Skip to content
Friends on the frontend (Friends) episode artwork

Mar 14, 2025 1 hr 47 min

Friends on the frontend (Friends)

from The Changelog: Software Development, Open Source

Adam's friend on the frontend, John Long joins the show to explore his usage of AI, design tools and the stack he prefers. We talk Next.js vs Rails, maintaining open source, building websites with Framer, their mutual love for Figma, and more.

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

Friends on the frontend (Friends)

0:00 1:47:30

Chapters

  1. 0:00 Let's talk!
  2. 0:38 Sponsor: Retool
  3. 3:39 Start the show!
  4. 5:40 Consuming generated content
  5. 9:24 Ai is similar in quality and mistakes
  6. 10:58 What does John do?
  7. 12:45 Rails has changed so much
  8. 14:10 Next.js lover?
  9. 16:25 Rails 3 release date
  10. 19:52 Let's talk Zest
  11. 23:55 Tools of a designer/frontender
  12. 29:54 Sponsor: Augment Code
  13. 33:03 Coming off a Framer high
  14. 38:43 Having a Frontend background
  15. 48:57 Using AI in design processes
  16. 57:31 What tools are in your stack?
  17. 1:09:14 Sponsor: Temporal
  18. 1:11:36 Figma is awesome
  19. 1:17:47 Generated.photos
  20. 1:19:55 Createing logos with ChatGPT
  21. 1:23:59 Iconic.app from Orman Clark
  22. 1:27:05 The state of Zest Icons
  23. 1:34:03 Heroicons!
  24. 1:36:18 1.5 vs 2 pixel lines
  25. 1:38:58 Devs working with Design
  26. 1:40:54 Figma School
  27. 1:43:44 Wrapping up
  28. 1:45:47 That's all friends!
of matches

Transcript Auto-generated, may contain errors

Welcome to Change Login Friends, your weekly talk show about tools on the front end. A big thank you to our friends over at fly.io. Deploying 35 regions, fork off VMs like their processes, and so much more. Learn more at fly.io. Okay, let's talk.

Well, friends, I'm here with a good friend of mine, David Hsu, the founder and CEO of Retool. So David, I know so many developers who use Retool to solve problems, but I'm curious, help me to understand the specific user, the particular developer who is just loving Retool. Who's your ideal user? Yeah, so for us, the ideal user of Retool is someone whose goal, first and foremost, is to either deliver value to the business or to be effective. Where we candidly have a little bit less success is with people that are extremely opinionated about their tools. If, for example, you're like, hey, I need to go use WebAssembly. And if I'm not using WebAssembly, I'm quitting my job. You're probably not the best Retool user, honestly. However, if you're like, hey, I see problems in the business. I want to have an impact. I want to solve those problems. Retool is right up your alley. And the reason for that is Retool allows you to have an impact so quickly. You could go from an idea. You go from a meeting like, hey, you know, this is an app that we need to literally having the app built in 30 minutes, which is super, super impactful on the business. So I think that's the kind of partnership or that's the kind of impact that we'd like to see with our customers. You know, from my perspective, my thought is that, well, Retool is well known.

Retool is somewhat even saturated. I know a lot of people who know Retool, but you've said this before. What makes you think that Retool is not that well known? Retool today is really quite well known amongst a certain crowd. Like I think if you had to pull like engineers in San Francisco or engineers in Silicon Valley, even, they could probably get like a 50, 60, 70% recognition of Retool.

I think where you're less likely to have heard of Retool is if you're a random developer at a random company in a random location like the Midwest, for example, or like a developer in Argentina, for example, you're probably less likely. And the reason is I think we have a lot of really strong word of mouth from a lot of Silicon Valley companies like the Brexit, Coinbase, DoorDash, Stripes, et cetera, of the world. There's a lot of chat. Airbnb is another customer. Nvidia is another customer. There's a lot of chatter about Retool in the Valley. But I think outside of the Valley, I think we're not as well done. And that's one goal of ours is to go change that.

Well, friends, now you know what Retool is. You know who they are. You're aware that Retool exists. And if you're trying to solve problems for your company, you're in a meeting, as David mentioned, and someone mentions something where a problem exists. And you can easily go and solve that problem in 30 minutes, an hour, or some margin of time that is basically a nominal amount of time. And you go use Retool to solve that problem. That's amazing. Go to Retool.com and get started for free or book a demo. It is too easy to use Retool. And now you know. So go and try it. Once again, Retool.com. Okay, John. So I saw a post on LinkedIn.

from you three months ago. This is kind of like maybe old news, technically, especially in the world of AI. And it has to do with AI. As many know, they may know you as a product designer, a developer, a front-ender, maybe even just a designer, you know, not just you, but just a designer. But design, front-end, all that good stuff. And you used AI to do some cool stuff. And I thought, hey, you're a friend. Let's get you on this show and talk like friends do about fun stuff. Yeah, sure. There you go.

I guess in my daily work, I've been doing kind of a mix of product design and marketing work. And, you know, I've been working with a local startup, Opine. They kind of make software for salespeople. And, you know, we're a handful of people. Everybody's sort of sharing roles, doing, you know, spread across multiple things. And so we use AI a lot for different things. And basically for the documentation side of things.

And then the marketing side of things, when I'm writing content, I created a custom GPT just through chat. GPT is like standard interface, uploaded all the content I could find from the website and LinkedIn. So it knows about everybody that's involved and then started writing marketing content. And eventually we did the documentation site as well with chat GPT. So what do you think about that? Are you obviously you're for it because you did it, but like in retrospect, was it, you know, in your job, I'm sure it made your job a lot easier, but, you know, as a human being consuming content generated.

How does that make you feel? Yeah, I mean, I would describe it, if you haven't done it, as very much like working with someone who is fast at generating drafts, right? Like, and Chapit GPT writes better marketing content than, dare I say it, most marketing people I know that are kind of doing that in their day job. So if you're not using it kind of as your drafting system, and I think there are certainly times where you want to... You want to slow down and just write out what you're doing. But think about your positioning and get all of those things figured out. Those kinds of documents on the marketing side, you need to spend time writing that. Write about your angle, all that kind of thing. But then once you have that, why not create a custom GPD and let it write your drafts for you? You can give instructions about the style that you want it to compose in. For me, I'm one of those people that can come to the blank page and...

I don't know what to do kind of thing. And it's not that I can't get there. It's just, it takes me like 40 minutes to write my first rough draft. So like chat GPT will shortcut that for me so that in five minutes I have my draft, I go back and forth with it a little bit. And then I jump in there and edit it myself because, you know, I can't quite get what I want it to do. So there's a lot of sort of back and forth. And I have, I have some of these chats where, you know, it depends on what I was doing. I think the.

The documentation one, the most epic thing that I did was to do a full walkthrough as kind of the... the documentation. So it was a full walkthrough of the product. And that was a ton of back and forth with ChatGPT. First, I had to write the outline for what I wanted to do. And then I took the outline major point by major point and asked it to write the content. I would review it. I'd say, no, that's too long, make it shorter. I would tell it when it got things wrong because it's really good at reading in between the lines, but sometimes it will say things that aren't true about your product too. So you have to hold it accountable.

But I mean, I don't think that's any different from working with a writer on things. So it's really interesting. My perspective on the GPTs is like they're here to stay, whether we like it or not. But learning how to work with them, whatever your role is, if you write marketing content in your day job, if you're a developer that needs to write marketing content, whatever it is, they can definitely increase your throughput, be the way that I'd describe it. And you can do things in days now with GPTs that used to take weeks to pull off when it comes to writing. So you think the writing is similar to the drafts you might get with a staff writer or something like that? And you think that...

Some of the they call them hallucinations, you know, I hallucinates or it makes things up. Essentially, they call them the terminology has been hallucinations. So it hallucinates. You think those are similar to the way, you know, a staff writer, a human writer would be not fully correct on everything because they're trying to massage words and make it sound cool. It's not much different than that. Yeah, I would say it's not much different than that. You might have to.

So another usage that your listeners might be curious about is I've been using it to write cover letters for job applications, right? It's been in between jobs. And it is similar in the sense that sometimes it gets things wrong and sometimes it repeatedly gets things wrong. I think humans are better at not getting repeatedly things wrong than ChatGPT is, would be the way that I would describe it. But it's probably better at writing than maybe 80% of the people out there.

So I don't know. I don't know if those numbers are exact, but it does write remarkably good copy. How much, so you're a product designer. You teeter in marketing a little bit, as you've said, in your most recent role at Opine. We've had you on the show years ago, Creative Serve, an open source project that written in Ruby, I think a lot of people may know of. You've designed icon sets. I think, is it open source? Is it open source?

200 of them are. 200 of them are. Prior contributor to Radiance CMS. I don't know. Did you start it or did you contribute to it? Yeah, I created it. Creator of. I mean, geez, I'm forgetting. I mean, I'm so old. I'd forgotten what you've made. You've been around doing this stuff and now you find yourself more.

more in designer land and marketing land than maybe developer land. Is that pretty accurate to like your current, maybe not your most current role, but like in the last couple of years? Yeah. So originally, I think when my career kind of kicked off, I was doing a bit of both. I did programming and design for a period of time. I think in my first role, it was more front end than it was design. And I sort of like encouraged the people that like, hey, I could actually like do the full designs. And then in my second role, it was, you know, design and coding.

I think it was by the time I made it to my third, like I wasn't doing the coding anymore. So I mean, that was 10 years of code. I've maintained code on the side. I have my own projects. I'm pretty into Next.js these days. And of course, I went deep down the Rails rabbit hole years ago, but Rails has changed so much.

Since I used it. Yeah. Would you say good or bad, this change? You know, I have a ton of admiration for DHH and his ability to just say, no, this is the simpler way. Let's walk in it. It's very different from the way the rest of the world seems to have evolved. I love the React ecosystem. I'm very aware, of course, you can use React with Rails, but the Rails way is different than that, right?

So, yeah, I have a lot of respect for that. I haven't used Rails in the new configuration, so I can't speak to it. What's the last version you play with? Oh, gosh, it probably would be like Rails 5. Oh, gosh. It's been a bit then.

Yeah, it's been a long while. I believe we're on Rails 8, I believe, right? Something like that, yeah. Yeah, Rails 8 is like the latest thing out there, I want to say. We're about to play with some Rails again over here. We've been, yeah, Rails 8.0.1. I had to make sure. Released December 13th this last year. So basically on the tip, 8.0.1. So you're a Next.js lover or just user? Oh yeah, it's a great framework. Next.js for life? Oh, I mean, as much as anything, right? Like I would start another project in it tomorrow, right? But like...

So to me, Next.js is a really great choice if you're using, if you decide you want to go with React on the front end. You know, I have mixed feelings about it. I do think that the Rails 8, like there was something, well, not Rails 8, but back at Rails 3, let's just stay there. There was something sure about like doing views and CSS and like a little bit of JavaScript. Like it was great. That was a, I mean, it was so easy to sort of comprehend the ecosystem compared to like Next today, right?

But at the same time, and there's stuff when I look at Next where I'm like, man, if they only provided a default ORM database system out of the box, it could be so much more. So I look at it like that. But every language or framework seems to pick a line as to how far they will go. And I get it. I mean, there's different choices on the ORM side for Next, although everything I've done has been like Prisma with it. And I think Prisma might be the leading candidate.

And there is a sense in which it seems like the default stack is very much evolving around Vercel and like the options that they give you that are easy, I guess you could say. But Nexus still, you know, doesn't have as many defaults as Rails does. You have to configure it. I wonder what that is, honestly, especially since that Vercel had an opportunity to create a convention over configuration, right? Which is the whole adage that Dave has been pushing for a very long time. It still does to this day, like it's not changed. Just to give a date for Rails 3, okay? The release date. You want to take a guess? Rails 3's release date?

I mean, it's like 20 years ago, isn't it? I mean, it's pretty close to 20. What is it? 2015. No, really? 15 years ago. August 29th, 2010. 2010. What is Rails 3? Rails 4 was released in 2013. Okay. And you didn't touch Rails 4, right? Rails 3 is the last version you touched? So we did some upgrading with Radiant before. The really brutal thing with maintaining Radiant was upgrading between the major Rails versions because so many things broke.

And we would spend a couple of weeks trying to, well, it's probably more like a couple of days, but when you're working part-time on it, it's a couple of weeks each time. And eventually it was just, it was too difficult for, I think, us to maintain at the current, you know. You're not on the project anymore? Radiant's done for you? Radiant is kind of parked. There were two different maintainers after me that took over the helm. Jim Gay is the current maintainer, but there's not a lot of activity on the project and it has not been upgraded to the latest version of Rails. So it's unlike, it's...

It's unlikely to have any users. And if they're users, they're not choosing it, they're still there. Every once in a while, something will come along and it seemed like it will have new life. There was a fork of it at one point. I think I feel kind of gratified on two fronts. It was a lot of effort early in my career to maintain. And I think a lot of people got something out of it and were inspired by it. And I mean, it's incredible as a developer to have gone through that experience. But there comes a point where like...

is open source going to be your life right and for some of us like you know that's what they that's what you choose but for me i i've like i i sort of feel like i have two professional careers design and development right and maintaining both of those is difficult um and then family life as well i've got um an eight-year-old a six-year-old and a three-year-old now so things are busy at our house so busy yeah wow three kids will keep you busy that's for sure it's so good that's a triple zigzag right there yeah it's tough because they're all close to the same age at the moment but um Yeah. At least. Yeah. Yeah. They're, they're, they're each incredible, wonderful people. We love kids around here. Of course I got three myself. Jared's got six, gosh, six children. You got three.

I'm looking at this commit history for Radiant, and I don't know if this is accurate or not, but it says four years ago was the last commit. That makes sense. I think Radiant, which is the core templating language under it, we've had some commits in the last six months. Gotcha. But the main project, it doesn't surprise me that it's been four years. So back to sort of establishing some of the things that you're dabbled in. So an icon designer, some of them open source. Is that Zest? Is that right, Zest icons? That's Zest, yep. Now you're more front-ender, designer, right, than you would, say, back-end. Definitely, yep.

and no longer dealing with Ruby, hanging out in JavaScript land, loving as best you can, Next.js, missing some convention over configuration, it seems, but pretty happy. Yeah. I mean, like I'm not coding in my day job, right? Like in my day job, I'm generally just focused on like Figma and I've done a lot of framer recently for the marketing site side, but I have a side project I've been working on in Next. I often find myself in React land from time to time needing to do a component to demonstrate something to a developer, right? So I do a lot of CodePen React stuff from time to time. It depends. But when you're doing developer handoff, that's kind of when I might do a little bit of code on something. And it would typically be more complicated interactions, the point where it's too difficult to just write it out or draw pictures of what you want it to do.

And it's just easier to, you know, do some code. Well, it's good you have that history of being a developer. Not that you're not today, but in practice, on the day-to-day, you're not developing. You're designing, primarily. But you have that language, that history, that ability to sort of still spike something out, like you mentioned a component to.

to sort of get somebody a developer to speed to an idea you have it's funny i i know it's a sort of controversial thing that comes up does actually designers learn how to code from from time to time and i'm i'm of the opinion like heck yeah like and learn how to do business too and learn like all the things like take it take it one at a time don't just stay in one late that's that's that's my opinion because it like makes you better at communicating better as a designer uh across the board um there's so much stuff right now where even just looking at Figma, it seems like it's going about 90% of the way there. Framer's better on that side on some things, but both of them, I'm just like, could you just learn a little bit more about what components are and provide the things that we need to make handoff easier, essentially? And I've talked with a lot of designers from time to time, and one of the big challenges when you come just from UX land is understanding

how it's going to be built into a system. Because we're all designing systems now, whether you want to or not. And if you design with the grain of the system, you might say, things are easier. If you design against the grain of the way that the system has been built, then things are hard. Yeah, learn a little bit about coding. Get into the CSS a little bit, but also get into the React side a little bit and try and learn some of that. What would you say that you like a lot about what you're doing now? Like if you're on that front-end side of things and you're sort of in design marketing world and you're building with Framer even, like I think Framer's pretty awesome. Haven't personally used it myself, but I think there's a lot happening in the ability to build out a front-end website that looks really good, that doesn't require you to touch a lot of code, if any, really at all. But I think familiarity, like a past front-ender.

who has hand coded front to back would thrive, in my opinion, on some of these newer tooling because you know a lot of what it can do. You know what the web can do. Even though you're not doing it, you know where the limitations and peripherals are in terms of where the boundaries might be. Yeah, it's very much like that. So the thing that I'd say is, as a designer, there's only so far that you can go with just like static pictures. And particularly when it comes to like motion and scroll driven designs and that kind of thing. Almost anything that you produce in Figma is going to be kind of hand wavy as you try and hand off to the developer. And there's this thing that happens, designer fatigue, where you're trying to get a project out the door. You have some ideas for how the motion should work. And you could spend a lot of time in...

in figma like trying to articulate that and hopefully your developer can come along and pick up those notes and like make it awesome right but you're not you're not necessarily always working with people who who can do that and tools like framer and webflow now are they're they're sort of speaking both languages and sitting really well in this niche of you can't do anything you want with the tools but you can do a lot like 90 of the common use cases now and what that means is that as a designer you spend less time like trying to articulate these interactions because you can just design and prototype them in the tool. And that's pretty incredible. It's really neat to see these web-based tools now. The hilarious part about this is I remember Dreamweaver. Dreamweaver is what I cut my teeth on, right? Yeah, man. But it's taken us, what, 20 years to get to this point where we have web-based tools that are kind of like of a similar caliber to what they had in desktop apps way back when.

And when you're doing that, one of the really cool things about Framer is that it's managing deployments in kind of a modern way and it does it all for you, but you can roll back to older versions, right? Which is kind of like a GitHub, you know, kind of feel to it. So, yeah, I mean... I'm coming off of Framer high right now, so you might take it with a grain of salt. That's the best time to espouse your thoughts. Yeah, I kind of think, like, why, if you were a web startup, would you not use Framer right now? Or maybe Webflow. I haven't done anything with Webflow. Are they similar? They are similar. Yeah, they are similar. I've heard that and kind of looked at Webflow. I need a Webflow project is what it is, but Webflow's component system looks like it's a little more developed than Framer at this point. It's funny, I would say that Framer feels like Figma.

but a tool that you can build a website with. And then Webflow feels like Dreamweaver to me. Like it's kind of this, like on the web. That's kind of the difference between the tools. It's really cool. For those who don't know Dreamweaver, what do you mean by that? Is it like code and the view in the same tool kind of thing? Is that what you mean by that? Yeah, Webflow from my memory, and it's been about five years since I played with her, maybe four years. From my memory, it seemed like it had more like you could code to do things in it when you needed to. And Framer doesn't...

really do the same things. Framer doesn't have events. like an events system would be the way I would describe it. It can do like simple, like interactions on click and things like this for components and component transitions, which is enough for like most things that you're going to do. But if you're trying to build something a little bit more interactive, you almost, it feels like you have to break out of the matrix in Framer and write, you can write React that works with Framer, right? But like the worlds don't really coexist very well would be the way that I would describe it. I think Framer will eventually get better and better in this regard. But it feels very much like,

I have to be a pretty smart coder to do coding in Framer, or I just don't have access to that world, essentially. Well, friends, I'm here with Scott Dietzen, CEO of Augment Code. Augment is the first AI coding assistant that is built for professional software engineers and large code bases. That means context aware, not novice, but senior level engineering abilities. Scott Flex for me, who are you working with? Who's getting real value from using Augment Code?

So we've had the opportunity to go into hundreds of customers over the course of the past year and show them how much more AI could do for them. Companies like Lemonade, companies like Codem, companies like Lineage and Webflow. All of these companies have complex code bases. If I take Codem, for example, they help their customers modernize their e-commerce infrastructure. They're showing up and having to digest code they've never seen before in order to go through and make these essential changes to it. We cut their migration time in half because they're able to much more rapidly ramp, find the areas of the code base, the customer code base that they need to perfect and update in order to take advantage of their new features. And that work gets done dramatically more quickly.

and predictably as a result. Okay, that sounds like not novice, right? Sounds like senior level engineering abilities. Sounds like serious coding ability required from this type of AI to be that effective. 100%.

You know, these large code bases, when you've got tens of millions of lines in a code base, you're not going to pass that along as context to a model, right? That would be so horrifically inefficient. Being able to mine the correct subsets of that code base in order to deliver AI insight to help tackle the problems at hand. How much better can we make software? How much wealth can we release and productivity can we improve if we can deliver on the promise of all these feature gaps and tech debt? AIs love to add code into existing software. You know, our dream is an AI that wants to delete code, make the software more reliable rather than bigger. I think we can improve software quality, liberate ourselves from tech debt and security gaps and software being hacked.

and software being fragile and brittle. There's a huge opportunity to make software dramatically better, but it's going to take an AI that understands your software, not one that's a novice. Well, friends, Augment taps into your team's collective knowledge, your code base, your documentation, dependencies, the full context. You don't have to prompt it with context. It just knows. Ask it the unknown unknowns and be surprised. It is the most context aware developer AI that you can even tap into today. So you won't just write code faster. You'll build smarter. It is truly an ask me anything for your code. It's your deep thinking buddy. It is your stay in flow antidote. And the first step is to go to augmentcode.com. That's A-U-G-M-E-N-T-C-O-D-E dot com. Create your account today. Start your free 30 day trial. No credit card required.

Once again, augmentcode.com. You said you're coming off a high. What did you build? So we built the Opine website, tryopine.com in Framer. How many people were involved in this project? I did all the design on it. We worked with a company, Fletch. They're also based in Texas on the homepage copy. But everything else I wrote or designed on the site. So everything you see at tryopine.com is you? That's right. Aside from copy. And I mean, you can see the sort of scroll animations and things like that. And we did. It was really interesting.

This is what I'm saying. It's hard to prototype interactions. We did actually more complicated animations in Framer. And as we were talking about it, the founders and I, we kind of backed away from it because it was feeling like we were hijacking the scroll bar. I don't know if you've ever been on a site where you scroll and everything's reacting to it. But I say that because you can go overboard with it.

And I don't know how I would have articulated to a developer to build the site without having a lot of back and forth and iterations and things like that. And just being able to get that instant feedback of designing it in the tool. The other thing that's interesting there is they have an import from Figma, which is pretty good. And it basically turns all your Figma. stuff into devs and you know with borders and you know a little bit of svg code here and there and what's really interesting about that is you can build responsive screenshots in a way that like if you had if you had screenshot them or exported them to png or even svg like you can't quite achieve um so on both of those sides like i'm pretty stoked about like what we were able to achieve without development and the tool so i mean just browsing the site i mean not that i'm trying to um Say that frame or web flow or tools like this are bad But I can't tell that it's not a hand-coded kind of front-end marketing site. It's not It's not easy to tell that it's

Something more simple, I suppose. Or maybe you can speak to the, in quotes, more simple than I'm suggesting. I'd imagine it's just as hard as you've got to put the design work into it. You probably have less control over everything. And so maybe it's actually slightly harder than hand-rolling your own code to get to the result. Was there much code involved in building this site out? This one, they have something called a behavior. And there were a couple of little things that I had to work around. Framer.

stuff with like a, I don't know, a 10 line behavior. What's the language of that behavior? It's JavaScript. Obviously, right? It's kind of like, I think it's basically, what do they call that? It's in React where you compose components by like, they call them higher order components.

That's how you decorate things, if I'm using the right terms. This is what I'm saying. So I come to React and I enjoy using just React, the library. I've never used state management stuff, things like that. And I know JavaScript very well. And I love using just React on its own. It's such a beautiful little library. It feels like jQuery to me. And I have a lot of joy using it. There's so little code required to pull off interactive things. It's crazy with it.

But then when you try and do it in Framer, you kind of have to learn the Framer way. And it does use some of the things that I'm less familiar with. So yeah. But Framer has so much that's just available from like drag and drop sort of perspective. It's interesting too. Like I run across designers that don't use auto layout in...

Figma, right? And auto layout is just like Figma's version of Fluxbox, right? And Framer is built on auto layout, right? Because you're actually trying to build something responsive, essentially. So there is a little bit of a change to it when you go to the tool, but...

If you're not used to drawing in Figma with auto layout and building things in Figma with auto layout, then you'll have to change that as you learn how to use Framer. Framer does have absolute positioning. You can do that, but of course, you don't want to absolute position everything. Your website will be a nightmare to maintain, right? See, this is where I think having that background in CSS makes sense. Somebody who doesn't come to a tool like Framer.

I actually came up in the day whenever you would absolutely position everything, but you would float everything. The whole layout was float-based. This is prior to the new tools these kids have these days. The best era of CSS is today, in my opinion, but it's still hard. Yeah, it's still hard, and I think it's a little more overwhelming than it was. I guess there was a lot of things.

In the beginning with CSS, you would come across somebody's site and you'd be like, whoa, how did they achieve that? And then you'd look at it and it'd be like, oh, it's all floats everywhere. This is weird. Why would they do that? That's so gross. And then you'd start doing it, right? I mean, I remember reading a book. Who wrote the book? There was a famous book. I want to say pre-2010. I'll find it and I'll put it in the show notes because it's the oldest of old school books. Yeah, yeah. Are you talking about Mike Myers? It was like a black and white book. It read more like a novel than a CSS training tutorial book.

I might even still have it in the library. But Mike Myers, I think it's my Eric Myers. So Eric Myers was the godfather, I think still might be, of CSS. And I think you and I both came up in the days whenever the industry was transitioning from tables, literally from tables. Yeah, so some of the first websites I did were in the new mindset. So using CSS for layout, essentially. That wasn't heard of. It was just formatting, just decoration for a little while, if you use it at all. It was always in the head of the document. There wasn't a separate document that was strictly for the CSS.

Well, no. Tables is the old way. We have screen readers, so we have to, you know, screen readers can't read tables well. And you need to do all these different things. And then you had Web Standards and Jeffrey Zeldman and all these things that popped up. And then you had even things like, what was it called? Microformats. Remember microformats? Yeah, microformats are still around. They're still around, yeah. Now you just sort of like...

you know, expect that to be metadata that you don't even know how it gets there these days, honestly. Well, it should be like your platform knows about it and spits it out, right? As a developer, you're not like hand coding those things anymore. Now you're putting in variables and you're putting into a database and you're part of a front end code to some sort of backend. And, you know, the CMS or the platform takes care of sprinkling in the goods. There's so many good things though now in CSS. The grid layout was kind of the biggest thing recently that has made it so much easier to build these things. But Flexbox, of course. And then there's all kinds of...

niches to get into like um the scrolling stuff in css to get the snapping and everything just right things like that well i think we got there by talking about really framer but from a from the lens of having been steeped in building front ends yourself yeah and i think having that those battle scars if you want to call them that going into a tool like framer one you could probably appreciate it more but then command it more because you literally know how the web should work if you were hand rolling the code yeah i mean it's it certainly helps um I will say it's wonderful to see a tool like Framer taking an approach that's probably more for people like me, I guess you could say. I think there was a lot originally of envisioning the tool in a way that was maybe not that practical, right? No one would argue over using tables, but like...

When we first got started, you would export your fireworks image to a table. And then in Dreamweaver, you would tell it how to size it appropriately. It was crazy. But the way that the tools are built, like Framer is built on all that good layout stuff and everything that CSS has. And they have catered to designer terminology for that kind of stuff. I don't think it says Flexbox anywhere, but it supports the Flexbox stuff. So if you're already in that mindset, it does give you an edge for using a tool like that.

I also used a lot of YouTube stuff to learn Framer as I was doing it. And I do more with YouTube learning for some of this stuff now than I did in the past. Yeah, I'm with you on that. I feel that I waste a lot of time, especially on beginner content. I feel like almost everyone on YouTube assumes the person is new.

and novice, and that just drives me nuts. I was going to ask you, in regards to that, because if you're using YouTube, you're probably leveraging ChatGPT, for example, to say, okay, I'm using Framer, and I want to do this kind of thing that you could do in CSS. What's the equivalent in X? Yeah, so actually, it's interesting to say that, because I did use ChatGPT from time to time. I found it had mixed results, and part of it is that Framer has different libraries that it knows about, so sometimes it couldn't discern whether I was asking about the application or about the programming APIs that they have.

um so that was a little interesting one of the great things about framer i know i'm glowing about it a little bit but like there really is a large community and there's a large community on youtube too so You know, in my most recent role, I was building a content management system, content management world, right? And there were a lot of things, it was called Unstack. There were a lot of things about it that were great. And we were building the component system and things like that. But we didn't have a ton of users in the way that like Framer does. And so using a tool like Framer does cater towards a more technical kind of person to be able to pull off a site in it than what we were doing with Unstack. Unstack was strictly for the business user, marketing people.

But there's such a big community, like learning the two different tools is a very different process. Like you can just go and, you know, they have beginner-focused stuff, but they also have a lot of niche stuff as well. Some of the things I was looking at was like, how do I get my menus to be responsive? So that was like one thing that I was consulting on YouTube. And there were two or three videos that I looked at for that. So you can get kind of niche. And that's a really interesting spot to be in for a tool when you're on top, I guess you could say. Interesting. So what is this unstack? What were you doing there again?

What were you? You were building the CMS itself? You were doing the front end stuff. At Unstack, I was head of product and design. It was my first role doing both product and design.

It's funny. So the company I was at before then, we were going through this phase where they had a WordPress website and they brought me on as a designer. And I was, again, in a sort of like mixed marketing. I was helping marketing a lot role and doing product design. And we were looking at a website and it was something that the founders had stood up like six months ago before they had funding. And it was really just to begin to put content on the web about what they were doing. So the design was not great.

So I'm a designer. Of course, I want to upgrade that. And so I was looking for tools, and I stumbled across Unstack. And Unstack at the time was really focused on the sort of SaaS businesses. That was their heart, to make it easy for SaaS businesses to have a website, something that's not as complicated as WordPress to set up with a lot of pre-built components to do things. And I was blown away because I built the first version of our homepage in about four hours on Unstack.

You have to design for the system, but it was pretty good. And it was pretty good in the sense that the founders could use the tool as well to build pages. And that was important to me. I love tools that you can be collaborative in with people and build content together and build things together. I still think Instac is a great tool for this, but we did pivot more towards e-commerce for a while. We were focused on Shopify.

We were eventually acquired by this company, Elastic Path. And the tool is now called Studio. And it's been integrated with Elastic Path's core offering. So, you know, you can build Elastic Path storefronts in Studio. Are you using AI a lot in your processes? Aside from this obvious one with Opon where you did the, you know, the awesome sort of guide, I suppose. I imagine it was a PDF as well as maybe a web version of it potentially. But do you find yourself using a lot of, or even being open to, tools that help you as a designer, developer get?

further faster better yeah you know it's it's one of those things that on the design side sometimes it's easy to feel sort of conflicted about it because now it's my profession um the way that i think things are evolving and not just my opinion but it seems like a lot of people's opinion is really more that people who use ai like you'll become more It's not that designers will disappear. It's that a lot of the designers' work will be augmented by it. And the designer will be more of an operator of the AIs kind of thing in order to build things out. That's kind of my view of it. And I think you'll see...

you know, lightweight design being done by the, you know, there's things like Vercels V0 right now, Bold.new, Lovable, that are doing the full coding, one shot prompt kind of things. I think you'll see these things get better and better as things go by. So people with expertise in them will be able to do more and better. And they'll be, you won't even know they're using AI sometimes, but they know the right AI tool to reach for to do the job and get it done quickly. ChatGPT I've done a lot with.

More recently, we did a logo for our swim club. And it's very interesting using it as a tool. I could generate lots of different versions with different texts, get different ideas, and then you can mash up, right? And so your skills in Figma are still important because you're going to take the different ideas and put them together. But it's a different workflow in order to build these things. Figma is getting more and more AI as well.

It's not quite to the point where it can use my design system to build screens. I think that's the killer application when you can give it your design system and say, hey, build a workflow for this. And then the hard part of design comes, which is that like editing, like what really needs to be on the screen. And right now, the work that we do as designers, a lot of times we're like building up to that point where we can do editing, right? But imagine if you could spend that time that you normally spend just building up to it, like focusing on the, you know, the really important.

editing, getting it just right, dialed in sort of activities. I think that's the near future. I think you're right. I think in a lot of applications, which are, you're on the edge of like, and what I think will happen is, it's already happening, is you have a lot of entrepreneurs who have a vision or a big idea or a direction and they don't necessarily need to camp out and let me learn how to build the best XYZ in Figma specifically. It's more like, what does my user need communicated to?

You know, and so that might be A-B testing, landing page, that might be A-B testing, or like even just iterating on the design of and the editing, as you said, of a homepage. What really captivates the essence of our brand and what we offer and what we aim to offer? Sometimes you call that working on the business versus in the business, and that kind of reminds me of that. It's like, do I want to be working on the business, which is vision, systems, people, direction, product, customer, analysts, you know, analysis, capital raising, or do I want to be... you know, do I want to work in the business, which is moving this from here to there, literally hammering if it's a physical thing or literally coding it if it's a digital thing. And obviously we're talking about digital here, but I feel like this is like this era where that line begins to blend because you've got a lot more possibility for the non, I don't want to say non-talented necessarily, but someone who doesn't really need to gain the talents in a particular area just to achieve their outcome. You know what I mean? And that might seem like a shortcut, but it's sort of this weird Wolverine where that's more real than it has ever been before or more possible.

Yeah. I mean, let me give you an analogy. When I was a kid growing up, I was homeschooled in high school, interestingly enough. And I can remember having this argument with my mom that I didn't need to keep doing spelling. She had me doing spelling in high school because I was bad at it. Because Microsoft Word can spell check my work. Why do I need to learn it kind of thing? And I didn't get me out of my vocabulary. I still had to learn the spelling of all these things.

But later on in life, I really look back on that and came to value it because there were so many, it's getting better, but there's so many applications where if you don't know how to spell, it just shows up in your work, right? You know, whether you're hand coding a webpage or some component or doing something in Figma, Figma's finally got some spell checking there. So for much of my professional career, it's been valuable to like know how to spell, knowing words are spelled wrong, right? I think it's going to be similar for us in the future. We're probably going to have arguments with our kids about like, why do I need to learn how to do this? The AI can do it, right? But the people that are going to be like the true craftsmen in the trades of what we're doing are the people that are going to understand like how to do it at the very lowest level kind of thing.

sort of always been true in this type of work, right? Yeah. That's what gives you an edge because you're able to quickly jump down to the point where, I mean, playing around with GPTs today, right, in code, they can get pretty far, but they tend to get stuck. And if you don't know how to code, you don't know how to give them the instructions to tell them what's wrong, right? So it can be...

There will be a lot of lightweight applications of this where I'm a marketing person. I need to have a tool to collect email addresses for some event or something that I'm doing. I want it to be done a certain way and feed it into this spreadsheet on Google. ChatGPT can do that stuff all day kind of thing. But you're not going to see those people transitioning into the full-on development stuff until they learn a little bit more as well.

I think these things will get better. They'll be, you know, ChatGPT will be able to do more complicated applications, but there's always that little bit that's not quite good at yet that's left over, I guess you could say. And then for the practitioners, like the people that like are doing design or doing development, they're going to be a benefit because again, they can spend more time focusing on what really matters, but they're still going to know their stuff to do it well. Right. When you look at your, the stack you use.

and your process. I think you got a new job. Did you get a new job? I did, yeah, just over the weekend. So over the weekend you accepted something. Can you mention that yet or is it still? Well, I'm going to Geard. I'll be head of design at Geard. Geard is another startup in the e-commerce space. So it's kind of my background right now. And what they do is they help brands like AT&T with a lot of refurbished cell phones, for instance, be able to list them on platforms like eBay and Amazon. So we're building a product that does that.

So, yeah. How do you spell Geared? Is it just Geared? It's G-I-E-R-D.com. G-I-E-R-D.com. Well, first I would start with like, gosh, let's spell that a way that... It's a short domain name, right? I mean, it's awesome, short domain name, but you have to ask, you know, don't make me think. Yeah. It's old, but Steve Krug had it best. Don't make me think, you know, when it comes to...

Web UX or brand UX. It's true. But I mean, there's trade-offs. Like, I think once you know it, you'll know how to spell it. And you can get there every time. You know what I mean? G-I-E-R-D dot com. It is just her domain name. So you're starting there soon. What's your role there? Same? So I'll be head of design there. Head of design. So is there an existing design system that you're aware of or designer? Yeah, yeah. So I'll be the first designer. Okay. And yeah, we'll ultimately build out a team. But yeah, I mean, right now it's still relatively small. I think Garrett is like,

I haven't actually asked the exact number, but I think it's around 60 people at the moment. They were really interesting because they have kind of built the relationship with customers and have done the effort manually. And now they're building the product that will replace a lot of the manual effort of the processes right now. So the website looks good. I mean, I'm on geared.com, G-I-E-R-D.com. Not a bad website. I mean, you couldn't tell if they don't have a designer or haven't traditionally hired one. I'm sure they've probably outsourced or consulted. Yeah, they've worked with it.

agency. Right. I guess what I'm trying to drive towards really is when you, not that you know your marching orders already, but if you had to choose your stack, when it comes to rethinking or implementing something new, what will you do? Will you sketch something out on paper first? Will you begin literally directly in Framer? How will you go about it? Will you do some market analysis with a GPT or an LLM or some sort of inference model? How will you approach what your next role will be here and apply your skill set there? Well, it sort of depends on like what the ask would be, right?

So at Geared, I will be supporting the marketing team. So I think there will be some things that we collaborate on, essentially. I think I will have a relationship with the agency as well. Some of these details you've still got to kind of work out. Let's maybe zoom out instead of literally what's going to happen. Let's make an assumption. Well, I think what the question you're asking is, if I'm a startup today, what would you do? So some of these questions are what your objective is.

For me, I would probably go with Framer because it's a tool that I know right now. And I know it can get us really far. It does all the standard startup-y stuff that I'm going to want. I might compare it with a tool like Webflow because I haven't used it yet in earnest. So I might look at that.

There's other options that are more technical. One of the ones that we were exploring for Opine was just doing it with Next.js. You know, even in the next six months, depending on how the LLMs do, like there's been talk of the idea that like perhaps marketing people in the future just ask the LLM to code the page, right? To bypass the coding side of things. Maybe, maybe. But in this moment, like I think Framer is the perfect choice for a SaaS-based startup. If you have design talent in one of your founders or your early people. So you're not going to paper. How are you thinking about this? You're going Framer. You just start building a website?

You got artifacts in Figma, you got artifacts in, are you using Illustrator still yet? Like what other tool are you using? I'm still enough of a designer where I would have to start in Figma on this stuff. But the interesting thing about the OpenSight is that... The way it came together was I left Figma a lot earlier than I normally would have. So the conceptual side of it started in Figma with the development of the brand, I guess you could say, and the implementation of the brand on the website started in Figma. But after I had kind of done the homepage, basically gotten it figured out, then...

It was pulled into Framer, right? And I got it all set up, I got header, footer, and then just iterated. And it's iterated to the point where what's in Framer, it resembles what was in Figma, but there's a lot of little changes. And most of the other pages were just built out in Framer without any work in Figma. I tend to have to go back to Figma if I need to do illustration.

of some kind. But when you kind of build your brand assets and get them set up inside of Framer, you know, if you're just reusing, you can build up full pages without ever touching, you know, Figma. So Figma and Framer. Figma is, I mean, your blank page beginner spot, right? I mean, that's, back in a different day, you may have used Illustrator. You may have used potentially even Photoshop. Yeah. I mean, I think Illustrator has its edge on some things still. Photoshop as well, of course, if you need it for photo editing. For a lot of things, I think people tend to be using other tools and it's only for

like very specific things. So in my last role, speaking of AI, I did use Photoshop for the AI features, which is interesting. So combining it with Midjourney, Midjourney could do the initial conceptual work, and then you could get Photoshop to replace this part of the picture that Midjourney could quite get right. So combining the tools together, that was a really nice workflow. But personally, I tried to stay out of the Adobe.

landscape. It's funny because I don't know if I have any legitimate reason to not like Adobe these days, but there was a point where it felt like they took over your computer. In order to use it, you had to have their little thing sitting in your taskbar that ran all the apps and updated all the apps. I remember any time I installed Adobe products, all of a sudden my computer ran slowly. With Figma running in the browser now, it seems like everything runs in the browser. It just feels so much more native than some of their tools. But Adobe has also been a kind of like acquirer of good software. And killer, particularly. And sometimes killer. So, yeah, I mean, they killed fireworks, which was my first love in design tools, right? So many people love it, yeah. And then there was a short period of using Illustrator and then Sketch came along and then it was Figma. So even with like Zest icons, that was all in Sketch? No, I did Zest in Figma from the beginning. Really?

I would say, from what I understand, you should probably compare Illustrator to Figma if you're doing icon work, but I'm such a lover of Figma at this point, and Figma has something called vector networks is what they call them, which is a fancy way of saying that you can join any point to any point in Figma. which was a novel concept at the time that they invented it. I'm not sure if it's supported by Illustrator now. It might be. But essentially, you can build shapes that are all connected and then tell it where you want the fill to be in Figma, inside that shape. And that is a nice sort of upgrade for doing icons.

I'm not convinced that Figma is the ultimate tool for icons. I feel like something native to SVG probably is. And I say that because I often have to export from Figma to SVG. at least historically, I'm not sure if they can prove this, but it would convert the paths to filled shapes, which is interesting because Figment has slightly different rules for the way that it curves paths and then SVG. So there's some things where it feels like, Figment, why'd you go the other direction? Why'd you choose to be a non sort of web tool technology? From what I understand, the originally implemented or an early version of Figment was SVG, but they went away from it because the...

WebGL was so much faster than SVG. SVG has a hard time rendering in browsers, or it did. I don't know if it still does. Massive SVGs are very hard to render. But for icons, your output you want is SVG. So if you could have a tool that was, there probably is one out there, that spoke pure SVG and was kind of tailored to the icon workflow, that would be awesome.

Well, friends, I'm here with Samar Abbas, co-founder and CEO of Temporal. Temporal is the platform developers use to build invincible applications. So Samar, I want you to tell me the story of Snapchat. I know they're one of your big customers, well-known, obviously operating at scale, but how do they find you? Did they start with open source, then move to cloud? What's their story?

Yes, Snapchat has a very interesting story. So first of all, the thing which attracted them to the platform was the awesome developer experience it brings in for building reliable applications. One of the use cases for Snap was Snap Discover Team, where every time you post a Snap story, there is a lot of background processing that needs to happen before that story starts showing up in other people's timelines. And all of that architecture was built, composing, using queues, databases, timers, and all sorts of other glue that people kind of deal with while building these large-scale asynchronous applications. And with Temporal, the developer model, the programming model is what attracted them to the technology. So they start using our open source first.

but then eventually start running into issues because you can imagine how many snap stories are being posted every second, especially, let's say, on a New Year's Eve. So this is where Temporal Cloud was a differentiated place for them to power those core mission-critical workloads, which has very, very high scalability needs. Although they started with open source, but then very quickly moved to Temporal Cloud and then start leveraging our cloud platform. And they've been running on top of Temporal Cloud for the last two, three years and then a pretty happy customer. Okay, so maybe your application doesn't require the scale and resilience that Snapchat requires, but there are certain things that your application may need, and that's where Temporal can come in. So if you're ready to leave the 90s and develop like it's 2025, and you're ready to learn why companies like Netflix, DoorDash and Stripe trust Temporal as their secure, scalable way to build invincible applications, go to Temporal.io. Once again, Temporal.io. You can try the cloud for free or get started with open source. It all starts at Temporal.io. I can speak to the...

your sentiment about Figma. I had to go recently to help somebody out with something. I was messing with Figma. And I happen to have two computers, you know, I'm like that kind of person where I've got like a laptop and I've got an older machine that I still use every once in a while. I actually, I retire it and I bring it back and I retire and I bring it back sometimes. But it's an older iMac Pro that still has so much firepower in it that I just can't, in good conscience, get rid of it. And so I'll use it. And so I've had my studio, which I'm in now.

And then my home office, which is about to merge. You know, I'm about to start producing my version, my side of the show from a different location. But it's been a while. And so I've been a two location person for almost three years now. And so I've I've had to embrace this mobility requirement to my to my flows. And so I was I don't do a lot of front end stuff in terms of design. And I had to be in Figma for some things recently. I was like, wow, this is kind of cool because I can be on my laptop. Figma. Cool.

Then I'm on my desktop, you know, in air quotes, desktop, this iMac Pro, different machine, Figma. Everything's still there. Extensions are still there. All the things are still there. I didn't have to like install the application. I didn't have to make sure plugins were enabled. I didn't have to do anything besides just log into Figma. And like, it was the first time I personally experienced it. And I was like.

That's pretty crazy. That's pretty crazy. Like the thing I need to use is a web-based tool, which means it's not operating, you know, operating specific, operating system specific. I can probably even hop on a Windows machine and open up Figma and Figma is still Figma. Is that, is that pretty, is that true? That's true. Yeah. I mean, I feel like even asking, is that true? I mean, like, obviously it's true, right? Yeah. So that's true. And this was true, let's see, six years ago. So this is a six-year-old thing. This is not new news to anybody. Yeah. So this, this was.

The funny part about this is I was at a company at the time and I was leading the design team. We had about eight designers working for us. And the team had just grown from three to eight. And the new guys were all Sketch people. And they were like, we got to switch back to Sketch. So we had reviewed tools and we decided to go with Figma when there were just three of us. But the new people all wanted to do Sketch. And I was pushing against it, pushing against it. Well, I eventually left that company and the team adopted Sketch, right?

And at the time, the only thing Sketch had going for it was the plugin system. And there were a lot of designers that were dependent on the plugin system for certain workflows, things like that. And Figma was new and different, but it could do almost everything Sketch could do. except for the plugin system. So that was sort of the big beef. I think they eventually switched back to Figma. I can't see how they wouldn't at this point. But it was funny to me that I got in on it a little too early with that team. And then Figma introduced its plugin system, which is great. And yeah, it seems like that was the point where a lot of design teams started choosing Figma. And it really is because of that collaboration. The crazy part is doing any kind of design review now with other people, you want to do it in the tool if you can.

Like, give them that Figma link. Don't share your video. I mean, you can if you want, but do it in the tool. Click on that person's face and follow them around the tool as they are presenting. And you get it so much clearer. You know, you can have the window as big as you want on your screen. You can zoom in on a particular detail. It's the way. It's so much fun. It seems like your stack is pretty simple then. Figma is where you begin. Yeah. Framer, because, yeah. Maybe some mid-journey.

If you need to do some image generation, some finishing touches in Photoshop, progressively, reluctantly, instead of Adobe tools. So using MidJourney and Photoshop, what I was doing was generating a product catalog for a fake store, right? And it was an amazing experience. You could do different things with it. We also used, so we had two different ones. One was a furniture store and the other was apparel. And we used a tool called humans.ai for humans, generating humans. And that one's pretty incredible. But you would end up sometimes with something not quite being right. Like in either case, like furniture or people, you could just quickly pull that into Photoshop, select the area, have it regenerate it. Photoshop didn't upscale pretty easily with AI and it can generate more of the background. So one of the things sometimes it's hard about,

So mid-journey would be that it's not zoomed in quite right for you. And you can do some of this with mid-journey. They've added more and more tools on that. But you have to send it off to the server to, like, regenerate, and it's kind of slow. So when you get it pretty close to what you want, just jumping into Photoshop can be a time saver.

I'm on the humans AI website. So this will do what? What does humans do? Human AI. Generated.photos. That's the one I use. Sorry, I gave you the wrong thing. Generated.photos. .photos. Well, we just gave humans.ai a little permission unnecessarily. Well, maybe they should be promoted. I don't know. Generated.photos. Human generator. This is crazy. So unique real-time face generator. Features here. You can create a unique person with your parameters in one click. Skin tone, hair color, emotion, age, head pose, gender.

The last time I used this, my one critique of the tool is that it tends to generate really beautiful people only. Really beautiful people. So try and generate an average programmer in it. And they're either like too heavy or they are like really skinny. Right. Like there's a profession. No, I'm just saying like try to generate someone who looks like a programmer. Like you pick their clothes. Well, that's biased. Come on, John. What are you trying to say?

No, no, I'm like an average human. This is a joke, come on. One of my criteria for this type of stuff is I don't want it to look too agitated, even though I'm using an agitated tool. So if they're too pretty, then you're just like, oh, these are all fake people. But it's pretty good at what it does. And one of the things that's cool about that particular use case was you can use basically the same model in different poses, different clothing, whatever. And that's useful for generating fake storefronts because

you need multiple images of the same person wearing the clothes for different angles. So that tool has a lot of promise, and it's already very, very good. Okay, so generated.photos, framer, obviously Figma, mid-journey here and there if you really need to, Photoshop for the final touch if you really need to go and replace something in the image. What else is in your stack? Well, ChatGPT for copy, and then it tends to do, because they've got Dolly built into it now, it tends to do better.

line art sometimes you have to compare with mid journey but if you're doing a logo i like chat gpt okay let's let's go there then so what do you do you just log into your own chat gpt you open up a new chat are you uh are you wordsmithing for a bit there you're kind of getting some emotion some words you go right into i'm probably the worst at this it's always like i know that i know the idea of prompt engineering but i'll just be like generate a logo for my swim club uh competition named you know carry city swimming because that's that's what we were doing and then It'll do it. A lot of times I'll forget to say, oh, I want line art or, oh, I want a particular style. I want hand lettered. You can say things like that. What's the swim club named? It's Cary Swim Club. How do you spell that? Cary is our town. C-A-R-Y. C-A-R-Y, okay. Cary, North Carolina. It is generally referred to as the containment area for relocated Yankees. There's a lot of us that have come down to North Carolina. So Cary Swim Club is what it's called? Yes, Cary Swim Club.

So we have a meet every two years called Carry City, and we did a logo just this week. We were playing around with it. I found a Figma plugin, speaking of plugins, that I could take that generated text and vectorize it, and it vectorized it for me. And it's funny, you'll find that when it generates logos, so the one that we really liked, it said Siri City. uh swim club instead of carry for some reason it changed the c into an s so you have to do touch-ups on it right now but i think it's a great way to get inspiration uh and you might just get lucky with a one shot did the right thing for you it can do icons as well i think it does pretty pretty good at icons But I mean, if you really need icons, you should check Zest out first, right? But after that chat, GPD. Last time I checked, Zest was a little dated, though, wasn't it? Is it dated? I mean, it's been a few years. I mean, you know, it's not as on the up as some of these other ones, you know? Oh, I don't know. At the time, I still do a lot of custom icons as part of my job. If you want two pixel stroke, Zest is.

Zest, I think, was leading the pack at the time. I found another one. There's an untitled UI that has the two pixel strokes. Why is it important to have two pixels? So if you're an icon nerd like I am, the great thing about two pixel strokes is that it's bolder than one pixel, which people tend to want a little bit bolder. And you can align it perfectly to pixels, right? Because one is uneven, right? It'll be sharper as long as it's not curved. Now, curved icons are going to look great at whatever resolution. Are you familiar with iconic.app, by any chance?

Iconic, I think I have played with this one. Is this what you use these days? I really don't use much, but I pay attention to Ormond Clark. He started dunked.com. I think he's done some design at Stripe. I've just generally followed some of his work over the years. I would say like literally probably 15 years or more.

And I've always enjoyed it. I've always enjoyed his style and his attention to detail. And I know he has Iconic out there, which is like free icons for designers and developers. And so there's like a mix of free ones there. But I'm not an icon nerd like you are, where I'm like, well, there's two pixels here. I guess I never thought about it like that. It just depends. I mean, I can't quite tell. I'm looking at it right now. I think, is he using 20 pixels or at least 20 by 20? No, this is 24 by 24. You've got some free ones in there. So if you go to the free ones, you can at least, let's say, use the AirPlay to TV option or the archive option because that's something you can use everywhere.

I think it's at different sizes. You can do it different sizes, obviously. Yeah, I think if I'm looking at this, so if I'm looking at the AirPlay TV one in particular, you'll notice that it's sort of blurry around the inside of it. So this is the challenge when you choose like a 1.5, which is probably what he's done, pixel stroke. It means that you have to sort it, when you do straight lines, you have to decide like...

Is it going to be right in between two pixels? Or is it going to be aligned to one pixel so it'll be darker on one edge and the inside, in this case, is fuzzier, essentially. Because it's using gray, right? Because it doesn't anti-aliasing. So yeah, as a nerd, I love it when my lines can line up with the pixels so it's as crisp as it can be. So that was what I did. The one that, if you like a little bit lighter touch, I've been using on projects lately is Heroicons. Have you played with that one? No. Heroicons is the tailwind, guys.

You know, some of this is a style thing, but they seem to mostly make the same stylistic choices. Like they're in the same family. See, it says here's the outline. It says 24 by 24, 1.5 pixel stroke. Yeah, so they use 1.5 pixel strokes. So what's interesting is you can scale up, I think, to 32 pixels. Yeah, 32 pixels. And it'll align.

All right, see what I did there? Math is fun. But yeah, I mean, there's differences. I have thought for a long while that I might want to do a Zest medium and a Zest light. but it's finding the time to kind of pull it off. So have you abandoned Zest completely in your updating or is it active? How active is it if it is active? I think it's been about two years since I did an update on it. So that's pretty inactive, John. Yeah, yeah. But in this day and age, you got to be, you know, here's what I like about Icon, iconic at least, is they're frequently adding things to the library. There's a business model here.

They're frequently adding things to it, and you can, I believe, get the Pro. I'm not trying to advertise for them, but I think their model is kind of cool. Let me go back there so I can speak eloquently about it. Okay, what is their Pro? Okay, so GoPro. Oh, wait, hang on. It says up here. 24 by 24 bounding box, 1.5 pixel stroke. So you're right. That one is a little blurred. So you're a fan of 2 pixel because it's even. Okay, so the iconic Pro package. Hey, Orman, if you like this, pay us. But listeners, just check it out. Don't buy this. Don't go and buy this. I'm kidding. You do what you want.

Solo is 99 bucks a year and team is 249 a year. Yeah. And you get access to all the icons. There's new icons every week. So I think if you revisited this as your, as your passive income plan, John, you know, steal their model, borrow their model. Yeah. It's, it's interesting. I think, I think it takes me about four hours to do an icon update. Right. And so when I do it, the, the, so that's just to do a release and it could be faster. Don't be wrong with like automating all the stuff. I believe I can speak up. Maybe it could, maybe it could.

But the last time I did it, it took four hours. And some of that's that there's not an easy way to automate sketch output, automate Figma output. So there's a number of manual steps that are in the way of it. So where I am as a dad right now is that... I've got a side project that I'm working on that takes my time. And so once a week, I meet together with my friend who we're working on together. And that's where I'm spending my time. Zest is interesting. I think I make a sale at this point every couple of months. So it's not even great. But if I was to pitch it to you, it's a one-time price. It's now only $30. I've sold it at different price points. It's over 600 icons. And they are pixel aligned for 24 pixels.

So, yeah, it's a baby. I want to get back to it at some point, Adam. Well, I'm here to encourage. I'm not here to make you feel bad. You know what? When it comes to children, man, I'm the first person to say family comes first. I think you have to have some things to keep you creatively alive. Yeah, that's right. If your job doesn't do it for you, then you're going to have some version of a hobby, whether it's a project, money making or not, to sort of like keep your own personal creative juices going and flowing. Yeah. Because it's how it is as a creative, you know? Well, I mean, it's interesting too. It's cool putting stuff out because.

it's going to eventually become dated and you can kind of track your own progress by like seeing. And so I would make different choices today, like on that, particularly on the landing page for zest. Like I'd love to do some different things there, but I'm also pretty proud that like, it's got some staying power. It's, it's, it's lasted. It's still in style. It hasn't gone out of style. Maybe, maybe let's say you, let's say you took your 30 bucks and you turn it to 29 bucks a year.

Cause you sold it once for the license and let's say you started updating it somewhat frequently, whatever you would define that as somewhat frequently enough to make it valuable to somebody to pay you yearly, roughly 30 bucks, 29 bucks a month. And let's say you had 500 sales in a year. That's $15,000 rounded up. Of course it's 14. Yeah. It's funny though. Like, I don't know if I'm too much too close to it, like as a customer, but like, that's not how I want to pay for icons.

I want to pay once and be done with it. The subscription gives you a chance to keep them involved. Oh, I understand as a seller, but I'm saying as a person who purchases. And I also sort of want people that are buying it. So I did structure it so that it's not a multi-use license. It's purchased for your work on a project. If you are an agency, you would need to buy a new license for each site that you did for a customer kind of thing. But partly the reason is I wanted to have that purchased once and you can use as much as you want and you get the updates.

sort of model as a user. Like that's the way I want to purchase icons. I get, I get it. I get that there are people that are playing around with those models and, and, you know, particularly on the, I would say like there's other sites that have thousands and thousands of icons on them that use that subscription model. Yeah. It's almost too much. Yeah. It does. It does feel overwhelming too, too much, but for like picking it for a project to me, like I, I wouldn't, I wouldn't want to use a site like that. Cause I don't want to just like, I have to negotiate with my boss.

to purchase a new icon set. Like, that's the last thing I want to do. But if I can say like, Hey, it's 30 bucks. It's like, great. Do it. Like that's 600 icons. You don't have to design. Well, if it's any consolation, I am not an iconic pro purchaser. Oh, so I'm not saying this as a, as a buyer. I'm saying this as a, as somebody who pays it. I mean, I don't even pay attention to that many icon sets out there. I mean, I could, I didn't even know hero icons from the tailwind folks, but I think the table stakes though, in this game is you have to offer a Figma file. You have to offer.

react and view type components or some sort of integrations because that's the way that is the front end world we live in you still get SVG and PNG of course and I think that I don't know how is Hero Icons is it free Yeah. I think it's heroicons. It comes from like hero. Heroicons. Heroicons. Yes. Free. This is free. By the way, it's Zest icons, not Zesticons. Did I call it Zesticons? No, I'm just, if you read the domain name. Did I call it Zesticons, John? No, you did not. All right. Zest icons. Heroicons. Heroicons. I think this is a problem with icon sets. Heroicons.

That's weird, but whatever. I'm going to call them Hero Icons. So he did a set that was much larger originally. They were called Hero Icons, right? And then they developed a smaller set, which I think ended up being the more popular. It's the default sort of Tailwind-y. So in the Tailwind world, you're aware of it. Yeah. You know, it's really hard to compete with icons out there. There's so many of them out there. Free, not free. Paid, not paid. Subscription, not subscription. You know, it's like unless you're...

Unless you're in to win it, it's kind of hard to compete. That's the thing. I think it's great that there are people doing this. I had an itch. I was in between jobs at the time, and I did most of the icons during that short window while I was searching for a job.

You know, and one of the things that to me, like I still benefit from Zest, even today I was pulling an icon over and making it look like the Heroicon 1.5 because I'm using Heroicons on another project because Zest has things that Heroicons doesn't have. So it's great for me personally to just have a library of things that I can pull from and build from. It's better than starting from zero, right? Yeah. That's cool though. So as one note of feedback, maybe I'm just so used to this 1.5 world these designers force us into, but when I go from iconic to heroic,

to zest zest feels bold and thick yeah a little branding that's the branding that's that's by design huh bold and uh there's even a sass logo in here john oh yeah man i i'm deep on certain areas i'm pretty proud of so zest comes in three flavors you can have social which is like what you need for a website and it's got stuff like sass in there and other things that you might want but it's got you know instagram and all the different types of logos that you want and those are pretty juicy i haven't seen and i'm a nerd but i haven't seen a set that has as many and you know feels current and all that so there's that and then there's free free which is 200 free icons and i actually tried to put my best icons in zest free okay but when you're ready to pay for it you'll get all the extensions of the best icons and that's what zest pro is so there you go there you go well you know as as a as a word of encouragement i think if uh if you want to be seen as not

dated. And I don't want to say that's necessarily true, but it's been two years. You got to update the brain marks at least. Twitter's now X. That's true. Right? I'm not sure Google's that rounded. GitHub's Octocat is not that anymore. It's more a different silhouette version of it. Less of a caricature. Oh, no, no, no. These are my take on the best version. Oh, I see. But you are right. You're right. I do need to get X in there. I've actually have the icon designed. I just haven't done that for hours. Just got to put it in there. Shit, that update, John. Yeah.

So they queue up and I've got about 20 right now that I need to do another update. Well, need to and have time to do different things. Always choose children and your wife over Zest, you know, or your friend in this side project or whatever you're doing. Why make it a choice if it doesn't have to be a choice? Any final things to share? Any final, you know, wisdom pieces? You know, I think one thing you may have that not a lot of people have is this ability to be both a designer and a developer, and you speak development as well as design. We have a lot of developers in this audience, and so I'm hoping some of them come to this conversation with the curiosity of what a designer thinks of when they choose their stack, when they choose how to leverage these AI tools to get their job done in different things. And so I thought we'd have a grab bag of a conversation here, but any words of wisdom for a...

designer slash developer working with developers? What developers should know about a designer? Anything on that front? I mean, probably my biggest thing for developers would be like, hey, do a Figma course. It's easier than you think. And focus, like learn how to draw with it, but also focus on how to get the development stuff that you want out of it. Because there are plugins that are written that will like write half the code for you when you export it.

from Figma. So it's well worth spending a little bit of time on the Figma side, like doing that if you haven't. For designers, I mean, again, it's probably a challenge to spend some time on the other side of the fence. I think the AI coding tools are pretty impressive, but it is nice to just sit down and learn a little bit about CSS, learn a little bit about JavaScript, the basics of these things. But I don't know. On the design side, I think sharing what you know and what you've learned, maybe teach some developers how to use Figma.

kind of thing. Any particular courses you know of that you've seen or recommended for Figma? Or they have some of their own educational content? Figma is interesting. I haven't...

really explored that because I haven't been learning Figma in a while. But when people ask me about it, I generally say go to the Figma website. They do have a lot of resources and they've invested in that for like teaching people how to use Figma. So you can kind of go down the outline on Figma.com and learn it. You know, my experience recently with Framer would say, hey, go to YouTube as well. And you'd probably find some good people there. Maybe you can do a Figma crash course, you know, revisit Figma as a developer and say, well, but I want to know. Yeah. And all my 10 minutes or less.

All your free time, John. I love it, actually. I used to sit down with every developer that joined the company. My goal was to just walk through Figma, give them an introduction. But even in just 30 minutes, it's not quite possible to pull out how to get the most as a developer. So that's why doing some exploration, it will pay dividends. If your designers are asking you to look at Figma, then spend some time learning it, and it'll be easier for you. The other thing that's interesting is that Figma's recently gotten things like variables.

and things like that. So as a developer, you should be getting all your design system stuff into Figma. And a lot of your designers don't know how to do that. So you may end up having to educate them on how to do that. But the world is better when the systems that we're working in are very similar. And I see a lot of teams that don't seem to quite understand how to use Figma yet from either design or the development side. So yeah.

Well, this episode, as you may know, listeners, was not brought to you by Framer. Although we did talk about Framer.com quite a bit. I would happily have them as a sponsor. So if you're from Framer, say hello. You know, I'm down for that. I like Framer, though. We've been sponsored by Wix. We've been sponsored by, I think, what was the other one you mentioned before? Webflow. Webflow, I think Webflow sponsored us before. But I like Framer. If you like Framer, I like Framer, John. I trust your judgment. That's why I joined the show. Well, John, it's been a pleasure catching up. It's been a while since you and I have spoken.

face-to-face, I think probably years, right? You think it's been years? Yeah, it's been a while. I think our last conversation was on Radiant. The last time I talked to you, you may have been pre-kids. Yes, that was pre-kids. So, I mean...

You got three now, so it's been almost a decade, John. Yeah, I was unmarried as well the last time you talked to me. I think you were engaged. I think you're right. I think you were engaged. Okay. Now you're hopefully happily married. It's a podcast. Transcribe. Don't say it if you're not. Yeah, it is. And if you are, boast, of course. Yeah. Yes. So it's been fun. Yeah. For the listeners, I guess in this little extended outro, John and I have known each other for many, many years. We started the SaaS way together. What else do we do together, John? How did I meet you? Where did I meet you at?

I think we were aware of each other. So there was a time where we were going to similar conferences. Yeah, probably less conf. Did you do RailsConf? Did you do RailsConf back in the day? I think I've done one. I did not meet you there. I've only done one, and it was in Portland, I believe. Okay. 2008.

Yeah, I might have done that one. Yes, I think I did. Maybe you did. Maybe I met you there. I know Steve and Al were there. Yeah. I remember being with Steve in the trolley tram thing that goes around the town. I don't know. Yeah. It's cool stuff. Cool day. Less guys. Less guys are cool. Did you see Alan is running for mayor? Mayor. Vote for him if you're in Panama City. I mean, I would like to become a resident just to vote for Alan. I would vote for Alan for anything pretty much. Miss Alan. Love Alan. Obviously, Miss Steve as well. But don't make me cry on the show, man. John, it's been fun, man. Catching up. It's been fun. Stay cool. Always a pleasure.

OK, it's been a couple of years since I talked to John. I want to say, gosh, John and I spent a few minutes after the show actually catching up on some family stuff and some other things. And it's been, I think, a decade is what we said. A decade since John and I last talked. He wasn't married. He was engaged. He had no children. Now he has children. You know, it's just a different time then. So this show was a fun catch up with John. And I've always loved John's front end and design eye. And I've been really curious about his tools and stack and choices over the years because I've always appreciated his taste when it comes to front end work and design work.

If there's one thing worth taking away from me from this conversation, it is two pixel lines when it comes to icons, because John clearly believes in that. But I would say Framer. Framer got a big shout out on this show, and I think it's probably worth checking out. You know, I'd love to get somebody on the show from the Framer team. And if you know somebody, let them know. We'll have them on the show. It'll be fun. Okay. Big thank you to our friends over at Retool, Retool.com, AugmentCode.com, and our friends over at Temporal, Temporal.io. And of course, our friends and our partners over at Fly, Fly.io.

And those beats are banging the best beats in the business. Break master cylinder. So good. Okay, friends, we'll see you on Monday.

Frequently asked questions

How long is this episode of The Changelog: Software Development, Open Source?

This episode is 1 hour and 47 minutes long.

When was this The Changelog: Software Development, Open Source episode published?

This episode was published on March 14, 2025.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this The Changelog: Software Development, Open Source episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.