Supper Club × Arc Browser with Hursh Agrawal

I sure hope you're hungry.

Oh, I'm starving.

Wash those hands, pull up a chair, and secure that feed bag, because it's time to listen to Scott Tolinski and Wes Bos attempt to use human language to converse with and pick the brains of other developers. I thought there was gonna be food, so buckle up and grab that old handle because this ride is going to get wild.

This is the Syntax supper club.

Welcome to the Syntax supper club.

Today, We're talking to Hirsch Agarwal.

Was that was that good, Hirsch? That was perfect.

Oh, okay. Awesome.

And, We're gonna be talking about the browser company and the Arc Browser, which has been really my favorite browser here since I I picked it up. Since I got Just do it. I haven't put it down.

We're gonna be talking all about Arc and the browser company itself. And, we're gonna be talking to Hersh here.

And with me as always is Wes Bos. Wes, how's it going?

I'm doing good. Doing good. Excited to talk about Arc. It's It's kinda interesting.

We talked about browsers and browser engines a couple podcasts ago, And it's going to be interesting that I'll save it. I'll save it, but I'm very excited about it. You'll save it. Yeah. I would start I'm starting to go into the whole whole topic.

The whole thing. Yeah. Yeah.

We're sponsored today by 3 awesome companies. First one is Auth0. They They have a developer day, coming to London, September 29th in Berlin, October 5th. It's totally free, but you gotta register if you wanna be able to get in. We'll talk about Partway through the episode and Gatsby, the fastest front end for the headless web and fire hydrant. It's a one stop shop to develop and scale a world class Incident management program. We'll talk about all of them partway through the episode. Well, let's, let's introduce her her Hershier. Hersh. Who are you? What do you do, and what's what's your role with with the browser company?

Yeah. Sure. I'm one of the cofounders of the browser company. I'm also the CTO.

This is actually my cofounder and my 2nd company together, we started our 1st we dropped out of school to start our 1st company in 2011, many years ago, over a decade ago.

We've ridden that for about 3 years. What company was that? It was called Branch. It was kind of a, like, moving panel discussions online.

And so it was all text.

And what ended up happening was Time Magazine, New York Times, these big publications, would use it on their blog to record these, these panel discussions and then post it up there.

And then after 3 years of working on it, we ended up getting acquired by Facebook. We were at Facebook for some time. We'll do a number of other things.

And then 3 years ago, this thing started. Wow. Cool.

Wow. Yeah. Well, why don't you,

give us a rundown of what Is Arc, and why did you decide to make another browser? 3 years ago, when we were, you know, we were thinking about, do we wanna start another company? We're thinking about ideas.

And we started looking at the market and seeing this trend, which I'm sure you're quite familiar with, of productivity software, really taking this product lens, and a lot of companies sort of, emerging out of that. You know, Slack was perhaps the first one. Then you see it with superhuman Notion, where all these companies were seeing software that we largely thought was finished. You know, Excel, Word. There's a new version every 3 years, but it's, like, Kinda done. Yeah.

And people would take it and, like, sort of adjust core parts of that. And in doing so, build very popular products.

And we thought it was so strange that the the one product we spend the most time in during the day, you know, like 40 hours a week, 50 hours a week, our entire lives are in it, It's the one single one that hasn't really changed from a from product, from a UX point of view. Certainly, the the web has moved forward. The API is available to the web. The engines are are improved at a drastic pace, but, tabs are still tabs. You know, history is still history. Bookmarks are still bookmarks.

And that was quite strange to us. Especially because it's it's the biggest market and one of the most, you know, one of the biggest products out there. And so or part categories. And so, We maybe a little foolishly, we're like, okay. It seems like a cool idea.

2 weeks later, I raised a bunch of money, and then, you know, a month after that, we're like, oh my god. Now we have to build a browser.

And then Yeah. You know,

the rest is just straight. It does feel like it is funny just how little innovation there has been in those u UI elements. I mean, What did Safari, like, even slightly changed how their URL bar worked and everybody lost their minds? But, you know, Chroma Functionally, hasn't really even changed its appearance at all.

Firefox just kind of a a paint coat and maybe, like, container tabs. That was, like, their big thing was container tabs, and then you have Microsoft Edge. Their their big innovation was the side tabs. Alright. Now we got side tabs thanks to Microsoft Edge.

It's like, man, nobody's really doing a whole lot there, and that was one of the the joys about Arc. When you pick it up, it's not 1 or 2 innovations. It's A ton of innovations, and they all work together beautifully. And the design of the whole thing is so well executed that, like, It's really quite an experience, especially if you've been using browsers for a long time like me. You haven't seen a whole lot that's new. There's a lot new here, and I I think you guys really knocked it out of the park For that, I'm gonna be sounding like a, a big Arc fanboy today because You're so kind.

I've just been going nuts on this browser for so long here. So yeah. Why don't Oh, either Scott or Hersh, are you guys all together?

People listening. Like, what what why is it so good? Why is it different? Like, what did you do different than than Chrome? You know? When we first got into it, the 1st challenge was, okay, how do we how do we actually build a different browser? I think that was one of our first insights is that It both has to be a lot better, and that market is so saturated. It also has to be really different.

And so the first thing was getting into it and figuring out how do we iterate quickly enough and prototype these things. Because, we quickly realized a lot of what we Thought will be really cool. And then what turns out to be cool when you're using the browser every day? That Venn diagram, like, barely overlaps, you know, because we have So much muscle memory, like decades of muscle memory on how to use browsers. And so our initial prototypes, we would build stuff, and it would just be awful because, like, I didn't know I used control a all the time to move to, like, you know, the front or back of an input field, then we would remap it and then instantly feel the pain. So, the 1st year was really just a lot of iteration to really get that prototyping rhythm down and figure out, okay, we need to just Iterate on this stuff in a day, in 2 days to really get through to get to the good stuff that feels really good.

And then the rest has evolved more naturally.

The sidebar, we have tabs on the side. The first thing was figuring out this tab overload problem, and the RAM usage problem of just so many tabs open. And we We sort of realized, oh, this is it's not a technology problem. It's actually a product problem because nobody wants to close their tabs because you're just afraid stuff will be gone forever. I might need

that. Totally. And us too.

And so that was a real insight that, oh, man, all of the so many of the technology problems have just been, dealt with ad nauseam by the Chromium, team, by Firefox, by these folks, but nobody's taking, you know, one level of abstraction higher. Can we have a product and, You know, engineering approach to solving these.

And so that's what led to the sidebar to our pinned and and explore tab System, the auto archiving, man, that was when we first introduced that, that was very controversial internally.

How can you just get rid of these tabs? You know? And then we tried it, and not 1 person missed them.

And so West, you don't you might not know about that feature, but it what it is basically is, like, you can set A a timeline like, you can say archive the tabs that have been open for longer than 7 days.

So if your Tab had been open for 7 days or I mean, you can configure 30 days, 24 hours, 7 days, or whatever if you like to keep your workspace to various levels of clean.

You could have it auto close the tab after 30 days because that means if you haven't looked at it, I mean, you're probably not looking at it. Yeah. And maybe you Either bookmark it or or something, but you're probably not going to miss it if it's gone. Yeah. Honestly, like, you you think about

I don't know who your A target market is maybe you can shed some light on that. But I just think about, like, my wife and and and people who I know. You know, you get a laptop from someone and you're like, Oh my gosh. Is this how you're living? Like, you're like, why do you first of all, why do you have Imovie in your dock? You never make movies.

And you have 10,000 tabs open, and then they have, like, weird, like, like, ways of, like the way my wife bookmark stuff is She just keeps a tab open of, like, her tracking number, and it's just like, oh my gosh. Like, these people need this. Right? It do do you see that? Like, when people are, is that your target marker, or is it more technical people?

Exactly. I mean, in that at to some degree, it's everybody. Initially, it's been technical people who are willing to take the friction Shouldn't have a totally different approach to to browsers.

But that was one interesting thing is everybody does that, and then everybody blames themselves. They're like, Wow. I am so messy as opposed to the browser doing this, enabling behavior.

And so we felt like, wow. The browser really hasn't been taking a lot of the responsibility for Helping you manage your life and your, you know, overload.

Yeah. Yeah. That that's nuts. Let's talk about one of our sponsors, and that is Auth0. They got A little bit of a fun thing coming up. It's off the 0 developer day. It's coming to London, September 29th, in Berlin, October 5th.

You can join them for a one day event to connect with your peers and level up. Oh, Scott.

I don't know if that's okay that they're using that language there.

Everybody's using it. Yeah.

Level up your login experience. There's a 3 hour of, great content from Auth0 engineering product and DevRel Covers the state of identity and digs into new capabilities such as fine grain authorization and passkey. That's kinda interesting. I don't know what either of those things are. So maybe I should fly to London, to take this workshop. It's a 2 hours of workshop to get hands on with Auth0 SMS biometrics biometrics is like face ID and touch ID and all that stuff. Fido and more and one to 1 whiteboarding Opportunities with Auth0 Solutions Architects. Pretty pretty cool. So it's free to register, which is awesome, but you there's only limited spots.

So make sure you click the links in the show notes again. September 29th in London, October 5th in Berlin. Thank you, Auth0, for sponsoring.

So here's a question, how do you make a browser? Yeah. It turns out it's incredibly complex,

which was a shock to us in the beginning.

Browsers are Massive, massive C plus plus codebases.

So very gnarly. And, you know, there's no memory management. It's all smart pointers and stuff. So you get all these Memory issues, which cause security issues. And so, our first challenge was, okay, how do we actually build a new browser but also build a system that allows us to iterate quickly? Because if we're just, you know, futzing inside Chromiums, it just takes forever.

For example, a clean build inside Chromium without any build caching is about 4 and a half hours. So, you know, like, you switch a branch, and then you have to wait 4 and a half hours for the thing to compile.

Or, you know, even the the incremental compilations are, like, 4 minutes, 3 minutes. And so you, like, change a margin, and then you wait 3 minutes to see how it updated the UI. It's yeah. Oh my gosh.

Us developers are complaining about, like, millisecond lag times in our websites. Yeah.

Wow.

Totally.

And then there's this cultural difference of, like, you know, if you wanna hire if you want really craft focused engineers, a lot of them don't know c plus plus A lot of focus, you know, folks who have worked in Salesforce a lot are, you know, hardcore engineers who just are really good at optimizing performance. But a lot of the UX stuff, just a different scale. And so we had to first figure out that bridge. And so the 1st version of it, the 1st year, we actually took electron, and we Forked Electron and built it into a browser and then built all of the UI with JavaScript.

And our 1st prototype we launched, probably 8 months after we started, with this, like, forked Electron browser thing, with all the the UI in in React. We got some good learnings, but then quickly realized it just wasn't, performing enough.

JavaScript is single threaded, and that single thread keeps getting blocked, especially when you're loading web pages, and the UI is kinda stuttery.

Brave and Firefox have pulled it off, but it felt like this constant uphill battle for us. So, in the end, we switched over to use, Built a similar Electron out of Swift,

and went in that direction. Oh. So what are the UI elements done, and now it's they're all in Swift? They're all Swift. A lot of them are SwiftUI.

Others use AppKit. And so, it's all native Swift. And then we, you know, we built essentially an electron out of Chromium That allows us to build everything in swift with the web contents and extensions and stuff hooking into Chromium.

Wow. Cool. Oh, cool. Man, it's it's funny. We've had A couple people on the podcast, the guys behind Raycast.

I'm hoping we can have the WARP developers and, People who are building these products, and we're all just like, oh, like, is it built in web tech? And all of them are like, no. No. We tried it, but

Yeah. Does that does that mean that we'll probably never see a Windows version of this? No. We're working on Windows now. Oh, okay. I know. Aiming sometime next Sure. That's been, a can of worms too to sort of suss out how do we share code, and how do we actually get it working. So I can get into that as well. I get a lot of questions about if there's a Windows version coming, so that's nice to know that you're working on it. Yeah. That's great. What you know, one thing that can't be overstated too much, I I think, is just how Awesome.

The overall design the overall aesthetics of this app is, you know, from the moment you fired up, and there's, like, a sound Experience and and all these, like, really great things, but just the design of everything is pretty well crafted. And and I would go as far as saying, like, it's Super modern in a way that, like, doesn't feel like it's, trying to look like something else. It doesn't look like it's from this school of design or something. It it looks Really, really super good.

How much of a focus was designed for your team? And if It was a big focus, which I would assume it was. Where did you find your design designer's design work? How did you approach that whole process of making this thing look so good?

Yeah.

This may be a better question for my cofounder, because a lot of this came out of his sort of passion for product.

And so many of the designers we had met in our career already through our 1st company at Facebook, elsewhere.

I think a lot of it comes down to a few things. 1 is just realizing that the fact that we use this so much during the day, 40 hours a week, 50 hours a week, Those little interactions really matter and start to add up. And so just focusing on those heavily from day 1 because, you know, a little Tiny embellishment here and there when you're using it so often, you really notice it quite a bit. Yeah.

And then the second thing is that's just that prototyping Is you know, we don't really have specs for things. We don't plan that far ahead, more than a few weeks ahead. And most of our, design a lot of our design work It's done in code, you know, pair programming all day long. And so getting those interactions just right in code can be done because we don't have go through this mock phase To actually get the,

get the product out. Wow. A lot of these little design touches really make a a big big difference.

Even just like the little icons per Grouping of tabs that you have. It's like it's really fun to see the little I have a little, like, emoji money guy with the tongue sticking out for all of my finance pages. It's like, oh, that's just so Cute. And then when I go to that tab, the the whole background of the browser turns green. And it's just like it's so cool. There's a lot of little fun things. I think one of my favorite, innovations you guys have is like the feature of command t in a browser. So, typically, you're working in a browser. You hit command See, you pop open a new tab and you start typing.

But it seems like you guys really flip that, and and that seems kind of one of those things that, maybe it's something normal people wouldn't think to touch. So how did you come up with the command t functionality? Basically, for those of you who haven't Touched arc before, it opens up a command palette, which then kind of functions as a new tab. How did that how did that innovation come about? Because that was one of the coolest parts to me.

Totally. It it feels like 1 single thing, but it was many small little iterations over, you know, 2 years of us coming to where it is right now. I think one of the, axioms of our product team is trying to hook into existing behaviors that we already do in browsers and then making them a little more powerful.

I'll get to command t, but one other example of that is previews. Like, if you hover over, say, a Google Calendar tab that you've pinned, It shows you a little calendar preview, and you can join Zoom links and, you know, allows you to schedule stuff. And And so there, we didn't wanna make that a separate product, but rather we knew you were gonna hover over that tab anyway. And so let's hook into that And show you something along the way, so that these features are easily discoverable. Because there's there's so much stuff in Arc that you're not gonna find it all unless we have easy entry points. And so command t became an obvious one because we started with just being like, hey, why do you have to create a whole new tab? Let's open up the command palette And then create the tab when you're ready to go somewhere.

And then after that came this like switch 2 or like, oh, if we have the tab open already, let's just send you there.

And then, you know, since this command t entry point was open already and we were just hitting it so often, especially since we archive your tabs and so, you you can just create new tabs. There's no problem with it. You don't have to be, very, conservative about it.

We started adding more and more of these commands in there and continue to. You know, know, one thing we get often is adding a calculator or currency conversion or this kind of thing. We're really making that command entry entry point. Since it's so it's used so often, a discovery mechanism for other features we have. Yeah. One thing I really appreciate it because I have, you know, a system wide command palette with Raycast. Right? You know, you do Command space bar or whatever. I really appreciate that it's command t

because it doesn't replace the functionality of your command t in a normal browser. It does the exact same thing you'd expect, but it does so much more at the same time.

And I think that's just like a huge, huge plus, for

for Arc as a whole. That was one of the gnarlier parts of building a browser, which is how do you build the shortcut system When websites can Oh my god. Overrule almost any possible shortcut.

And so finding ways to have a rich shortcut system, but also not, You know, override websites or have ways that we can, like, fall back to ours.

The team has done an incredible job with, like, the customized shortcuts and how do we Deal with all of the various such cases with that. So that that took a little while. I can only imagine. I run into that on a a micro scale on my own machine. Like, oh, I wanna assign this to hyper right key. Oh, no. Wait. I have that doing something else system wide already. Shoot. Okay.

What do I do? I gotta back to the drawing board. What makes sense here? Yeah.

Let's take a minute to talk about another one of our sponsors today, which is Fire Hydrant.

Fire Hydrant has this really neat intuitive guided workflows that provide what they call turn by turn navigation or incident response. I thought that was a a pretty neat way to describe that.

It gives you thoughtful prompts and powerful integrations To capture all of your incident data and drive a useful retros and actionable analytics, there's some really neat stuff here that comes along with this incident timeline So you can automatically and easily add to an incident from Slack or any of the tools that you know and love. Never assign a note taking role and Copy pasta into a Google Doc ever again. Oh, we gotta love when they use copy pasta in in The ad right here. Before, you'd be digging through Slack channels and messages, and now you don't need to have that anymore. You get this full on incident timeline.

You can eliminate the note taker role, someone who should be sleeping instead of, just just taking down notes to try to figure out what the heck happened on thing. But now, again, you get this really neat incident timeline.

So you can get all of that and more, Including a super useful API that allows you to control fire hydrant itself.

So give fire hydrant a try because why? It's Free. So fire hydrant.comforward/ syntax.

It's free to try. Get started today, And give it a try and see if you need to level up your incident response with fire hydrant.

I got a question about The the tech a little bit more.

So you obviously didn't create an entire browser engine, from scratch, you it's just built on Chrome. Right? Totally.

Yeah. So how does that work? It's kinda interesting to see, Like Microsoft Edge.

We have Brave.

There's a whole bunch of different browsers where they're not creating a browser from scratch. We've never had more browsers, But we've never had less engines. So, do you just you just you just take the chrome engine and start building on top of that, or how does that work?

Because we were building in Swift and so much of our our we want our business logic modeled slightly differently, almost like a web app. You know? We have a Redux unidirectional data flow system, and all of the UI is reactive to that.

And that, you know, gives us obviates this whole class of consistency bugs And just makes it easier to develop and much faster. And so we really I think one of the insights we had was that Chromium, To us, it's largely infrastructure.

And so, you know, we treat the engine as such where it's just like separate pre compiled frameworks to get a lot around the compilation time, And then we hook into it from our app itself.

And so we actually it's swappable. We actually have a web kit version internally too that we don't ship because it doesn't have extension support and dev tools and stuff, but, the engine for us is is like a little piece that sort of hooks in on the bottom. That's interesting.

Wow. That's really cool. Yeah. Part of part of what is good about, the fact that you went with Chromium is that it's pretty easy to hook developers in when So many people are all you're already using Chrome for the dev tools and just pop it open, and it's all how you'd expect it to be. So that's always a a huge plus for web devs.

Yeah. Is there do you have any plans on, like, developer focused features? Sometimes you see like like, specifically Firefox pushes all kinds of developer

features. Do you would you ever look into that? Oh, yeah. I mean, it's a huge focus. I think one feature we launched recently was Boost, which was a beta feature.

And we have an ID In Arc, where you can instantly sort of create an extension in less than 5 or 10 seconds to adjust a website or change what it looks like. We don't have a sharing mechanism in there yet because we we still have to figure out some of the security stuff around it. Yeah.

But I think we're really starting to realize, okay, there's there's a Huge market for tools for web developers, where nobody's really approached them from a product point of view. They're still from an engineering perspective.

And so we've been knocking around this idea. Actually, one of our designers just posted a survey up for this. I can send it to you if any if folks wanna fill that out. But Yeah.

Adding an automated Automated developer mode. So if you go to a local host site, it'll just open up the browser window in a totally different mode that allows you to easily look at the URL.

The URL bar, maybe turning that into a sort of REST, request, you know, framework of sorts. You can edit cookies or Headers or whatever, and just leaning into adding better UI around dev tools, building in, you know, React tools, stuff like that. There's a lot we can do.

Wow. That sounds really cool. Yeah. That does sound really cool.

Here's a question. How does Arc or the browser company, I should say, because the browser company is the company that makes the browser. Arc is the browser.

How does the browser company make money? Is there, like, a a plan? Like, what's the business model here? Most a lot of our company is ex Facebook, and something we were quite used to there,

was something we heard a lot about at the in the Chrome organization too, where, so many of the user friendly features they wanted to push in the Chrome org. And we saw this at Facebook as well.

You know, to get actually to get to Production, you would have to experiment to 0.5% or 0.1, you know, 1% of users.

And if it touched at Facebook, it was feed ad conversion In any way, like, the whole population would get axed. And we heard this repeatedly in the Cromor too. You know? There was actually at some point in their development, six 7 years later, when it's a mammoth, it's really beholden to search and ad conversion metrics because, you know, it's and that's fair on Google's side, you know. If Chrome suddenly hurts search ad conversion by 2%. That's, like, 50 people's jobs or 200 people's jobs. You know? Like, what do you do? You're kinda stuck.

And so we repeatedly heard this, from folks in the in the Karam Organization. And that's that's why we realized, okay, there's this huge market opportunity.

So it's really important to us to make sure whatever, revenue mechanism we have retains that, incentive structure, so that we're only beholden to members.

That's likely gonna be some way of charging. Mhmm.

I I don't think we're very stoked about making money off of data or even search referrals. And so, it's gonna be some sort of subscription, maybe Teams based, maybe something else. We gotta figure that out over the next several years. Yeah.

That's something that people don't necessarily talk a whole lot about is that how do these companies afford to make a browser? I'm pretty sure. I mean, you could probably tell me. Is it How does Firefox make money? You know? Is it is it by they put Google as a search engine and the ads they get off? Is that how they make money? 90 I think it's, like, 98 or 99% of our revenue is from Google paying for search referrals. Yeah. That's unreal. And, like, that's Pretty crazy because that's the entire Mozilla. Right? Like, I don't they have products here and there, but, like, I'm sure it's It's a drop in the bucket compared to it. So, yeah, it's actually kinda nuts that Google pays for this well, the 2nd largest Yeah.

Browser ever, at least in my stats. So

It's wild. And it's symbiotic too because it keeps them out of antitrust, situations. Right? And so they Right. Yeah. They're willing to pay for that.

Which another reason this is an interesting market is there's a lot of antitrust Pressure right now on Google, and so it keeps, you know, all the other browsers safe somewhat. Yeah. It's so funny because at a at a surface level,

you know, An ignorant person might look at this as being like, oh, we have these rendering engines. Why don't I why don't we just build something on top of a rendering engine? Bingo. Bango. I got a new browser. Like But it's so much more deeply complex than that in terms of technical issues, but also, business issues as well. You know? How do you how do you actually How do you ship this thing? It's wild. Oh, I'm just curious.

Do you guys I don't think you do. I'm just opening up dev tools. Do you goof with the user Agent as well. And then my follow-up question is, are you using Arc right now? Because Riverside is yelling at us about you not using, it says, using an unsupported version of Google Chrome.

Oh, I'm using I'm using Arc on Riverside West. Really? Yep. Oh, it probably got some special sauce. Yeah. I'm getting an untested browser too. I wonder what that is. We don't yet actually, mess with the user agent, which is a huge Consideration for us is, like, when do we actually wade into, you know, helping with the APIs available, in the renderer, adjusting the user agent? We haven't really touched that portion, of the browser experience yet. Yeah.

Honestly, from a web developer, I I hope you don't because, Like, I I have enough browsers to test, you know? And it's kind of nice knowing that some of the browsers what about a a mobile version? Is there Do you have any thoughts on Ios having an Ios version,

both the engine as well as the UI? Yeah. There seems to be some movement on especially in the, EU around some of the antitrust movement to get more engines on Ios. But at the moment, it's just WebKit.

So we'd have to use WKWebView.

I think that's fine. You know, we're still torn on whether the iOS version should be a Browser or should be just an Arc app and, like, have all the other features that Arc contains?

And so that's something we'll have to prototype and play with to see sort of what we end up with. Yeah. I think some people would be concerned about that they can't, like, share their stuff between the 2. I don't find myself doing that, I guess, enough. But, so it's not really a deal breaker for me, but it's definitely a concern I have. Because I I posted a YouTube video, and it's funny to hear, like, Man, people got a lot of concerns that are, like, different than mine. A lot of concerns Totally. I'm like, oh, everything's wonderful, and people are like, I have to create an account for this thing. Like, I that's, like, not a I I have, like, 10,000 accounts. What's another another account? That I guess that's another question. Do you have plans so let people to use this anonymously or just with an account? I guess if it if there is an eventual, you know, payment model here, then you'd probably have to have an account. But Yeah. Account is helpful for payment model stuff. Eventually, you know, multiplayer

team stuff we wanna do. And so, once we, you know, as we wade into that, it's actually nice that everybody's logged in already. So we'll see. You know? Again, so much of our iteration is based on what people are most Complaining about any one time, and that sort of, like, rises to the top, and then you work on that, and there's the next one. And so if that becomes a major concern, we can sort of deal with it then.

Yeah. But I I think to your point on, you know, everybody has a different way of using browsers. That one that was one of the biggest shocks For us, when we started, which is there's 14 hot keys for everything, like, going back and forward. And everybody uses a different one. And everybody uses their browser totally differently.

Yeah. So, you know, working with all those different cases was was gnarly in the beginning. Yeah. What what's your mechanism for, like, collecting feedback? I'm sorry. My my camera keeps Overheating. That's why it keeps shutting off.

I don't know. I couldn't tell you why. It's it's fine in my office here, but, yeah. What's your what's your mechanism for collecting feedback. I know at this phase, it's all invite only. So,

I don't I haven't seen too many any or any annoying, like, tell us what you think, Things popping up. So Yeah. We I mean, we get a lot of folks reporting in into our feedback and bug reporter. And so our membership team is sort of, like, Hanging on for dear life, having those conversations, but we respond to every single one that comes in and read it and sort of that gets filtered over to our product team.

We take feedback very seriously. That's how we built this thing. You know? There's no grand visions about stuff. It's very feedback driven and iterative.

And so the first, I think, realization for us over the 1st year or 2 was that our instincts largely match What our members will complain about or like or dislike, I think, as knowledge workers in the tech industry, and those are initial users. And so, we generally have a good sense when something is annoying, and we have this dogfooding channel in our Slack where people are just ripping about the apart the product all day long, and, like, this sucks. This kinda thing. So much of our product development is just driven off of that. You know, employees complaining about stuff or this you know, Having entire long rants or whatever else. And then a lot of it's driven around, you know, just product feedback. People are so kind and helpful and sending So much feedback on a daily basis. All of that goes to the membership team. They read all that, but it also goes to a fire hose into Slack where people just sit there, read it, and Gives them you know, helps build their mental model around, okay, these products are interesting, or this is annoying, or this we need to simplify,

as we're going about it. Nice. Yeah. I'm a big fan of dog fooding as well. So that and your team seems to have good opinions. So so so So I've been on some dev teams where the the opinions about something are like, why is this this way? And you're just like, that's not important. Let's not worry about that. So it seems like,

y'all have a a really nice handle on it. Kind of. Yeah. I it's shaped over time. I think we had really bad opinions when we started as non browser Folk.

And that changed pretty quickly. And, oh, we should say for people listening that, like, because it's built on Chromium, you could just use Chrome extensions. And I think that's the other thing is people say, I would use whatever, but my extensions don't don't port over. And and that's Almost entirely a moot point now because you can use Chrome extensions in Firefox. You can use Chrome extensions in Arc. I think the only weird thing is that Safari Doesn't always have a extension.

Let's talk about one of our sponsors, Gatsby. Gatsby makes the hard parts of performant website much Simpler with prefetching, lazy loading, and image optimization baked in. I have my own website, westboss.com, is built in Gatsby, and I use All of those things, it's pretty sweet. If so, if you're building a headless content rich site, Gatsby has a GraphQL data layer That allows you to source content from literally anywhere. I do it myself. I source it from APIs over the web. I source some of it from a lot of it from markdown or MDX files, they have thousands and thousands of plugins. You can write your own. You name it. They got it.

Take your website To Gatsby cloud after and, it's the only cloud that's optimized for Gatsby for the fastest builds and most reliable site previews. You're gonna wanna check Out. Visit gatsby.devforward/ syntax to get your 1st Gatsby site up and running in minutes And put their claims to the test. Thank you, Gatsby, for sponsoring.

I actually I I wanna know about this.

One one of neat things is that you have 1 developing themes that you have these different, like, areas in the tab window.

You can theme each of these different groups, essentially, tab groups.

You have, this really awesome gradient designer where it's like a four Quadrant color picker of bluish, reddish, yellowish, and greenish, and then there's a graininess slider. Like, what was the process for coming up with this as a theming engine because I I would imagine most theming engines are just, you know, click a few boxes to change some colors here. But this thing is maybe the coolest,

not the fact that it has the graininess stuff in here. I gotta love some some grain sliders on something. Oh, just wait a few weeks. There's a there's a new version of it coming out No way. In a much worse stepdad.

I I wish, Dustin, who, heads our design team, was here, because that was A lot of his brainchild alongside some other folks on the team. But they spent a long time really dissecting the interaction Model what people want out of this and how to make it feel so fun and and enjoyable to actually pick a theme.

Both because it's fun to do, but also then it makes the browser feel like yours. You know? I picked that theme, and I, like, got it Exactly perfect to feel more personalized, and that makes it feel like your space when you're there every day.

And so it was both tactical and just

know, wanting to build a really fun experience. Yeah. And accessibility folks out there, you know, a lot of times, something that gets really, forgotten on about these little accessibility tweaks. Right? You're having to pick the color and the background size. One thing that you'll notice in Arc is that if you you're dragging the slider for the colors, The moment you pass that threshold, the font color changes from white to black automatically.

It's not something you have to touch or think about, it's like little touches like that. It just, like, makes it besides the interface itself, but it just makes it flawless. You can't make a bad looking theme on this thing. It just Looks nice for you without, giving you a bunch of preset themes. You're still getting that ownership of it. So yeah. Just great job. I don't know. I know. I could gush about this thing all day. I think I could spend a whole lot just saying nice things about this. So either way, these are our supper club questions or dessert questions. There are Our finale questions here that we wanna ask you the same questions we ask everybody who comes onto the show. Wes, do you wanna take the first question? I do.

What computer do you use?

I use a 2019 MacBook Pro Sixteen inch that constantly overheats.

Yeah. It's the it's the worst performing computer we could find in the office, mostly to test performance.

And so I haven't upgraded to an m one yet.

Has anyone on your team have a m one and has tried to compile the browser with it? Yeah. Most folks do 4 specifically compilation time. You know, it's such a drastic increase in compilation time. Most people have switched over to m one pros. Yeah. Interesting. I was always I love seeing that. Like, I looked at my little JavaScript start times, and they they got really fast. But it's it's always interesting to see people who have, like, Hours. They can save hours, which is unreal.

It's a it's a massive difference, actually. It turned out to be pretty worth Totally worth the money to replace the entire team's computers once the m one pro came out.

Yeah. I think about that a lot with mice being, like, doing 3 d rendering and stuff, and you just leave it on a computer overnight. I, like, wonder how how big of an impact that's had for them. What about a keyboard? Do you use any sort of fancy keyboard or, just stock?

I, I used to use one of the crazy, Kinesis Advantages for the longest time. Oh, yeah.

Because I had pretty bad RSI, but now the, regular Apple compute Apple keyboards work great. They're so flat For whatever reason, they don't trigger my RSI.

So I would love to use a mechanical keyboard, but these are these are the ideal these days.

What about, Put, text editor theme and font.

Yeah. These days, we do all of our coding in Xcode.

So there's a lot of that. But thank God it has a VI mode now. Besides that, I used Neovim for the longest time. Oh, yeah. Color. I think I used solarized dark. And then Okay. I think it's just the default.

I don't know what what Xcode has. It's the default,

text. Yeah. You get some people who are really into changing them. I've set mine once a while ago to be like Mona Lisa, and I just kind of, like, set it there, and then I'm sorry. I've gotten so used to that. Now looking at anything else feels, wrong.

It's funny how your brain accommodates. Yeah. Whenever there's a different Oh, better exhausted.

Classic. What about a terminal and shell?

We use warp.

You know, we've got a lot of respect for what they're doing, and so, we've been using Warp internally. It's great. Oh, that's good. That's good. And just with Bash

or Fish or Z

Shell? Z Shell. I've used Z Shell for a long time. And now I think it's the default on macOS.

Yeah. Yeah. I I remember they switched over to that a while ago. It's it's a good one. I'm a big fan.

Here's the interesting one. If you were to start Or coding from scratch, what would you learn?

Oh.

It's hard to argue against JavaScript in this Yeah. Day and age.

I feel like native programming is making a little bit of a comeback. I mean, you know, certainly mobile is all there. But then, I don't know, with Flutter and stuff, React Native, maybe not That much. Yeah. Yeah. JavaScript ecosystem, I would probably start with, and then, Swift or Kotlin as a as a solid second. Yeah. The Kotlin The call the multiplatform stuff is really interesting. They're really pushing that forward where you can Netflix is, I think, where you can use it for any platform now. That's Yeah. Kinda feels like when Kotlin got into Android,

its usage got picked up a whole ton. What about, like, staying up to date? Do you have any resources? Where do you where do you go to check out new stuff?

And Hacker News is the classic. I do have a few listservs I, really like.

Changelog.com has a bunch of that's a listserv that has a lot of interesting stuff that comes in.

Every.

That's more business than programming, but, like, that's also a great one.

So, yeah, Hacker News and Listers. Awesome. Okay. Here's a question that we normally ask about, like, web web stuff, but I'm gonna ask you about browsers. Like, what's what What are you most excited for in the future of browsers themselves? In terms of, like, web technology,

I'm still very bullish on, Assembly, I don't think we've seen everything that can the impact that has, on the ecosystem as more and more native applications move to the web.

I you know, the industry is moving in that direction. That was a huge reason for us to work on a browser.

I'm excited about how we can we can help that. I think Less so in terms of the the APIs available to the web, but the different sort of containers you can see your applications in. You know, as if we have different types of browser windows, this kind of thing. Yeah.

A good that's a self serving answer, though.

That's great.

What else? And, you know, the JavaScript ecosystem so quickly that it's tough to say one thing. But, you know, the bun and Dino are interesting as well. See how long they last. I'll get that. Yeah. We've been talking about that a lot A lot lately is that,

you're starting like, you you think, like, oh, yeah. We we're in a really good spot now where we got a couple good browsers And we got a a good JavaScript runtime, and then poof, out of nowhere, we've got all these new browsers. Explosion. And then we've got all these new, run times. And it's not that they're changing it. It's that they're saying, okay. Now that we have somewhat of a standard, like, how can we iterate on that? Yeah. We have stability. Let's push it. It says Ben Thompson has this, like, bundling and unbundling phase

thing. Right? And so it's, like, seems like we're in a bundling phase with Yeah. Deno and

Browser. This kind of thing? What else? Yeah. Oh, sorry. One one last question. Where can people go to stay up to date about new features and stuff that is, coming to Arc. Yes.

The main place is we have a Twitter handle, Arc Internet.

I would Please follow that. That has if you're a member already, that's a member community, and that has a lot of prototypes, features, you know, animations, Stuff we're sharing and what we're working on.

And then you can always follow the browser company,

Twitter handle as well that has, just more broad announcements on it. And web developers out there, even if you don't plan on using this browser, go to their website because it is a it's it's website inspo, to the max. It's one of those really gorgeous websites. So check out their their site overall if you're, if you're a dev out there.

The browser dot company. Just launched our Arc website too, arc.net. So you can check that out as well. They did some fun stuff there with the animations. I'll have to check that out. Awesome. Let's move to the next section, which is, shameless plugs and Sick picks. So a sick pick is you literally anything you can pick that is sick. So it could be, an app, a chocolate bar, a hat, like, anything in your life that you said, oh, this is really handy.

Scott picked the knife sharpener on our last one.

This is kinda boring. It it started from, you all mentioning getting a mic, but I got, like, one of these little Jabber things, and it's changed my, like, Zoom calling significantly.

I, like, don't have headphones on at the moment. It feels much more like the person is in the room.

The voice gets through really well. It has a good speaker built in. So, very boring and very, like, I don't think it's that sick, but it has changed.

What is it exactly?

It's like a little puck. It's just, like, little Jabra puck, and it it's not US, Bluetooth.

It, like, plugs into your computer, and then it's just like you you talk as if you're in Like a conference

call. You put it on the table. Yeah. Yeah. Totally. Yeah.

Well, It's funny that Jabra has come so far. Like, weren't they the Yeah. They they used to be a sick company where they had the little The Bluetooth. You're set. Remember people moving around with those? Yeah. Jawbone.

Jawbone. Yeah.

That would be sick. Yeah.

We should bring the Jawbone back.

That would That's true. Imagine yeah. Bring up style Jawbone. Yeah.

Oh, I saw someone the other day wearing, one of those those jeans, the really, really white Jinko jinko. Jinko. Yeah. Yeah. Jinko.

Yeah. My mom didn't let me have any of those. Yeah.

Oh, that's great. Awesome. Well, that's that's Sick pick. We'll move on to the next section, which is shameless plugs.

What would you like to plug? Where can people find you? All of that. You got the floor.

Most of the browser. Please check it out, arc.net.

Sign up for our wait list. We're taking people off the wait list pretty pretty quickly, So it's not a long wait a week, a few weeks at most.

And we'd love to have you try the browser.

Check it out, send feedback.

And we, you know, we iterate on it pretty quickly. And so, if you try it out, please send feedback. Please send what you like, what you don't like. We largely base our product off of that. And so, we'd love to make a browser for you. Awesome.

Yeah. And, you're doing a great job so far. So everybody out there, if you haven't tried to get on that waitlist, sign Sign up, give it a try, and this is not me, chilling for any reason other than the fact that, it's gorgeous. It's an awesome experience. So Shout out shout out to everyone at, everyone is working on Arq. It's a lovely product. It's very much a team effort. Just wanna plug the team as well. They're incredible.

We're very bottoms up as a company, and it's been this entire team's work, over the last two and a half years. I've got this here. That's great. Thanks so much for coming on. Appreciate your time. Have a good one. Suresh. Peace. Take care. Bye. Later.

Head on over to syntax.fm for a full archive of all of our shows, And don't forget to subscribe in your podcast player or drop a review if you like this show.