about enshittification and like some topics are maybe related to that as it relates to web development, but maybe also more broadly. 00:00:55.340 --> 00:01:01.380 Matt: For those that don't know, enshittification is a fairly popular term in the recent maybe five years, maybe longer. 00:01:02.910 --> 00:01:08.520 Matt: Just sort of like things are getting worse is the maybe general consensus. 00:01:10.400 --> 00:01:18.720 Matt: It's like the quality of products that we buy is deteriorating or you pay more, but you get less. 00:01:20.030 --> 00:01:21.080 Matt: So those kinds of concepts. 00:01:21.300 --> 00:01:25.940 Matt: But then I think also looking at it from the lens of web development or just development in general. 00:01:27.800 --> 00:01:30.320 Matt: Yeah, I think that's kind of like a rough concept. 00:01:30.580 --> 00:01:38.060 Matt: I think with that, we also want to kind of talk a little bit about this blog post from Grant Slatton called Nobody Cares, 00:01:38.140 --> 00:01:42.040 Matt: which is kind of an interesting, maybe sort of related. 00:01:43.400 --> 00:01:50.460 Matt: And then also maybe dig into a term I think one of us coined, the "shadification" of the web. 00:01:51.320 --> 00:01:53.660 Matt: so when we get to that we can dig into that a little bit more 00:01:55.600 --> 00:01:57.460 Matt: yeah Scott like what's your take on 00:01:57.640 --> 00:01:58.760 Matt: enshittification like is it like, how do you see it? 00:02:02.020 --> 00:02:03.960 Scott: I think you summed it up quite well to be honest 00:02:04.900 --> 00:02:07.740 Scott: I kind of read it and did a little research about 00:02:08.110 --> 00:02:10.020 Scott: how people were using it and it was very much 00:02:11.020 --> 00:02:12.700 Scott: kind of how you're describing it is 00:02:13.630 --> 00:02:17.060 Scott: we have these software products out there that were once great 00:02:17.680 --> 00:02:20.480 Scott: and then now they're just currently 00:02:20.500 --> 00:02:26.040 Scott: trending downward they're not really adding new value i can i can name some examples of like 00:02:26.680 --> 00:02:31.580 Scott: things that i've noticed recently like i i listen to some podcasts on spotify and now i get 00:02:31.740 --> 00:02:38.860 Scott: commercials so the the price just goes up you're not really getting much new things i will say 00:02:39.000 --> 00:02:45.640 Scott: spotify has had some some good new stuff like um being able to do jam sessions and whatnot they're 00:02:45.660 --> 00:02:50.460 Scott: still creating features but an example of like you're paying money and now you're getting ads 00:02:51.060 --> 00:02:56.940 Scott: that seems to be counterintuitive or netflix is a good example where um you know you used to just 00:02:56.940 --> 00:03:04.560 Scott: pay $7.99 it's just the now you pay $25 for the the premium if i'm if i'm correct otherwise you don't 00:03:04.640 --> 00:03:14.260 Scott: get um hd 4k or or you get commercials so they're starting to just say oh how can we make more money 00:03:14.780 --> 00:03:18.360 Scott: and not give more, but rather make people pay for the top tier, 00:03:18.730 --> 00:03:20.020 Scott: which is smart marketing, 00:03:20.640 --> 00:03:26.020 Scott: but it's not innovating on what the product was prior. 00:03:27.300 --> 00:03:30.240 Scott: And that's not to say that those are actually two companies 00:03:30.620 --> 00:03:34.180 Scott: that I actually think are good in the tech space 00:03:34.310 --> 00:03:35.320 Scott: and still do good things. 00:03:35.920 --> 00:03:38.580 Scott: But there is just this product's been around for a while. 00:03:39.380 --> 00:03:42.120 Scott: And instead of trying to see where we can innovate 00:03:42.120 --> 00:03:49.500 Scott: and add new value it's been very much how can we monetize what we have and um the product is kind 00:03:49.500 --> 00:03:55.320 Scott: of going uh south i guess a better example would have been facebook something that people loved 00:03:55.620 --> 00:04:01.920 Scott: maybe a decade and a half ago when it first came out it was very exclusive uh and even like when 00:04:02.260 --> 00:04:08.280 Scott: instagram came out what i don't know 10 years ago it was like i was in college um like people 00:04:08.300 --> 00:04:13.760 Scott: flocked to that and then older folks use Facebook and maybe maybe some people like Facebook but 00:04:14.840 --> 00:04:19.220 Scott: a hot take I feel like software engineers move further and further away from 00:04:19.820 --> 00:04:27.320 Scott: popular social media I just always feel that's been the case what about you Matt 00:04:19.820 --> 00:04:27.320 Matt: yeah like I mean 00:04:27.780 --> 00:04:31.260 Matt: on your last point I don't think it's limited to software engineers I think it's maybe just 00:04:31.520 --> 00:04:37.380 Matt: people our age is probably sure which is like maybe dating us or whatever but 00:04:38.620 --> 00:04:42.520 Matt: But yeah, I think, yeah, I mean, you're talking about some really good examples. 00:04:42.670 --> 00:04:50.920 Matt: I think other sort of trends that come to mind is like, you know, I think like every product out there now is like some has to include AI as part of it. 00:04:50.950 --> 00:05:04.720 Matt: And like, you know, it's like, OK, there's some cool there's like some cool AI integrations, but there's places where it's just like they're just like tacking it on just to like say that they are an AI company or, you know, offer AI features or whatever. 00:05:05.040 --> 00:05:08.480 Matt: but I think sometimes those just become more annoying. 00:05:08.930 --> 00:05:10.460 Matt: It's, I think, a really good example. 00:05:10.660 --> 00:05:12.260 Matt: This is even predating AI, 00:05:12.310 --> 00:05:16.620 Matt: but Comcast's sort of customer service, right? 00:05:16.760 --> 00:05:20.800 Matt: It's intentionally designed to not help you. 00:05:21.420 --> 00:05:23.720 Matt: And now that they have an AI chatbot 00:05:23.720 --> 00:05:26.920 Matt: in front of the actual person answering the call, 00:05:26.980 --> 00:05:31.980 Matt: it's like, you know, now it's even more so designed for that. 00:05:33.560 --> 00:05:34.100 Matt: So I don't know. 00:05:34.260 --> 00:05:45.580 Matt: I think there's a lot of examples of that, where it's just like all these companies are maybe building features that don't really serve the need of the original product. 00:05:45.760 --> 00:05:55.540 Matt: I think that's maybe a more broad sort of topic of companies eventually grow to the scale or the size or the product grows to the point where it's no longer serving the original use case. 00:05:56.220 --> 00:05:57.640 Matt: Sometimes that works out, sometimes it doesn't. 00:05:58.500 --> 00:06:06.340 Matt: But yeah, I think just in recent times, it feels like everything, like quality-wise of everything is like dropped off a cliff. 00:06:08.380 --> 00:06:08.900 Matt: Yeah, I don't know. 00:06:09.020 --> 00:06:11.580 Matt: But like hindsight is probably more important there, though. 00:06:11.700 --> 00:06:18.560 Matt: It's like, you know, it's like using Microsoft Word 20 years ago or something was probably, you know, probably also had a lot of pain points and a lot of issues. 00:06:18.820 --> 00:06:20.380 Matt: But using it today is probably even worse. 00:06:20.510 --> 00:06:20.900 Matt: I don't know. 00:06:22.100 --> 00:06:28.360 Scott: You brought up bringing up AI is really great because that is definitely the current conundrum, I guess. 00:06:29.000 --> 00:06:43.140 Scott: But you made me think of like recently I was looking at product reviews and on Amazon and they have AI to summarize like everything in the review to tell you like maybe like what's included with the product to try to find an answer. 00:06:43.880 --> 00:06:46.740 Scott: And I feel like, yeah, that's kind of helping the user. 00:06:46.930 --> 00:06:53.800 Scott: But instead of just making a better UI and we're going to get into this or make it easier to discover product information. 00:06:54.940 --> 00:06:56.000 Scott: Amazon is a great example. 00:06:56.300 --> 00:07:06.720 Scott: I've wanted to talk about this forever that like a lot of product sites like we've built product sites, web websites. 00:07:06.930 --> 00:07:13.420 Scott: They don't really care about selling the actual product or they just want to get as many products up. 00:07:13.500 --> 00:07:19.120 Scott: there as they can and the product details they at one point were more like we need to get the 00:07:19.380 --> 00:07:26.220 Scott: details up up front but now they seem very secondary it's like how many products can we throw on a page 00:07:26.840 --> 00:07:32.500 Scott: to try to just sell something so it's like again all about monetization and less about 00:07:32.820 --> 00:07:37.660 Scott: the customer finding what they need but like i mean i use this example of amazon ai it's like 00:07:37.920 --> 00:07:43.460 Scott: yeah it helps me find like what i need but instead of just like making the page work for me 00:07:43.480 --> 00:07:53.240 Scott: throw AI in there to summarize it, which probably takes more effort and more page load or request 00:07:53.520 --> 00:07:56.440 Scott: time rather, just to get it to load for me. 00:07:56.460 --> 00:07:58.880 Scott: So that's one thing that you made me think of. 00:07:59.040 --> 00:08:05.920 Scott: And then the other, you talked about product growing where the original use case might 00:08:05.920 --> 00:08:06.260 Scott: be abandoned. 00:08:06.740 --> 00:08:12.400 Scott: Obviously, sometimes the products evolve over time and that mission statement changes. 00:08:13.320 --> 00:08:17.400 Scott: But this is something, you know, I like to harp on is like, what is our goal? 00:08:17.400 --> 00:08:18.020 Scott: What is our mission? 00:08:18.340 --> 00:08:25.340 Scott: I think when we were at Wayfair early, that was always really hitting home, hit home like that. 00:08:25.540 --> 00:08:28.880 Scott: We were trying to make it easy for people to find everything for their house. 00:08:30.500 --> 00:08:33.240 Scott: And I think that really does drive decision making downward. 00:08:33.479 --> 00:08:39.120 Scott: But we saw firsthand as products grow, people come and go. 00:08:39.360 --> 00:08:40.099 Scott: Missions change. 00:08:40.280 --> 00:08:41.500 Scott: People aren't always aligned. 00:08:41.979 --> 00:08:47.300 Scott: And what they're trying to build maybe is a different understanding to everyone. 00:08:47.400 --> 00:08:55.240 Scott: So just having that alignment to make sure that people are trying to build towards the same end goal or to see that the space that you're in might change. 00:08:56.820 --> 00:08:58.980 Scott: Try to think of a product, maybe not with an example. 00:08:59.080 --> 00:09:05.260 Scott: You might build something for a specific use case, but realize, hey, there's this use case over here and we need to pivot towards that. 00:09:07.399 --> 00:09:15.740 Scott: So inherently, I feel like enshittification is partially when they lose sight of a goal. 00:09:15.910 --> 00:09:17.980 Scott: And it's more of just like, how can we make money? 00:09:19.240 --> 00:09:21.920 Scott: Maybe it's not always a corporation, but we're really big. 00:09:22.290 --> 00:09:29.760 Scott: And we're kind of just throwing things at the wall to see what sticks for monetization value over customer satisfaction value. 00:09:31.300 --> 00:09:34.440 Matt: Yeah, I've been noodling on this topic. 00:09:35.400 --> 00:09:48.480 Matt: I think we've worked at companies where there's that sort of common phrase of as soon as a measure becomes metric, it's no longer valuable or something like that. 00:09:48.680 --> 00:10:00.780 Matt: I'm probably getting that wrong, but I think we both worked at places where the KPIs became far more important than actually delivering value. 00:10:01.880 --> 00:10:04.780 Matt: And I think that's an interesting thread. 00:10:06.820 --> 00:10:09.860 Matt: That's probably become something that's far more important 00:10:10.480 --> 00:10:14.400 Matt: maybe in the past 20 years or something than in prior times. 00:10:16.320 --> 00:10:22.100 Matt: And we've both also worked at places that compare themselves 00:10:22.360 --> 00:10:26.500 Matt: to a lot of other competitors and see how they rank up against others 00:10:26.560 --> 00:10:27.440 Matt: on different benchmarks. 00:10:28.820 --> 00:10:30.780 Matt: I've had this blog post stewing in my mind. 00:10:30.820 --> 00:10:37.320 Matt: of like you know like you like working to like get better at a benchmark does not make you a better 00:10:37.900 --> 00:10:43.260 Matt: product company um i think it's just like it you're focusing on the wrong problem you're sort 00:10:43.300 --> 00:10:47.800 Matt: of focusing on like how do you rank versus others versus like how do you actually solve the customer's 00:10:47.900 --> 00:10:53.320 Matt: needs um and so yeah i wonder if that's like yeah you know kind of like what you're saying like i 00:10:53.380 --> 00:11:00.119 Matt: wonder if this enshittification is like inherent in teams that just grow to a size that um start going 00:11:00.140 --> 00:11:02.860 Matt: after KPIs rather than solving customer needs. 00:11:03.800 --> 00:11:05.920 Matt: I think a lot of maybe CEOs or whatever 00:11:06.120 --> 00:11:07.120 Matt: sort of convince themselves like, 00:11:07.150 --> 00:11:11.800 Matt: oh yeah, if the KPI is a way to go after the customer need, 00:11:11.860 --> 00:11:14.100 Matt: I think that easily slips. 00:11:14.150 --> 00:11:15.400 Matt: It becomes a proxy metric 00:11:15.600 --> 00:11:17.300 Matt: and then that proxy gets further and further away 00:11:17.480 --> 00:11:18.480 Matt: from what they're actually trying to solve. 00:11:20.500 --> 00:11:20.900 Matt: Yeah, I don't know. 00:11:21.260 --> 00:11:25.360 Matt: It's interesting that there's this, I don't know, 00:11:27.120 --> 00:11:28.340 Matt: companies that start small 00:11:28.370 --> 00:11:29.840 Matt: and sort of really solve product needs. 00:11:30.120 --> 00:11:31.140 Matt: eventually grow to that size. 00:11:31.140 --> 00:11:31.780 Matt: I don't know if there's like, 00:11:32.240 --> 00:11:32.940 Matt: there's probably something else 00:11:33.140 --> 00:11:35.000 Matt: that sort of attributes to that. 00:11:37.360 --> 00:11:39.500 Scott: Yeah, I also think when like folks are aligned 00:11:40.660 --> 00:11:43.980 Scott: and they were like constantly building new products 00:11:44.260 --> 00:11:45.780 Scott: or just trying to build something small, 00:11:46.040 --> 00:11:47.880 Scott: get it out there, get some testing 00:11:48.360 --> 00:11:49.500 Scott: and then expand on that, 00:11:50.160 --> 00:11:51.460 Scott: we're always more successful 00:11:51.640 --> 00:11:52.440 Scott: because we were like, 00:11:53.420 --> 00:11:54.680 Scott: I guess for lack of better terms, 00:11:54.900 --> 00:11:56.780 Scott: sussing out what the user really wants. 00:11:57.200 --> 00:11:58.320 Scott: Because sometimes it's unclear 00:11:58.580 --> 00:11:59.560 Scott: and especially in a new space, 00:11:59.780 --> 00:12:01.020 Scott: Like, AI is a good example. 00:12:01.170 --> 00:12:02.960 Scott: Like, everyone's just building a chatbot, 00:12:03.290 --> 00:12:04.480 Scott: but is that really what a user needs? 00:12:04.920 --> 00:12:06.420 Scott: Like, maybe, yeah, sure. 00:12:07.900 --> 00:12:10.620 Scott: But there are probably more applicable use cases that, 00:12:11.160 --> 00:12:12.800 Scott: I mean, I had the Amazon one. 00:12:13.030 --> 00:12:14.520 Scott: I don't know if that's, like, the best example, 00:12:14.610 --> 00:12:15.600 Scott: but it is helpful. 00:12:16.860 --> 00:12:18.320 Scott: Maybe clean up the rest of your UI. 00:12:19.240 --> 00:12:20.480 Scott: But you also made me think of, like, 00:12:20.530 --> 00:12:23.780 Scott: when companies grow, you'll see, 00:12:24.060 --> 00:12:25.660 Scott: and people come and go, 00:12:26.220 --> 00:12:28.120 Scott: you'll see the same tests ran over and over. 00:12:28.140 --> 00:12:36.740 Scott: um we did i'll never forget i don't know how many header tests tested different navigations uh 00:12:37.520 --> 00:12:47.480 Scott: in my tenure at w but it's more than 20 plus and we always just like got the same data and 00:12:47.840 --> 00:12:51.640 Scott: we did record data places but like clearly it wasn't 00:12:53.800 --> 00:12:59.920 Scott: like enough of a necessity to record it and make it available and easy to find so that we could 00:13:00.580 --> 00:13:06.360 Scott: learn from it instead of doing it again and again and again with this concept that like oh maybe the 00:13:06.440 --> 00:13:13.860 Scott: way users shop for things changes sure i do think that like there is some of that but that kind of 00:13:13.870 --> 00:13:20.980 Scott: gets more towards the what we'll get to as our next topic about how a lot of web apps are starting 00:13:21.000 --> 00:13:27.420 Scott: to use like similar pieces or the same pieces but we were never really creating anything net new that 00:13:27.620 --> 00:13:32.480 Scott: hadn't been done before it was just how we were going to interpret for the user the easiest path 00:13:32.690 --> 00:13:41.080 Scott: forward um and i i wanted to just kind of go down this lane of often i think as companies grow or 00:13:41.080 --> 00:13:45.860 Scott: at least what we i have seen is they start prioritizing feature feature feature feature 00:13:46.300 --> 00:14:01.540 Scott: They're not thinking enough about the value that feature builds, but rather appeasing that stakeholder, that these features are being delivered and that they might have value, and weighing the value and pros and cons of different features. 00:14:02.740 --> 00:14:10.180 Scott: For years, we didn't have a very specific feature, and that was like comparisons. 00:14:10.300 --> 00:14:15.220 Scott: like a lot of the sites that could compare products and we never had that but we also would rebuild 00:14:15.360 --> 00:14:19.040 Scott: these features man i know you want to say something i have another thought after this but go ahead 00:14:19.520 --> 00:14:24.160 Matt: yeah yeah i guess like on the on like the feature building like i think it's less 00:14:25.300 --> 00:14:28.960 Matt: like releasing a lot of features i think it's more of like releasing a lot of optimizations 00:14:29.590 --> 00:14:34.120 Matt: i think those optimizations aren't like i think internally in those big companies they sell it as 00:14:34.140 --> 00:14:40.980 Matt: features but but like for the end user product right it's like it's rarely is it actually a 00:14:41.160 --> 00:14:44.920 Matt: feature versus just like like you were saying in terms of like the a-b testing sort of flow it's 00:14:45.000 --> 00:14:48.800 Matt: like they they they you know with all those different tests for the header they weren't 00:14:49.439 --> 00:14:54.140 Matt: necessarily always releasing new features for the header they were just like reshaping what the 00:14:54.140 --> 00:14:57.880 Matt: header looks like or changing some of the functionality of the header but like not actually 00:14:58.020 --> 00:15:03.420 Matt: like you know it's like you know at that time of wayfair we weren't introducing like brand new 00:15:03.460 --> 00:15:05.420 Matt: paradigms on how headers 00:15:05.690 --> 00:15:06.420 Matt: work on websites. 00:15:07.900 --> 00:15:09.020 Matt: Yeah, I have a further point for that. 00:15:09.860 --> 00:15:10.720 Matt: I'll let you get back to it. 00:15:11.540 --> 00:15:13.060 Scott: I absolutely agree with that. 00:15:13.790 --> 00:15:15.760 Scott: But I guess more so often... 00:15:16.720 --> 00:15:17.640 Scott: When I would 00:15:17.740 --> 00:15:18.780 Scott: do it, I can only speak for myself, 00:15:19.460 --> 00:15:21.380 Scott: we would eradicate the old code 00:15:21.690 --> 00:15:23.400 Scott: and write net new code. 00:15:23.880 --> 00:15:25.100 Scott: It is definitely true that 00:15:25.540 --> 00:15:27.380 Scott: as we grew, using 00:15:27.450 --> 00:15:29.300 Scott: the header again as an example, it seemed like teams 00:15:29.330 --> 00:15:31.040 Scott: just moved a thing around. 00:15:31.440 --> 00:15:33.400 Scott: They were not really... 00:15:34.020 --> 00:15:35.380 Scott: starting and rebuilding a thing. 00:15:36.070 --> 00:15:39.080 Scott: So the way I see it as it being a new feature 00:15:39.320 --> 00:15:40.760 Scott: is like we just built some new code 00:15:40.940 --> 00:15:42.080 Scott: and replaced some old code, 00:15:42.280 --> 00:15:44.840 Scott: but that may not be how everyone does it, but continue. 00:15:46.180 --> 00:15:49.700 Matt: No, yeah, I think, yeah, like I think internally, right, 00:15:49.880 --> 00:15:51.940 Matt: like you can look at it as like a new feature, 00:15:53.200 --> 00:15:54.240 Matt: it's like new, right? 00:15:54.360 --> 00:15:55.700 Matt: But I think if you look at it 00:15:55.710 --> 00:15:57.420 Matt: from the perspective of like a customer, right, 00:15:57.560 --> 00:15:59.340 Matt: like the visitor to the website, 00:15:59.530 --> 00:16:02.580 Matt: like they don't care if, you know, for example, 00:16:02.860 --> 00:16:04.560 Matt: Like migrating tech stacks, right? 00:16:04.740 --> 00:16:06.420 Matt: Like they don't care if the previous tech stack was, 00:16:07.220 --> 00:16:10.080 Matt: you know, or like if the team behind the product 00:16:10.250 --> 00:16:11.680 Matt: like swapped out the tech stack, right? 00:16:11.860 --> 00:16:14.880 Matt: Like the features generally usually are like on parity 00:16:14.970 --> 00:16:16.920 Matt: with what they're using, you're used to before. 00:16:18.300 --> 00:16:20.760 Matt: I think more broadly, it's like making me think of like, 00:16:21.160 --> 00:16:21.880 Matt: you know, companies become, 00:16:23.340 --> 00:16:24.620 Matt: and this like phrase sort of came to me 00:16:24.630 --> 00:16:25.620 Matt: as you were talking about like companies 00:16:25.760 --> 00:16:27.260 Matt: sort of become optimization engines 00:16:27.640 --> 00:16:28.880 Matt: over innovation engines, right? 00:16:28.960 --> 00:16:30.980 Matt: Like I think when you're a small company, 00:16:31.280 --> 00:16:32.380 Matt: you focus on innovation, right? 00:16:32.460 --> 00:16:43.600 Matt: You focus on how can you solve a problem in a new way compared to people already in the market or people that have tried solving that problem in the past. 00:16:44.200 --> 00:16:50.920 Matt: And then when you're a big company, you start to focus more on these A-B tests, like these slight tweaks to different things. 00:16:51.280 --> 00:16:53.960 Matt: It's like we've probably heard that popular article. 00:16:54.020 --> 00:16:58.620 Matt: I don't know if it's true or not, but like Google A-B tests the color blue on the links. 00:16:59.640 --> 00:17:06.500 Matt: So, yeah, and, you know, it's like, oh, we changed the hue of the blue and it's like we get 10 percent more conversion on people clicking through to ads or whatever it is. 00:17:08.300 --> 00:17:11.280 Matt: And so it's like, yeah, that's like an optimization, not a innovation. 00:17:11.589 --> 00:17:23.600 Matt: Right. So I wonder if it's just like so many people are just like used to or like maybe like those optimizations are seen as just like not making the product better. 00:17:23.880 --> 00:17:25.800 Matt: Like they're not innovating in the space that people expect. 00:17:26.120 --> 00:17:32.700 Matt: You know, it's like, it's like completely different scale, but it's like people asking like where the flying car is, right? 00:17:32.840 --> 00:17:36.700 Matt: Like, you know, we're not seeing that kind of growth happen. 00:17:38.500 --> 00:17:38.580 Matt: Yeah. 00:17:39.580 --> 00:17:41.620 Scott: So you're actually, you're changing my perspective a little. 00:17:41.680 --> 00:17:45.940 Scott: I do think there were like some, I guess, net new features that they were trying to build. 00:17:46.140 --> 00:17:54.260 Scott: But they're actually, you're now making me think that we were at our strongest when we were trying to build new things. 00:17:54.660 --> 00:17:57.980 Scott: Like, that's like a new product as part of the product. 00:17:58.240 --> 00:17:59.720 Scott: It could be like a feature within the product. 00:18:00.420 --> 00:18:02.120 Scott: And we were trying to iterate on that. 00:18:02.660 --> 00:18:04.960 Scott: And that was probably when we were at our best. 00:18:05.420 --> 00:18:08.660 Scott: But what I wanted to bring up is what you're talking about, like optimizations. 00:18:10.260 --> 00:18:13.780 Scott: They were optimizations, but at the cost of not looking at performance. 00:18:16.500 --> 00:18:21.780 Scott: I think maybe other companies did different, but we wouldn't prioritize performance enough. 00:18:21.980 --> 00:18:50.440 Scott: It would constantly be brought up that we either needed some sort of performance metric to work against that teams would adhere to, or people believed performance was important, but it wasn't like, I guess, a first-class citizen in the work people did, or well-taught how to make sure that they delivered that, but rather it was very secondary in thought. 00:18:51.980 --> 00:18:57.620 Scott: my original point was like before you deliver something net new and and if you're a small place 00:18:57.720 --> 00:19:04.020 Scott: like performance can be a little easier you still have to care about it but like you might be more 00:19:04.160 --> 00:19:08.020 Scott: likely to have a smaller app and therefore be able to get more out it's not always the case 00:19:08.500 --> 00:19:14.460 Scott: um definitely have had implications growing growing apps at smaller places but you're able to 00:19:14.960 --> 00:19:19.699 Scott: like there's no one else there when something isn't performing in a small company and you're on the 00:19:19.720 --> 00:19:27.280 Scott: hook at a big company your goal is more to appease the stakeholder that they like what you do and 00:19:27.680 --> 00:19:32.560 Scott: it's easier to there's maybe a strong term but like scapegoat another team has to solve 00:19:33.190 --> 00:19:41.220 Scott: problems um which maybe this is like a more big company problem but um it's easier to 00:19:42.460 --> 00:19:44.080 Scott: act as if you have less 00:19:45.520 --> 00:19:46.160 Scott: ownership 00:19:46.280 --> 00:19:48.060 Scott: of a thing when 00:19:49.220 --> 00:19:50.120 Scott: you're described 00:19:50.160 --> 00:19:51.860 Scott: as owning a feature 00:19:52.520 --> 00:19:54.200 Scott: and you view it 00:19:54.660 --> 00:19:55.400 Scott: in a perspective 00:19:56.240 --> 00:19:57.640 Scott: that like I just own this section 00:19:58.799 --> 00:19:59.840 Scott: and especially 00:20:00.040 --> 00:20:00.220 Scott: like 00:20:01.440 --> 00:20:03.860 Scott: well this is maybe more specific 00:20:03.860 --> 00:20:05.860 Scott: to the background of like some folks 00:20:05.900 --> 00:20:08.200 Scott: at our company at the time 00:20:08.820 --> 00:20:10.079 Scott: just owned visuals 00:20:10.740 --> 00:20:13.900 Scott: and they never really cared about APIs and endpoints. 00:20:16.399 --> 00:20:18.980 Scott: And maybe it wasn't trained well enough that, 00:20:19.080 --> 00:20:20.720 Scott: hey, that is part of the front end these days. 00:20:21.500 --> 00:20:24.780 Scott: But understanding how bit, 00:20:24.850 --> 00:20:28.040 Scott: well, also having service teams that provide these for you, 00:20:28.590 --> 00:20:31.040 Scott: that blurs the lines a little bit more. 00:20:31.430 --> 00:20:36.300 Scott: But essentially, it's harder to draw ownership lines 00:20:36.500 --> 00:20:39.059 Scott: when there are more people owning things 00:20:39.220 --> 00:20:43.740 Scott: and there isn't clear defined guides but when you're in a smaller place you don't need clear 00:20:43.940 --> 00:20:50.820 Scott: defined guides like everyone has a stake in care and ownership so um the purpose of this getting 00:20:50.830 --> 00:20:58.920 Scott: back to and enshittification is sometimes it can happen because lines become blurred and people's 00:20:59.170 --> 00:21:06.219 Scott: motives are to a specific problem that they've been told is what their ownership area is or 00:21:06.720 --> 00:21:12.440 Scott: it's just all they think their ownership area is and not clear guidelines about working together 00:21:12.510 --> 00:21:18.720 Scott: as a cohesive unit and making the whole product better i mean people talk about 10x engineers all 00:21:18.800 --> 00:21:24.220 Scott: the time but like really all it means is that like you want to learn and understand the whole sequence 00:21:24.760 --> 00:21:31.120 Scott: of how the applications and the services work together and that's what makes you a strong 00:21:31.360 --> 00:21:36.200 Scott: engineer like i'll toot your horn a little bit matt but one of the things that made you so strong 00:21:36.220 --> 00:21:43.260 Scott: at W is you understood better than most how all the services worked and came together to make the 00:21:43.380 --> 00:21:49.440 Scott: application work and that is like a superpower and I've learned from that that you need to just 00:21:49.500 --> 00:21:56.260 Scott: make sure you understand the architecture first and foremost 00:21:49.500 --> 00:21:56.260 Matt: yeah I think I mean I think yeah yeah 00:21:56.260 --> 00:22:01.280 Matt: I think a lot of that stuff yeah I think it rings true but I think it's also like you know people 00:22:01.420 --> 00:22:06.180 Matt: listening might not understand like that you know the structure they're at internally I think really 00:22:06.560 --> 00:22:11.140 Matt: sort of the core emphasis is like at a larger company your focus your area of focus is like 00:22:11.780 --> 00:22:16.900 Matt: well is pretty small generally speaking you know it's like okay like as you get higher up in the 00:22:17.000 --> 00:22:21.340 Matt: stack like in the career you know career ladder at that company it's like yes you're going to have 00:22:21.340 --> 00:22:28.920 Matt: a broader perspective um but generally speaking most IC's are just very focused on one specific 00:22:29.040 --> 00:22:33.420 Matt: feature one specific product area and so yeah exactly like what you were saying it's like that 00:22:34.040 --> 00:22:40.840 Matt: leads to like, you know, sometimes people just don't really care about, you know, the broad, 00:22:41.070 --> 00:22:44.920 Matt: like how does this feature fit into the rest of the product or like the customer, you know, 00:22:44.920 --> 00:22:50.240 Matt: like what customer pain point is this solving compared to like the other aspects of the product 00:22:50.420 --> 00:22:57.480 Matt: that the customer is using, which I think like that kind of leads into maybe our like next sort 00:22:57.480 --> 00:23:01.700 Matt: of little section of this like blog post titled Nobody Cares. I think that's like, 00:23:02.840 --> 00:23:05.560 Matt: I think, I mean, maybe this is like not the way that the original, 00:23:05.750 --> 00:23:07.520 Matt: like originally the blog post was meant to be taken, 00:23:07.700 --> 00:23:09.940 Matt: but like, I think it's like, 00:23:10.180 --> 00:23:13.460 Matt: I've seen this sort of ring true where the, like, yeah, 00:23:13.640 --> 00:23:18.100 Matt: where engineers aren't empowered to take a more holistic point of view on the 00:23:18.260 --> 00:23:21.820 Matt: entire product. Right. It's like you're, 00:23:22.140 --> 00:23:24.600 Matt: if you're told that you can only work on, you know, feature a, 00:23:25.260 --> 00:23:29.200 Matt: you might not feel like you have the ability, the agency, you know, 00:23:29.400 --> 00:23:31.460 Matt: whatever the leverage to start, you know, 00:23:31.540 --> 00:23:35.380 Matt: sort of recommending changes or applying changes to feature B, C, and D. 00:23:35.760 --> 00:23:35.820 Matt: Right. 00:23:35.940 --> 00:23:39.000 Matt: So it's like you, in many ways, 00:23:39.000 --> 00:23:43.380 Matt: it's like whether or not it's like the actual like engineer doesn't care or 00:23:43.660 --> 00:23:46.800 Matt: they're just like sort of forced to a position where they can't, 00:23:48.280 --> 00:23:50.980 Matt: can't apply the energy to care for those other things. 00:23:51.540 --> 00:23:54.120 Matt: I think that's like the, like, I think, 00:23:54.120 --> 00:23:57.400 Matt: I think it's just like an interesting sort of perspective on that blog post. 00:23:59.280 --> 00:24:01.100 Matt: Yeah. I don't know. It's a, yeah. 00:24:02.200 --> 00:24:10.680 Scott: yeah and i think it's a scaling factor right like or it's also a strategy of how do we create 00:24:11.420 --> 00:24:17.700 Scott: like we talked about like pods like that was like the big thing that was and it never really changed 00:24:17.840 --> 00:24:24.560 Scott: but it was like okay people own these specific areas i would have argued where we were we there 00:24:24.560 --> 00:24:29.320 Scott: could have been larger pools of ownership that crossed over that would help with the the blurring 00:24:29.340 --> 00:24:37.560 Scott: of these lines but um i think the bigger point is when you can't define from the top 00:24:39.770 --> 00:24:47.440 Scott: clear responsibilities for the ownership of your job um and be held accountable so like i'm thinking 00:24:47.560 --> 00:24:53.200 Scott: kind of the RACI principles like if you can't define who's responsible or accountable for 00:24:53.340 --> 00:25:01.660 Scott: something like an endpoint speed or you don't have mechanisms to make it clear about who owns what 00:25:02.440 --> 00:25:07.880 Scott: because the size of your org is really big you this is you start to have these like fractions of 00:25:08.360 --> 00:25:14.220 Scott: you have a problem you know about it but you don't think you own it you might but assuming good intent 00:25:14.860 --> 00:25:19.780 Scott: you might not think that you own performance when there's a team that like is called performance 00:25:19.800 --> 00:25:35.400 Scott: Right. But then again, if there isn't like some clear guideline to go by to know what performance they own or who to go to to ask about these things or to care, you know, that that's where the breakdown kind of begins, too. 00:25:35.830 --> 00:25:48.200 Scott: So it's just like really setting expectations well and defining those expectations, because I do think like these way these things are broken down could be successful, but it's more maybe of a lack of education. 00:25:49.520 --> 00:26:00.000 Matt: Yeah, I would even argue that it's not even the, like, I think that the root problem isn't that it's like you don't know who to talk to about it. 00:26:00.280 --> 00:26:05.360 Matt: It's like that that's even the mindset that you come into that with, right? 00:26:05.480 --> 00:26:22.300 Matt: Like, it's like, if you're in a culture where you're not empowered to just make change, then you are going to naturally say, like, okay, I need to ask someone about, you know, I need to ask someone about if it's all right to make this change or if I should be caring about this or whatever. 00:26:22.960 --> 00:26:29.760 Matt: Versus if you're just empowered to make that change, which I think that's the case usually for most startups. 00:26:30.460 --> 00:26:37.280 Matt: The company is so small that there's literally no one else knows the answer to your question anyway, so you just have to do it. 00:26:38.480 --> 00:26:51.540 Matt: But in a big company, I think that usually the culture is very much like, yeah, get sign off, go through some red tape to sort of verify that you're allowed to make that change and then make the change. 00:26:53.340 --> 00:26:58.440 Matt: yeah I don't know I think it yeah like it is a I think that's just like another sort of cultural 00:26:58.880 --> 00:27:03.700 Matt: factor that leads into why you know why you know Grant in his blog post talked about like nobody 00:27:03.900 --> 00:27:09.120 Matt: caring right like they're not empowered to at least at engineering companies right they're not 00:27:09.340 --> 00:27:16.440 Matt: empowered to you know make the change they want to see or actually like empowered to like have 00:27:16.460 --> 00:27:19.540 Matt: the facility to care about the problems they're solving. 00:27:20.500 --> 00:27:25.340 Scott: Yeah, there's definitely more bureaucracy of change making. 00:27:25.440 --> 00:27:29.540 Scott: I remember someone would propose a change and people would be like, 00:27:29.540 --> 00:27:33.500 Scott: whoa, whoa, whoa, looking at the change, really scrutinizing that change. 00:27:33.900 --> 00:27:37.460 Scott: Because when you're siloed on a small team and you all start to work together and gel, 00:27:37.500 --> 00:27:39.180 Scott: you start to have common beliefs. 00:27:39.380 --> 00:27:43.260 Scott: But getting that out there to the rest of the organization 00:27:43.500 --> 00:27:46.160 Scott: or sharing and spreading those beliefs gets more difficult. 00:27:46.260 --> 00:27:52.020 Scott: unless your team is literally given time to document patterns for everyone to use. 00:27:52.650 --> 00:27:59.420 Scott: I'm clearly taking the stance of a team that is setting up a platform for my platform experience. 00:27:59.860 --> 00:28:03.180 Scott: But yeah, I am saying there's some lack of structure that is to blame, 00:28:03.360 --> 00:28:11.040 Scott: but I do agree with you to take that culturally working together 00:28:11.820 --> 00:28:15.440 Scott: and understanding and creating a culture of how we solve problems 00:28:15.460 --> 00:28:17.520 Scott: is also big to that. 00:28:17.560 --> 00:28:18.580 Scott: But I think that's like, 00:28:19.080 --> 00:28:20.960 Scott: like basically structure, 00:28:21.920 --> 00:28:22.360 Scott: a structure, 00:28:22.580 --> 00:28:24.020 Scott: like we need to like enforce structure 00:28:24.100 --> 00:28:25.820 Scott: from the top down to help inform. 00:28:27.940 --> 00:28:28.100 Matt: Yeah. 00:28:28.860 --> 00:28:29.200 Matt: Yeah, I think. 00:28:29.960 --> 00:28:30.080 Matt: Yeah. 00:28:30.580 --> 00:28:31.060 Matt: Yeah, I agree. 00:28:32.300 --> 00:28:32.400 Matt: Yeah. 00:28:32.480 --> 00:28:33.280 Matt: So I guess like, 00:28:33.680 --> 00:28:34.740 Matt: I think maybe we've gotten 00:28:34.960 --> 00:28:36.400 Matt: a little bit far away 00:28:36.640 --> 00:28:37.200 Matt: from gentrification, 00:28:37.580 --> 00:28:38.960 Matt: but I think another topic 00:28:39.220 --> 00:28:40.160 Matt: that came to mind 00:28:40.340 --> 00:28:41.900 Matt: as we were talking about this topic 00:28:42.120 --> 00:28:43.260 Matt: as like being something 00:28:43.260 --> 00:28:44.340 Matt: we wanted to talk about on the podcast 00:28:44.660 --> 00:28:44.960 Matt: was like, 00:28:46.700 --> 00:28:50.440 Matt: was sort of the, I can't think of a good word, 00:28:50.450 --> 00:28:53.620 Matt: but like sort of the explosion of websites 00:28:54.140 --> 00:28:56.740 Matt: that have started to use shadcn/ui, 00:28:56.910 --> 00:28:58.140 Matt: which I don't know if I'm pronouncing correctly, 00:28:58.230 --> 00:29:00.000 Matt: but I think that's the way you pronounce it. 00:29:01.040 --> 00:29:03.880 Matt: There's a pretty popular like component library 00:29:04.080 --> 00:29:05.900 Matt: that's a little bit different from most component libraries 00:29:05.950 --> 00:29:07.040 Matt: that people have used in the past 00:29:07.210 --> 00:29:10.440 Matt: in the sense that it's like a sort of like copy the code 00:29:10.510 --> 00:29:12.240 Matt: into your own code base instead of installing it 00:29:12.420 --> 00:29:13.540 Matt: from NPM or something. 00:29:16.360 --> 00:29:21.420 Matt: but I think what we've seen over maybe the past year or two I think it's been around at least two 00:29:21.540 --> 00:29:28.200 Matt: years is that more and more sites yeah more and more sites are just like like they all look the 00:29:28.360 --> 00:29:32.020 Matt: same you know it's like if any if anyone's sort of announcing like oh I'm building this like 00:29:33.040 --> 00:29:36.920 Matt: new you know generally these days it's like an AI product but like sort of new website 00:29:37.720 --> 00:29:42.320 Matt: and they share a screenshot it's like you can 100% tell that they just like pulled in 00:29:42.540 --> 00:29:45.140 shadcn and used it, like used the defaults. 00:29:46.540 --> 00:29:50.500 Matt: And I think it's like, it's an interesting pattern that we've seen. 00:29:50.910 --> 00:29:55.940 Matt: I think like a very clear telltale sign these days is if it's like monochrome. 00:29:56.510 --> 00:30:01.300 Matt: Like shadcn has colors, but like the default for everything is like black and white. 00:30:02.640 --> 00:30:07.360 Matt: I think it's like maybe people don't realize that they were meant to customize it when they 00:30:07.530 --> 00:30:07.920 Matt: added it. 00:30:08.210 --> 00:30:09.980 Matt: I think he's just like pulled it in and said, that's it. 00:30:10.540 --> 00:30:12.520 Matt: But yeah, I don't know. 00:30:12.680 --> 00:30:15.120 Matt: Yeah, Scott, do you have some thoughts there? 00:30:16.760 --> 00:30:18.140 Scott: Yeah, all of that. 00:30:18.340 --> 00:30:19.200 Scott: Yes, you're right on. 00:30:20.679 --> 00:30:22.180 Scott: You'd notice a lot of... 00:30:22.430 --> 00:30:24.040 Scott: And the style for websites, 00:30:25.050 --> 00:30:26.760 Scott: there is these trends that it goes through. 00:30:26.870 --> 00:30:29.820 Scott: But now it's like I have this marketing site 00:30:30.200 --> 00:30:33.260 Scott: where I have this really large hero 00:30:33.540 --> 00:30:38.700 Scott: with slight white or dark black to gray gradient text on it. 00:30:39.040 --> 00:30:40.180 Scott: It's really subtle. 00:30:40.860 --> 00:30:45.660 Scott: but there's also these like glowy like things going on in the background you can see it almost 00:30:46.260 --> 00:30:52.780 Scott: on like every every little startup site um but you are right shud's shadcn and he's even said it 00:30:52.900 --> 00:30:58.600 Scott: himself uh the the writer of it that like the goal is to give it to you this way so you could 00:30:59.040 --> 00:31:06.320 Scott: be a jump off point for you to create what you wanted um although that's doesn't happen often 00:31:06.340 --> 00:31:14.060 Scott: but anyway so yeah like it's meant to be customized but um i think what it's led to though and it's 00:31:14.060 --> 00:31:21.380 Scott: not just shadcn but it's like a lot of these ui systems and i guess design systems they they start 00:31:21.520 --> 00:31:27.280 Scott: to solve things the same like even when we worked on design systems we would look for like third 00:31:27.480 --> 00:31:31.700 Scott: parties that would solve certain little problems and now everyone's like okay i'm gonna solve that 00:31:31.720 --> 00:31:38.780 Scott: problem that way there's become like standards like radix right radix radix is one um but like 00:31:39.440 --> 00:31:45.640 Scott: you know handling handling accessible navigation or tabbing and whatnot like those kinds of 00:31:45.900 --> 00:31:51.180 Scott: standards start to creep in which those are fine but more so that it's just like okay we have this 00:31:51.380 --> 00:31:55.880 Scott: concept that's a modal and now we're getting like tags for it we have this concept that's a drawer 00:31:56.440 --> 00:32:01.680 Scott: and we start to just be like okay these are the pieces um and even when we were working on it was 00:32:01.900 --> 00:32:05.040 Scott: We would tell the design team, use the pieces, use the pieces. 00:32:05.340 --> 00:32:12.740 Scott: So the design team is constantly, or the design systems team is constantly being asked to fix those things. 00:32:13.000 --> 00:32:16.300 Scott: And then when they're looking for inspiration, they're just all looking at each other. 00:32:16.780 --> 00:32:19.500 Scott: We start to have all these same patterns that emerge. 00:32:20.000 --> 00:32:25.160 Scott: And like, yeah, you see like a sidebar nav or like a drop top drop down nav, 00:32:25.230 --> 00:32:30.060 Scott: but you don't see as much innovation or chances to like, how can we do things? 00:32:32.060 --> 00:32:33.800 Scott: how can we do things in a new light? 00:32:35.160 --> 00:32:35.880 Matt: Yeah, 100%. 00:32:35.910 --> 00:32:37.060 Matt: Yeah, I think it's like, you know, 00:32:38.140 --> 00:32:40.460 Matt: it's very similar to what we were just talking about before, 00:32:40.660 --> 00:32:41.640 Matt: where it's like these large companies 00:32:42.320 --> 00:32:44.600 Matt: start to focus on optimizations rather than innovation. 00:32:44.770 --> 00:32:46.180 Matt: And it's like, I think we've kind of seen that 00:32:46.190 --> 00:32:48.780 Matt: in the open source space with these component libraries 00:32:49.040 --> 00:32:49.820 Matt: like ShadCN, right? 00:32:50.000 --> 00:32:53.080 Matt: Like, you know, there's now probably, you know, 00:32:53.320 --> 00:32:55.960 Matt: more than we can count copies or clones of ShadCN 00:32:56.140 --> 00:32:58.760 Matt: that, you know, it's like, does the same exact thing, 00:32:59.000 --> 00:33:02.300 Matt: But, you know, maybe it's just, like, maintained slightly differently. 00:33:02.800 --> 00:33:05.840 Matt: Or maybe it's for a different web framework than React or whatnot. 00:33:06.140 --> 00:33:11.220 Matt: But, yeah, I think, you know, assuming the listener is still listening, 00:33:11.700 --> 00:33:14.600 Matt: it's like you're probably wondering, like, okay, like, okay, yeah, 00:33:14.820 --> 00:33:17.340 Matt: you know, like, how are we connecting shadcn to this, like, 00:33:17.460 --> 00:33:18.300 Matt: in enshittification topic? 00:33:18.840 --> 00:33:21.180 Matt: I think really the, in my mind at least, it's, like, 00:33:22.220 --> 00:33:25.020 Matt: pretty similar to, like, what we saw, like, the explosion of, like, 00:33:25.100 --> 00:33:26.340 Matt: sort of bootstrap based sites 00:33:28.200 --> 00:33:29.040 Matt: back in the day. 00:33:29.960 --> 00:33:30.980 Matt: Scott, you probably have more 00:33:31.840 --> 00:33:33.060 Matt: historical knowledge there than I do, 00:33:34.340 --> 00:33:35.360 Matt: since you've been doing web development 00:33:35.500 --> 00:33:37.180 Matt: longer than I have, but I feel like 00:33:37.180 --> 00:33:38.620 Matt: back in the day, it's like when bootstrap was 00:33:39.080 --> 00:33:40.260 Matt: maybe bootstrap 4 or something, 00:33:40.960 --> 00:33:41.700 Matt: when that came out, 00:33:43.060 --> 00:33:45.100 Matt: it's just like every website looked the same. 00:33:45.720 --> 00:33:47.260 Matt: The same could be said with material UI. 00:33:48.080 --> 00:33:49.040 Matt: I remember material UI 00:33:49.220 --> 00:33:50.080 Matt: being really big, 00:33:51.660 --> 00:33:52.700 Matt: and just like every site 00:33:53.040 --> 00:33:54.000 Matt: wanted to use material UI. 00:33:54.160 --> 00:33:56.540 Matt: I mean, heck, I built a few side projects 00:33:56.640 --> 00:33:58.660 Matt: and they all used MaterialUI and they all looked the same. 00:34:00.020 --> 00:34:00.840 Matt: And so I think it's just like, 00:34:01.320 --> 00:34:03.680 Matt: it's interesting that there's these waves of like, 00:34:04.820 --> 00:34:06.280 Matt: I don't know, like sort of waves of, 00:34:07.060 --> 00:34:08.080 Matt: you know, everyone looking the same 00:34:08.320 --> 00:34:10.179 Matt: instead of like this, you know, 00:34:10.260 --> 00:34:12.480 Matt: it just feels like there's like a lack of creativity 00:34:13.860 --> 00:34:16.020 Matt: happening around web design, web development. 00:34:17.780 --> 00:34:19.320 Matt: And I feel like that's just like, 00:34:19.480 --> 00:34:21.780 Matt: that is like an aspect of in-shittification. 00:34:21.899 --> 00:34:23.700 Matt: It's like that's the connection my brain is making. 00:34:25.420 --> 00:34:26.300 Matt: but yeah 00:34:26.530 --> 00:34:26.659 Matt: yeah 00:34:28.760 --> 00:34:30.600 Scott: yeah a lack of creativity is my 00:34:30.860 --> 00:34:32.379 Scott: thought but like it's almost like 00:34:33.660 --> 00:34:34.580 Scott: I'm trying to say 00:34:34.580 --> 00:34:36.560 Scott: I can define this but like a prescriptive 00:34:36.899 --> 00:34:38.639 Scott: solution to solve 00:34:38.919 --> 00:34:40.700 Scott: UI and every UI problem 00:34:41.179 --> 00:34:42.379 Scott: so like UI 00:34:42.639 --> 00:34:43.919 Scott: is used to be very 00:34:44.600 --> 00:34:46.520 Scott: for us at least very creative like how can we solve 00:34:46.550 --> 00:34:48.159 Scott: this problem like what can it do 00:34:48.720 --> 00:34:50.500 Scott: and now it's almost like okay here are the 00:34:50.600 --> 00:34:51.940 Scott: prescribed solutions 00:34:52.780 --> 00:34:55.720 Scott: within the boundaries, pick them. 00:34:56.210 --> 00:34:58.500 Scott: And not like, well, how can we try to break that mold? 00:34:58.550 --> 00:35:03.460 Scott: I remember a project real specific on the design system 00:35:03.680 --> 00:35:07.640 Scott: that I couldn't gain traction on where we had this concept 00:35:07.710 --> 00:35:13.700 Scott: of having sidebars and modals within different applications. 00:35:13.870 --> 00:35:15.860 Scott: I think I called it contextual modals. 00:35:16.100 --> 00:35:19.720 Scott: That was just the idea of having an application 00:35:19.790 --> 00:35:21.720 Scott: and an application and how could we handle that. 00:35:22.760 --> 00:35:28.820 Scott: and like that was so novel i think to sell because it was really just like okay we have an admin 00:35:29.000 --> 00:35:33.100 Scott: toolbar and we need to wrap it around a whole application how do we do it would solve something 00:35:33.260 --> 00:35:39.420 Scott: like that's problem today people would probably just be like make a chrome extension or something 00:35:40.020 --> 00:35:45.640 Scott: like don't have it in the application at all and maybe that's valid but what the purpose of this 00:35:45.660 --> 00:35:53.520 Scott: is just getting at trying to now break the mold of what people have normalized as prescriptive 00:35:53.740 --> 00:35:57.940 Scott: patterns to solve a problem has become harder than it once was. 00:35:59.580 --> 00:36:00.120 Matt: Yeah, for sure. 00:36:00.140 --> 00:36:05.220 Matt: I think there's this huge wave of momentum around that stuff. 00:36:05.420 --> 00:36:09.520 Matt: I mean, you can kind of see it internally at big companies also where a lot of companies 00:36:09.680 --> 00:36:12.640 Matt: invest in design systems because they... 00:36:13.400 --> 00:36:16.560 Matt: I don't know if it's that they don't value UI innovation, 00:36:18.230 --> 00:36:19.680 Matt: but I think it's just like there's this, 00:36:20.960 --> 00:36:25.040 Matt: they just say like, why rebuild the same UI over and over again? 00:36:25.230 --> 00:36:29.660 Matt: And at least they think it's the word same UI 00:36:29.800 --> 00:36:30.940 Matt: or that it's the same UI, 00:36:31.220 --> 00:36:33.680 Matt: but it's like, I think that, 00:36:35.580 --> 00:36:38.580 Matt: I think we've probably equally heard 00:36:38.960 --> 00:36:40.980 Matt: like a large number of questions around like, 00:36:41.440 --> 00:36:44.740 Matt: oh, if we're adopting a design system, does that mean that, you know, designers don't do new work? 00:36:44.980 --> 00:36:49.080 Matt: Or does that mean that engineers don't get to, like, explore different ideas? 00:36:50.340 --> 00:36:52.320 Matt: You know, it's like we're just, like, sort of doing cookie-cutter stuff. 00:36:52.500 --> 00:36:56.320 Matt: And I think it's like, you know, most organizations usually have an answer to that. 00:36:56.420 --> 00:36:57.960 Matt: And like, no, we're not stifling innovation. 00:36:58.180 --> 00:37:00.580 Matt: We're just, like, trying to ensure consistency. 00:37:01.780 --> 00:37:06.540 Matt: But I think it's an interesting, like, it's an interesting sort of, I don't know, 00:37:06.780 --> 00:37:11.060 Matt: momentum to like kind of combat and try to come up with something unique and 00:37:11.240 --> 00:37:14.420 Matt: innovative. Right. And it's like, now we're seeing it in, you know, 00:37:14.560 --> 00:37:16.780 Matt: not necessarily now in the open source space, but like, 00:37:17.080 --> 00:37:19.800 Matt: like it sort of happened like we were talking about bootstrap and material, 00:37:20.020 --> 00:37:24.960 Matt: but it's like, it's just like more like it's common outside of launch 00:37:25.100 --> 00:37:26.380 Matt: corporations also. 00:37:28.600 --> 00:37:29.600 Scott: Yeah. I almost feel like, 00:37:29.720 --> 00:37:34.000 Scott: so I remember being on the design system and having to sell the design system 00:37:34.640 --> 00:37:40.980 Scott: And the big sell for the design system is, hey, it allows us to move faster because we have consistency. 00:37:41.300 --> 00:37:44.380 Scott: Everything looks like the same, so it feels like one cohesive flow. 00:37:45.200 --> 00:37:53.800 Scott: And we have a starting point or we have pieces, and we can now just focus on the data layer so that the UI can be built quicker. 00:37:53.970 --> 00:37:56.340 Scott: And we're not rebuilding the same thing, so we're not wasting time. 00:37:56.860 --> 00:38:06.460 Scott: So it's almost like we did too good of a job as UI engineers at the time to sell the value of having building blocks ready to go. 00:38:07.900 --> 00:38:15.460 Scott: But the bigger problem is just that it seems like the more all design systems have the same pieces. 00:38:16.760 --> 00:38:18.300 Scott: And I try to do this all the time. 00:38:18.440 --> 00:38:27.080 Scott: I try to be like, okay, what should actually be a feature that is common enough that it's its own feature? 00:38:27.220 --> 00:38:28.920 Scott: I always talk about the Command-K palette. 00:38:29.460 --> 00:38:30.740 Scott: Sure, it's just a modal, maybe. 00:38:31.480 --> 00:38:36.580 Scott: But that itself, to me, is something we should just be a feature. 00:38:36.900 --> 00:38:40.280 Scott: You should be able to pull in a select dropdown and maybe a modal. 00:38:40.860 --> 00:38:41.380 Scott: Maybe not. 00:38:41.960 --> 00:38:47.400 Scott: But if you're building something like shadcn, where any new application might use it, 00:38:47.780 --> 00:38:52.460 Scott: I think that's almost like a common enough feature that should become compound enough to be created. 00:38:52.640 --> 00:38:54.400 Scott: But these are the things that I think like. 00:38:56.160 --> 00:38:59.220 Scott: Like shadcn doesn't really care about, like they do this unique thing, right? 00:38:59.800 --> 00:39:04.660 Scott: They're like, oh, here you take the code and download it and you own it now. 00:39:05.120 --> 00:39:09.000 Scott: So they almost they wipe their hands clean of like having to worry about things. 00:39:09.060 --> 00:39:13.600 Scott: I remember people talking about, well, what if like Radix updates and it destroys your UI? 00:39:14.180 --> 00:39:16.300 Scott: They don't care. Doesn't doesn't bother them. 00:39:16.460 --> 00:39:19.640 Scott: I mean, it hasn't, and I don't even think it's gonna. 00:39:20.620 --> 00:39:23.500 Scott: But the point being, now that ShadCN is so big too, 00:39:24.340 --> 00:39:29.380 Scott: likelihood that they have to make sure they don't break all sites becomes higher. 00:39:29.620 --> 00:39:34.820 Scott: But the point being, that simplification, 00:39:34.960 --> 00:39:37.960 Scott: it's almost an easier way than managing and owning dependencies, 00:39:39.080 --> 00:39:42.160 Scott: has just allowed them to be like, oh yeah, just copy the code. 00:39:42.420 --> 00:39:45.320 Scott: But they still get to staple and put their name on it. 00:39:45.840 --> 00:39:59.860 Scott: And it's just very interesting that something as simple as, like, just copy and paste the code, basically, or, you know, run a script that downloads the code into a file for you has become, like, the norm there. 00:40:01.100 --> 00:40:21.680 Scott: And I find that to be also like part of the problem, I guess, where like if everyone just does it this way and then we get used to it, it comes back to what I was just trying to say like slightly earlier is we just like accept that there are certain ways to solve a problem. 00:40:21.780 --> 00:40:30.340 Scott: And those are the only ways. And it just digs this hole deeper that it's harder for us to decide, let's try this new thing or sell this new thing. 00:40:30.810 --> 00:40:38.400 Scott: When people will point to and look at shadcn or I mean, like chakra or material and be like, no, this is the way these people do it. 00:40:38.650 --> 00:40:43.920 Scott: This is what works without using data. They just use the fact that it exists as the data. 00:40:44.320 --> 00:40:51.980 Scott: People in tech sometimes talk a lot about data, but they don't really use the numbers, which is something, a gripe of mine, I guess. 00:40:52.010 --> 00:40:52.460 Scott: Go ahead, Matt. 00:40:53.380 --> 00:40:53.640 Matt: Oh, yeah. 00:40:53.850 --> 00:40:57.660 Matt: I mean, you have to be data-driven in this time and age, of course. 00:40:58.840 --> 00:41:02.260 Matt: No, yeah, I think, yeah, it's interesting, right? 00:41:02.380 --> 00:41:16.580 Matt: I think the distribution format for shadcn maybe was, and I can't say for sure, but maybe the intention there was to say, hey, you have control to customize and to innovate. 00:41:17.620 --> 00:41:26.280 Matt: But I think a lot of people just took it at face value and said, okay, my components that I pulled in used to live in node modules, but now they live in ./ui. 00:41:28.100 --> 00:41:31.680 Matt: And I think most people just say, okay, that's just whatever. 00:41:32.220 --> 00:41:33.040 Matt: That's just a change. 00:41:34.420 --> 00:41:39.880 Matt: I don't think a lot of people maybe saw that as an opportunity to dig in and customize. 00:41:40.110 --> 00:41:43.700 Matt: I think there's still that maybe fear in the back of the head of, 00:41:44.190 --> 00:41:45.520 Matt: if we customize, now we can't upgrade. 00:41:47.180 --> 00:41:48.900 Matt: And they're maybe worried about that. 00:41:49.350 --> 00:41:54.160 Matt: I think in general, maybe this is more of a closing thought 00:41:55.460 --> 00:41:57.580 Matt: than maybe a continuation of the thread. 00:41:58.340 --> 00:42:01.500 Matt: I think really what we've been seeing is like this enshittification 00:42:01.620 --> 00:42:05.380 Matt: trend feels like it could be attributable to 00:42:06.380 --> 00:42:09.860 Matt: you know sort of lack of innovation or creativity and I think there's 00:42:10.020 --> 00:42:13.700 Matt: like a lot of maybe societal change like sort of pressures 00:42:13.860 --> 00:42:17.980 Matt: there to like sort of say yeah just like do the common thing like don't 00:42:18.160 --> 00:42:21.880 Matt: try to explore new avenues you know cultural aspects there and 00:42:22.000 --> 00:42:26.000 Matt: like whether it's like culture as a you know company or culture as 00:42:26.020 --> 00:42:34.700 Matt: like a country um and uh yeah i think it's just like i don't know it's people aren't willing to 00:42:34.820 --> 00:42:40.840 Matt: like sort of rethink or sort of maybe think from like um original like our you know first principles 00:42:41.640 --> 00:42:46.700 Matt: um i think that's like one of the core aspects right it's like even to that point of the blog 00:42:46.840 --> 00:42:54.040 Matt: post of like you know nobody cares it's like i think it's just like people are not um incentivized 00:42:54.040 --> 00:43:00.720 Matt: to apply that effort or to, you know, to like create new things 00:43:01.720 --> 00:43:04.900 Matt: and explore new boundaries and like provide value. 00:43:06.600 --> 00:43:06.980 Matt: Yeah, I don't know. 00:43:07.100 --> 00:43:11.820 Matt: It's a, yeah, I feel like I'm maybe like circling the drain on like that thread, 00:43:12.020 --> 00:43:15.620 Matt: but like maybe there's like some insight I haven't come to yet there. 00:43:15.960 --> 00:43:16.620 Matt: But yeah. 00:43:17.720 --> 00:43:21.560 Scott: I want to bring up the AI stuff too for a second because like, you know, 00:43:22.760 --> 00:43:25.960 Scott: Bricell also has an AI solution. 00:43:26.800 --> 00:43:28.300 Scott: The name is giving me right now. 00:43:28.960 --> 00:43:29.440 Scott: What is it, Matt? 00:43:29.540 --> 00:43:29.800 Matt: V0. 00:43:30.820 --> 00:43:31.500 Scott: V0, yeah. 00:43:32.320 --> 00:43:33.380 Scott: The most obvious name. 00:43:34.200 --> 00:43:37.440 Scott: V0, which basically spits out ChotCNUI for you. 00:43:37.740 --> 00:43:46.020 Scott: And the thing is, well and great, but it's not going to be creative or do something revolutionary and unique. 00:43:46.220 --> 00:43:49.460 Scott: It's only going to print out the patterns it's learned on or it's trained on. 00:43:50.020 --> 00:43:54.980 Scott: So that in itself, AI is another part of the problem. 00:43:55.660 --> 00:43:57.400 Scott: I mean, AI is going to have to be really smart. 00:43:57.680 --> 00:44:00.680 Scott: I don't think we're really that close. 00:44:01.360 --> 00:44:03.520 Scott: I know everything moves exponentially faster than we think, 00:44:03.700 --> 00:44:07.780 Scott: but I don't think we're really that close to an AI coming up with new ideas 00:44:08.730 --> 00:44:10.640 Scott: because that's just not what machine learning is. 00:44:11.320 --> 00:44:15.620 Scott: So we're going to see more of this, as I'll describe it, like soulless. 00:44:16.160 --> 00:44:17.060 Scott: This is how it works. 00:44:17.720 --> 00:44:28.500 Scott: And if people start thinking that AI is providing the right answers all of the time, it's stripping the creativity from what could happen. 00:44:30.120 --> 00:44:31.320 Matt: Yeah, yeah, 100%. 00:44:31.320 --> 00:44:34.820 Matt: Yeah, I think I saw a thread on this, like somewhere on Blue Sky. 00:44:35.060 --> 00:44:38.480 Matt: Maybe I'll try to dig it up and share it in our show notes. 00:44:38.640 --> 00:44:54.340 Matt: But yeah, there was like this thread about like, you know, a lot of companies these days are now asking themselves, you know, not just like, okay, if we're adopting an open source solution, it's like, does this, you know, like how many stars on GitHub does it have? 00:44:54.360 --> 00:44:55.240 Matt: Like how maintained is it? 00:44:55.400 --> 00:45:00.500 Matt: It's like the question they're asking is like, does AI like know about this, right? 00:45:00.560 --> 00:45:04.860 Matt: Like does ChatGPT understand this library, understand how it's used? 00:45:05.780 --> 00:45:08.320 Matt: And then that's like the decision making criteria. 00:45:08.440 --> 00:45:12.560 Matt: right like and it's like we've seen this in like in practice exactly in v0 right it's like the 00:45:13.360 --> 00:45:18.960 Matt: like next.js is going to become the react de facto framework because so many different ai models 00:45:19.200 --> 00:45:22.960 Matt: have been trained on it and it's just like you know sort of it's like a self-fulfilling prophecy 00:45:23.200 --> 00:45:29.000 Matt: where it's like ai recommends you use next yes to spin up a react app you use it it gets trained on 00:45:29.060 --> 00:45:34.160 Matt: that data for the next iteration and so you just keep getting into this like ingrained mode of like 00:45:34.580 --> 00:45:38.400 Matt: that is just going to become the way that we build stuff which is kind of unfortunate right it's like 00:45:39.180 --> 00:45:40.540 Matt: we're losing some of that 00:45:41.600 --> 00:45:42.420 Matt: niche part of 00:45:43.070 --> 00:45:44.940 Matt: open source community that's exploring new ideas 00:45:45.380 --> 00:45:45.620 Matt: because 00:45:47.740 --> 00:45:48.580 Matt: AI models 00:45:49.020 --> 00:45:50.820 Matt: don't know about it, and they're not trained on it, 00:45:50.820 --> 00:45:52.320 Matt: and they're not going to recommend it. 00:45:54.000 --> 00:45:54.620 Matt: Which, again, 00:45:55.140 --> 00:45:56.280 Matt: maybe leads to that in certification. 00:45:57.300 --> 00:45:58.740 Matt: It's like everything becomes 00:45:58.740 --> 00:46:00.680 Matt: the same, and you lose 00:46:01.840 --> 00:46:03.000 Matt: value or creativity, 00:46:03.170 --> 00:46:04.800 Matt: and it's just optimizations rather 00:46:04.800 --> 00:46:05.280 Matt: than innovations. 00:46:06.640 --> 00:46:07.820 Scott: It's so ingrained now. 00:46:07.860 --> 00:46:13.900 Scott: they deprecated create react app because they're just like use a framework at this point at this 00:46:14.040 --> 00:46:20.320 Scott: point it's like sure there's probably there's more but it's like use vite which is pretty much 00:46:20.530 --> 00:46:26.440 Scott: create react app with some speed and processor layers that help it and and then you or it's use 00:46:26.860 --> 00:46:32.720 Scott: next js or maybe there's a tan stack create i'm sure there tan stack start there i'm sure there 00:46:32.740 --> 00:46:39.300 Scott: is but it's like they don't even think that you should start at the bare bones react app anymore 00:46:39.920 --> 00:46:46.260 Scott: because there's so many formulated opinions on where you should start and that takes the 00:46:46.620 --> 00:46:55.660 Scott: creativity out of what you're building right there 00:46:46.620 --> 00:46:55.660 Matt: yeah 100 um yeah yeah i think yeah i mean i don't 00:46:55.720 --> 00:46:59.580 Matt: want to necessarily like cut off any credit any further thoughts you might have on the topic but 00:46:59.600 --> 00:47:03.440 Matt: I think we've kind of like talked about the core things that we had on the 00:47:03.560 --> 00:47:06.440 Matt: docket for today. Yeah. 00:47:06.720 --> 00:47:10.500 Matt: I, I, I, I fear that our episode is going to be very like doomer. 00:47:11.340 --> 00:47:12.640 Matt: You know, like I think this episode, 00:47:12.690 --> 00:47:15.980 Matt: the previous episode have been very like sort of doom or like negative outlook. 00:47:18.400 --> 00:47:19.120 Matt: Yeah. Yeah. 00:47:19.840 --> 00:47:22.520 Matt: We need to think about how we can get more maybe optimism into, 00:47:23.700 --> 00:47:26.220 Matt: into these discussions, but I think it's interesting for like, 00:47:26.360 --> 00:47:27.420 Matt: it's worth us digging into them. 00:47:29.300 --> 00:47:41.100 Scott: I don't know if, yeah, it may be like, maybe it seems that doomish in the sense that like we enjoyed some of these things and they're being taken away. 00:47:43.640 --> 00:47:52.260 Scott: But I don't think that's, I don't think that like that doesn't mean we can't still be creative and still try to push these boundaries in our own time. 00:47:52.360 --> 00:48:06.100 Scott: I did tell you as a silver lining, I've been trying to create my own library, and I want to create a Command K palette as a component, and maybe that'll be a way to get some momentum behind something like that. 00:48:06.430 --> 00:48:11.040 Scott: I think that there is opportunity for new patterns to continue to emerge. 00:48:12.800 --> 00:48:15.580 Scott: They will become maybe standardized at some point. 00:48:16.640 --> 00:48:17.920 Scott: And new things will definitely happen. 00:48:18.020 --> 00:48:25.360 Scott: And it's just that we are no longer trying to focus part of our job to discover them, which is maybe a little bit of a bummer. 00:48:25.460 --> 00:48:28.840 Scott: But I wouldn't I wouldn't say it's as much doomerism. 00:48:29.700 --> 00:48:36.200 Scott: I do get that, like, it really it's AI that is throwing the wrench into, like, what will our job become? 00:48:36.700 --> 00:48:37.860 Scott: It's very interesting. 00:48:39.000 --> 00:48:40.780 Scott: I did have some other thoughts on enshittification. 00:48:41.120 --> 00:48:47.660 Scott: I can't remember them right now, but it's very interesting that it now seems like corporations are like, oh, yeah, use AI and use it as much as possible. 00:48:48.180 --> 00:48:49.220 Scott: and let's move faster. 00:48:50.820 --> 00:48:51.660 Scott: That is just definitely 00:48:53.500 --> 00:48:54.440 Scott: an interesting future. 00:48:55.580 --> 00:48:57.960 Matt: Yeah, I think we probably have a whole other episode 00:48:57.980 --> 00:48:59.180 Matt: to dig into in terms of like, 00:49:00.740 --> 00:49:02.320 Matt: there's this sort of, 00:49:04.500 --> 00:49:06.120 Matt: yeah, I don't know the word for it, 00:49:06.200 --> 00:49:08.540 Matt: but like sort of a spectrum of like companies 00:49:08.880 --> 00:49:11.940 Matt: where they say that they're AI first or AI native. 00:49:12.180 --> 00:49:13.760 Matt: And I think there's sort of like, 00:49:13.980 --> 00:49:15.860 Matt: good and bad outcomes of that. 00:49:16.900 --> 00:49:19.800 Matt: And yeah, maybe we save that for the next episode. 00:49:20.460 --> 00:49:20.660 Matt: We'll see. 00:49:22.140 --> 00:49:22.320 Scott: Yeah. 00:49:23.480 --> 00:49:26.920 Scott: But like overall, this concept of enshittification, 00:49:27.340 --> 00:49:31.160 Scott: like I think there's like a lot that makes it occur. 00:49:31.640 --> 00:49:35.220 Scott: But I guess my final thoughts on it is, 00:49:35.340 --> 00:49:40.060 Scott: is it's a lot of the times it's like a platform gets so big 00:49:40.360 --> 00:49:44.100 Scott: and it's served its purpose that it stops innovating. 00:49:44.400 --> 00:49:48.540 Scott: or it's like you said, kind of stops, loses its mission of like, 00:49:48.800 --> 00:49:52.920 Scott: what were we here for in the first place? Like Facebook, um, 00:49:53.420 --> 00:49:57.940 Scott: for a long time added features, but these apps and these, 00:49:58.560 --> 00:50:02.360 Scott: especially the bigger apps, I think that they're where it's heading. 00:50:03.560 --> 00:50:05.840 Scott: Is there just be going to try to become everything apps? 00:50:06.040 --> 00:50:10.040 Scott: Like China has everything apps and like Facebook is a good example of that 00:50:10.140 --> 00:50:12.620 Scott: because it's a place to like connect with friends, 00:50:13.260 --> 00:50:19.260 Scott: but now it's a messenger app too but then also they own a messenger app but it's also got stories like 00:50:19.700 --> 00:50:25.460 Scott: um snapchat but also they own instagram and that adds stories like snapchat it's they're they're 00:50:25.540 --> 00:50:35.080 Scott: like these everything apps um you can do what you do on twitter or um blue sky on on there by posting 00:50:35.360 --> 00:50:42.180 Scott: you can do all of it so i think i mean i don't think it's the greatest but i think they're going 00:50:42.200 --> 00:50:52.100 Scott: to start to try to have currencies or um find ways where they can just be an app that's a super app 00:50:52.560 --> 00:50:57.560 Scott: um and i'm not saying that that's the right answer i actually don't think that's the right answer 00:50:57.800 --> 00:51:03.780 Scott: because that leads us down a path of like them have um these companies have this maybe this is 00:51:03.860 --> 00:51:09.980 Scott: doomerism again but companies having so much data about us that um they can start to market up prices 00:51:10.000 --> 00:51:12.220 Scott: based on the things that they know we need. 00:51:15.880 --> 00:51:20.980 Scott: But also, the point being that when they run out of innovation, 00:51:22.700 --> 00:51:25.900 Scott: they start to just saying, what's a whole new market we can get into? 00:51:26.310 --> 00:51:32.440 Scott: Or they don't look at how can we further serve the purpose of our original intent. 00:51:33.140 --> 00:51:35.240 Scott: And that's where these apps kind of have their downfall. 00:51:36.520 --> 00:51:40.500 Scott: Trying to think of like an example like Spotify I had from the beginning. 00:51:40.570 --> 00:51:44.580 Scott: I don't think that we're going to see it like have a downfall, 00:51:44.760 --> 00:51:47.620 Scott: but it is disheartening that you pay for this service. 00:51:48.560 --> 00:51:49.100 Scott: It's gone up. 00:51:49.280 --> 00:51:53.400 Scott: I've had it for probably like 12, 15 years. 00:51:53.560 --> 00:51:54.440 Scott: It's a long time now. 00:51:55.040 --> 00:51:57.160 Scott: And, you know, it's not too, too much more. 00:51:57.260 --> 00:52:03.260 Scott: It's like $13 a month from its original $7, $8, $10. 00:52:03.900 --> 00:52:08.300 Scott: But at the same time, it's very disheartening to see it add ads in it. 00:52:08.740 --> 00:52:12.600 Scott: It is adding video, so they are still trying to innovate on it. 00:52:13.000 --> 00:52:14.240 Scott: I talked about the jam feature, 00:52:14.900 --> 00:52:17.420 Scott: but it's when an application becomes a place 00:52:17.460 --> 00:52:20.220 Scott: that just seems to be losing its quality. 00:52:20.780 --> 00:52:22.640 Scott: Like Amazon website, for example, 00:52:24.440 --> 00:52:27.200 Scott: was the best place to get products cheap, 00:52:27.580 --> 00:52:28.720 Scott: and then they started raising their prices, 00:52:28.960 --> 00:52:30.120 Scott: and now it's just about convenience. 00:52:30.900 --> 00:52:32.880 Scott: And then they stopped guaranteeing two days. 00:52:33.320 --> 00:52:46.820 Scott: And then all of, and this is an episode in itself that we still want to talk about, but then all of the product display pages became about monetizing every product that they own and less about having a good quality clarity around the product that you're interested in. 00:52:47.440 --> 00:53:02.540 Scott: All of that is like, while Amazon is still like a pretty good place to shop, it shows you that like it's a downward trend of like they cared really much about the customer until they had all the customers they could pretty much have. 00:53:03.160 --> 00:53:09.320 Scott: And now they have these customers in their pocket a little bit with things like, you know, people are paying yearly for Amazon subscriptions. 00:53:10.440 --> 00:53:18.060 Scott: So now they just work a little less important about the customer and more about like, how can we sell more, make more money and monetize that? 00:53:19.480 --> 00:53:32.220 Scott: So if they pulled back a little bit and built a UI, and I guess it's very UI centric is why we talked about Shotsian, and built a UI that helped the customer almost more like theater to find the thing that they needed. 00:53:32.500 --> 00:53:35.980 Scott: I hear so many people say like, oh, yeah, Amazon sucks when I search. 00:53:36.200 --> 00:53:39.080 Scott: I can't find the thing I need for multiple pages. 00:53:40.510 --> 00:53:47.040 Scott: Or I know that I and some friends of mine are like, oh, you need to look up the reviews and make sure the reviews aren't faked. 00:53:47.540 --> 00:53:54.080 Scott: Or you need to look at this other website that exists to look at the price trend and see how much they've raised or lowered the price. 00:53:54.440 --> 00:53:59.460 Scott: When those kinds of things start infiltrating your app, it's no longer about giving the user the best experience. 00:53:59.980 --> 00:54:01.700 Scott: And it's more about just monetizing. 00:54:02.220 --> 00:54:05.320 Scott: And that's a downward trend that you don't want to be a part of. 00:54:07.380 --> 00:54:10.540 Matt: Capitalism just leads towards this. 00:54:12.280 --> 00:54:16.800 Matt: Trying to get the most out of, or maybe extreme capitalism, 00:54:17.180 --> 00:54:19.360 Matt: trying to get the most value out of every single customer 00:54:20.500 --> 00:54:25.980 Matt: just leads to this case where it's not a great customer experience, 00:54:26.240 --> 00:54:29.400 Matt: but it's a good enough experience that they're going to convert. 00:54:30.060 --> 00:54:31.060 Matt: and the company knows that. 00:54:32.720 --> 00:54:36.220 Matt: And so they build up a moat that is like, 00:54:36.460 --> 00:54:38.060 Matt: yeah, exactly like your convenience 00:54:38.260 --> 00:54:41.260 Matt: or you've shopped on Amazon for long enough 00:54:41.420 --> 00:54:43.860 Matt: that Amazon knows that you're not going to go 00:54:43.920 --> 00:54:45.460 Matt: to some other site to start shopping at 00:54:45.700 --> 00:54:47.500 Matt: just because all the convenience 00:54:47.660 --> 00:54:48.460 Matt: that they've already built up. 00:54:49.200 --> 00:54:51.340 Matt: And so they just know that they don't need 00:54:51.440 --> 00:54:52.760 Matt: to make it 10x better. 00:54:52.960 --> 00:54:58.280 Matt: They just need to make it not 10x worse or something. 00:54:59.580 --> 00:55:03.800 Scott: or just better than the next competitor in some form or fashion that doesn't have to be 00:55:04.000 --> 00:55:08.140 Scott: in the facet of ui and i also just want to point out we know that amazon you know makes their money 00:55:08.340 --> 00:55:16.260 Scott: through aws and that's another example of like a really bad ui it's not good uh but it's it's 00:55:16.260 --> 00:55:23.240 Scott: like cheaper than most and they're able to do that because they have that um capital essentially 00:55:23.260 --> 00:55:29.780 Scott: to provide services and they did it at the right time um but but it's again like 00:55:30.900 --> 00:55:36.100 Scott: i've been told they have like seven or eight design systems and when you look at aws it shows 00:55:36.620 --> 00:55:41.980 Scott: half the page is white then all of a sudden it's dark mode and then it's just not that that's not 00:55:41.980 --> 00:55:48.660 Scott: a that's a no-no for users like we need to uh i don't know why there's a design system at that 00:55:48.680 --> 00:55:53.400 Scott: point if the themes are completely different and the whole point is consistency having seven or 00:55:53.410 --> 00:56:01.700 Scott: eight of that is not a strong argument to me but um i digress all right yeah i think we've i think 00:56:01.720 --> 00:56:07.100 Matt: we've probably you know talked about this topic uh a decent amount maybe more than we maybe 00:56:07.420 --> 00:56:12.420 Matt: originally intended um we'll come back to it though yeah yeah we might yeah we might come back to in 00:56:12.440 --> 00:56:19.900 Matt: the future. Scott, what's new with you? What's a new and exciting thing have you been looking at 00:56:19.920 --> 00:56:26.040 Matt: over the past few weeks that you want to talk about? 00:56:19.920 --> 00:56:26.040 Scott: Yeah. So before I talk about that, we did 00:56:26.120 --> 00:56:31.460 Scott: an experiment from our last podcast about using and not using AI. We're going to hold off on 00:56:31.560 --> 00:56:37.480 Scott: talking about that until we have Dillon back. Dillon wasn't able to make it today. So we will talk 00:56:38.520 --> 00:56:41.120 Scott: about that in a future episode. 00:56:42.200 --> 00:56:43.880 Scott: The reason why I said that beforehand 00:56:45.500 --> 00:56:49.880 Scott: is because I started to dig into Avante for any of them. 00:56:49.960 --> 00:56:51.360 Scott: I've been watching some videos on it, 00:56:51.480 --> 00:56:53.020 Scott: and it is more copilot, 00:56:53.260 --> 00:56:55.780 Scott: sorry, more cursor AI-like than I thought, 00:56:56.000 --> 00:56:57.720 Scott: and I started to configure it better, 00:56:57.840 --> 00:56:58.960 Scott: and it has been stronger for me. 00:56:59.460 --> 00:57:01.020 Scott: And I want to say that it's helped me 00:57:02.360 --> 00:57:06.460 Scott: in my cop-out answer in my Golang quest 00:57:06.480 --> 00:57:07.980 Scott: of becoming a better Go engineer. 00:57:09.140 --> 00:57:11.900 Scott: But I've been using that, 00:57:11.950 --> 00:57:14.460 Scott: but also getting stronger at Go, 00:57:14.660 --> 00:57:15.900 Scott: where I'm starting to feel a lot better 00:57:16.140 --> 00:57:17.760 Scott: about my Go skill set. 00:57:18.340 --> 00:57:19.640 Scott: One of the things I haven't done yet, 00:57:19.680 --> 00:57:21.560 Scott: Bo was telling Matt actually last weekend 00:57:22.500 --> 00:57:26.020 Scott: that I'm really into just starting a Go project, 00:57:26.960 --> 00:57:28.560 Scott: or what I want to do is I want to build 00:57:28.870 --> 00:57:30.340 Scott: a backend service in Node, 00:57:30.980 --> 00:57:33.840 Scott: and then, you're looking at me funny, Matt. 00:57:34.240 --> 00:57:35.880 Scott: I want to build a backend service in Node 00:57:35.900 --> 00:57:37.060 Scott: and then refactor it to go. 00:57:37.740 --> 00:57:41.080 Scott: I think that would be a great exercise for me 00:57:41.200 --> 00:57:44.260 Scott: to continue to grow my backend skill set. 00:57:45.000 --> 00:57:46.980 Scott: So I've been getting a lot more comfortable, though, 00:57:47.100 --> 00:57:49.120 Scott: with the language and just understanding what's possible 00:57:49.290 --> 00:57:50.140 Scott: and how everything works. 00:57:50.720 --> 00:57:52.820 Scott: And basically the differences between that and JavaScript. 00:57:53.160 --> 00:57:54.860 Scott: So it's been a ride a little bit, 00:57:55.300 --> 00:57:58.540 Scott: but I'm starting to feel confident in the language. 00:57:58.760 --> 00:58:00.540 Scott: It's been maybe three months. 00:58:00.820 --> 00:58:04.200 Scott: So that's a real big win for me. 00:58:05.420 --> 00:58:06.800 Scott: Matt, what's up with you? 00:58:09.580 --> 00:58:10.600 Matt: Oh, yeah, that's good call-outs. 00:58:11.200 --> 00:58:13.860 Matt: I've been digging more and more into, 00:58:14.180 --> 00:58:15.280 Matt: so, like, even though, yeah, 00:58:15.340 --> 00:58:16.480 Matt: our last episode talked about, like, 00:58:16.720 --> 00:58:17.780 Matt: not using AI as much, 00:58:17.900 --> 00:58:19.340 Matt: I've actually been digging into using AI 00:58:19.380 --> 00:58:19.960 Matt: a little bit more. 00:58:21.260 --> 00:58:21.660 Scott: Womp Womp. 00:58:23.920 --> 00:58:26.160 Matt: Specifically using the, like, 00:58:26.160 --> 00:58:28.740 Matt: the Composer agent feature in Cursor. 00:58:29.840 --> 00:58:31.180 Matt: But I've also been testing out this thing, 00:58:32.200 --> 00:58:32.860 Matt: like, briefly. 00:58:34.040 --> 00:58:38.420 Matt: That's called, what is it, Codename Goose, 00:58:39.380 --> 00:58:44.920 Matt: which is apparently sort of a thing open source by block, I think, 00:58:45.340 --> 00:58:45.840 Matt: maybe by Square. 00:58:48.060 --> 00:58:50.400 Matt: And it's this pretty cool thing where it, like, 00:58:51.900 --> 00:58:55.740 Matt: it, like, infers functionality that's available on your computer 00:58:56.820 --> 00:58:59.300 Matt: and, like, can, like, sort of build up its own functions 00:58:59.600 --> 00:59:02.340 Matt: and, like, agentic sort of resources or tools. 00:59:03.000 --> 00:59:06.540 Matt: It's using the model context protocol from Anthropic, which is pretty neat. 00:59:08.320 --> 00:59:18.060 Matt: But it's as this maybe slightly scary onboarding where it's just like you put it up and then it starts spamming you with things saying like, 00:59:18.060 --> 00:59:20.620 Matt: oh, do you want to allow this thing access to your files? 00:59:20.700 --> 00:59:22.140 Matt: Do you want to allow this thing access to the network? 00:59:22.260 --> 00:59:24.280 Matt: Do you want to allow this thing access to whatever? 00:59:24.480 --> 00:59:28.160 Matt: And it's a little bit scary, but I kind of YOLO'd it. 00:59:28.500 --> 00:59:29.360 Matt: And it's pretty neat. 00:59:30.360 --> 00:59:32.220 Matt: It could do some pretty complex stuff. 00:59:34.020 --> 00:59:35.760 Matt: so I'll have a link to that in the show notes 00:59:35.920 --> 00:59:37.820 Matt: but it's something I've been messing around with 00:59:38.900 --> 00:59:41.340 Matt: along with Cursor's agent stuff 00:59:41.960 --> 00:59:42.820 Matt: and then also I've been leveraging 00:59:44.880 --> 00:59:47.780 Matt: this MacWhisper or SuperWhisperer I think is what I installed 00:59:48.400 --> 00:59:50.360 Matt: that allows me to talk to Cursor 00:59:50.700 --> 00:59:53.560 Matt: and dictate the prompt to Cursor 00:59:53.560 --> 00:59:55.180 Matt: and then Cursor then goes off and does the work 00:59:55.700 --> 00:59:56.920 Matt: which is actually really neat 00:59:58.200 --> 01:00:01.360 Matt: I won't even have to have a keyboard, I can just talk to my computer 01:00:01.380 --> 01:00:03.100 Matt: and have it do the work for me. 01:00:05.040 --> 01:00:05.660 Scott: That's pretty cool. 01:00:06.440 --> 01:00:07.720 Scott: I'm excited to see that link. 01:00:08.040 --> 01:00:10.080 Scott: I haven't checked anything like that out. 01:00:10.860 --> 01:00:13.760 Scott: I do want to make one note about, you know, 01:00:14.180 --> 01:00:16.480 Scott: you kind of have me thinking about this. 01:00:16.480 --> 01:00:18.480 Scott: We've been a little bit doomish about AI, 01:00:18.660 --> 01:00:21.000 Scott: but at the same time, you've been leaning in. 01:00:21.120 --> 01:00:22.980 Scott: And I actually, I do think it's the right thing 01:00:23.080 --> 01:00:23.900 Scott: to continue to lean in. 01:00:24.420 --> 01:00:26.160 Scott: I just think that as a software engineer, 01:00:27.380 --> 01:00:31.340 Scott: my advice to you, was it that great guru 01:00:31.360 --> 01:00:32.520 Scott: My advice to you. 01:00:33.420 --> 01:00:34.380 Scott: My advice is to you. 01:00:35.190 --> 01:00:37.520 Scott: To continue, Matt has no clue what I'm talking about. 01:00:37.520 --> 01:00:38.960 Scott: I made a reference that just dated me. 01:00:40.200 --> 01:00:41.940 Scott: Anyway, gang star guru. 01:00:42.360 --> 01:00:50.420 Scott: Anyway, what I would recommend is that, yeah, I think it's definitely fine to use AI, 01:00:50.630 --> 01:00:55.140 Scott: but you want to make sure when you prompt the AI that you don't just spit everything out 01:00:56.490 --> 01:00:59.300 Scott: and copy and paste it and not understand how it works. 01:00:59.560 --> 01:01:05.240 Scott: You need to find one thing about it that you can understand a little better. 01:01:05.560 --> 01:01:06.640 Scott: I'm really speaking about languages. 01:01:06.820 --> 01:01:07.600 Scott: Maybe you don't know that well. 01:01:08.400 --> 01:01:11.180 Scott: If you know JavaScript well, or sorry, any language well, 01:01:11.320 --> 01:01:12.400 Scott: and you're using it to spit that out, 01:01:13.320 --> 01:01:15.080 Scott: it's a lot easier for you to just look at it and say, 01:01:15.320 --> 01:01:16.420 Scott: okay, I know what it's doing. 01:01:16.700 --> 01:01:17.320 Scott: That makes sense. 01:01:17.660 --> 01:01:18.340 Scott: Here's an optimization. 01:01:18.980 --> 01:01:19.800 Scott: But when you don't know it, 01:01:19.800 --> 01:01:23.000 Scott: you want to make sure that you can break down at least one piece of it 01:01:23.180 --> 01:01:26.460 Scott: and learn from it so that you can continue to keep up with it. 01:01:27.500 --> 01:01:45.700 Scott: That's the route I have been taking, but also just like doing some things and some percentage capacity to just remind yourself, you know, how to solve a problem because it's so easy to just be like, wait, how would I do this? 01:01:46.000 --> 01:01:48.760 Scott: And then feel like you have this imposter syndrome that you don't know how to solve a problem. 01:01:49.400 --> 01:02:08.360 Scott: So always take on easy tasks and small tasks or think about what you're asking it to do and how you might solve it before you just take AI's word for it so that you're not just mindlessly connecting dots and you don't feel like you are worthless. 01:02:09.580 --> 01:02:10.380 Scott: Matt, you have thoughts on that? 01:02:10.980 --> 01:02:11.300 Matt: Oh, yeah. 01:02:11.940 --> 01:02:12.860 Matt: Maybe a little bit ironic. 01:02:13.200 --> 01:02:15.320 Matt: I think I've been maybe doing the opposite. 01:02:15.620 --> 01:02:17.400 Matt: Well, in certain cases. 01:02:17.660 --> 01:02:17.800 Matt: Yeah. 01:02:18.000 --> 01:02:24.380 Matt: There are certain projects where I'm like, yes, I want to be more involved and I want to know what the code's doing. 01:02:24.430 --> 01:02:25.740 Matt: I want to write it myself, maybe. 01:02:26.580 --> 01:02:29.420 Matt: But other cases where it's just like I've been YOLO-ing it. 01:02:29.720 --> 01:02:37.040 Matt: And I was actually talking to a friend of the pod that we both know, Joe, about this sort of topic. 01:02:37.150 --> 01:02:44.780 Matt: And he was saying that also that he's seen the trend of people want to be more involved, more hands-on with the AI recommendations, like the agentic sort of updates. 01:02:46.180 --> 01:02:48.400 Matt: And I was like telling them like, actually, I just want the opposite. 01:02:48.490 --> 01:02:56.120 Matt: I just want to like tell something like go do this task and then not check back on it until it's like done. 01:02:56.530 --> 01:03:05.120 Matt: Basically, like I want us to get to the point where I can like, you know, just tell tell an agent to like do a change and then have it deploy it and validate it and whatnot. 01:03:05.480 --> 01:03:08.200 Matt: And like I have, you know, I've just been sitting back doing nothing. 01:03:10.459 --> 01:03:11.780 Matt: So so I don't know. 01:03:11.900 --> 01:03:14.760 Matt: It's a maybe different perspectives there. 01:03:14.920 --> 01:03:16.080 Matt: But yeah. 01:03:18.340 --> 01:03:20.740 Scott: Yeah, well, you write code every day of the week. 01:03:20.940 --> 01:03:23.220 Scott: So maybe for you, you feel like you're still in it. 01:03:24.800 --> 01:03:27.220 Scott: But yeah, I know I do the same thing. 01:03:27.460 --> 01:03:29.400 Scott: Like, I don't think that it's wrong to do that. 01:03:29.430 --> 01:03:30.480 Scott: I think you just need to. 01:03:32.680 --> 01:03:37.340 Scott: One of the things I'm big on is taking time out of your day to make sure you're learning something new. 01:03:39.060 --> 01:03:41.680 Scott: Whether that's something that is completely unrelated to work. 01:03:42.620 --> 01:03:43.860 Scott: But it just needs to excite you. 01:03:44.320 --> 01:03:45.160 Scott: but it helps you with your career, 01:03:45.980 --> 01:03:47.940 Scott: which is one of the reasons why I use Neovim. 01:03:47.940 --> 01:03:50.100 Scott: I know I've talked about this before probably, 01:03:52.020 --> 01:03:54.880 Scott: but you want to make sure that you're getting something 01:03:56.660 --> 01:03:58.260 Scott: out of the work you do 01:03:59.020 --> 01:04:00.180 Scott: and that it's coming back to you 01:04:00.320 --> 01:04:02.240 Scott: that's growing your career in any way, shape, or form 01:04:02.320 --> 01:04:04.140 Scott: that makes you feel stronger. 01:04:04.560 --> 01:04:06.600 Scott: I think that if you want to... 01:04:06.640 --> 01:04:08.120 Scott: I also do what you do, Matt. 01:04:08.120 --> 01:04:13.820 Scott: I eat stuff in the AI and just take it and paste it and go. 01:04:13.860 --> 01:04:17.680 Scott: sometimes you work on tasks that aren't that exciting and sometimes that's what you do but 01:04:18.240 --> 01:04:24.760 Scott: you want to try to make sure that you are aware um and that something engages you and makes you feel 01:04:24.960 --> 01:04:30.640 Scott: like like you are you are solving problems too so you still have that sense of ownership or 01:04:31.160 --> 01:04:37.859 Scott: sense of completeness i guess when you're done with with your work 01:04:31.160 --> 01:04:37.859 Matt: yeah for sure all right should 01:04:37.880 --> 01:04:39.340 Matt: Should we call it there? 01:04:40.400 --> 01:04:40.820 Scott: All right. 01:04:41.600 --> 01:04:41.840 Scott: Peace! 01:04:44.400 --> 01:04:45.400 Matt: See you guys next week. 01:04:46.640 --> 01:04:46.960 Take care.