Mini Panel June 2024
===

[00:00:00] Welcome back to PodRocket, a web development podcast brought to you by LogRocket. LogRocket helps software teams improve user experience with session replay, error tracking, and product analytics. Try it for free today at logrocket. com. I'm Emily, producer for PodRocket, and it's been a little while, but we're back with a little mid year panel review going over what happened in the first half of 2024 in the web development world, going back to some of our predictions that we made at the end of last year, see if anything has happened since then. And then we're going to go through some hot takes at the end. And today we have our OG PodRocket host Noel and Paul here to talk about these things with me. How are you

guys doing today? Woo.

Doing great. Happy to dig into some of the madness of web dev today with you, Emily.

Oh, this is madness, isn't it?

All right. 

I'm good. I'm excited as well. We'll see how it goes.

We'll [00:01:00] see how it goes. Okay. Let's get into our first topic. 2024 is already halfway over, which is honestly a little crazy. We're officially halfway through the year, and like I said at the top of this, I think it would be good to reflect on some of the things we talked about at the end of 2023 going into 2024. So first off the bat, ~what have you both seen personally that you thought was influential or has promised to become? Let me rephrase this question. ~What have you guys each seen personally in the first half of 2024 that you thought was influential or becoming a bigger product, idea, topic, et cetera, in the back half of 2024?

I can start. I feel like ~my big one, ~this has been the year of react server components thus far. I think I've been surprised at how much ~like ~the dev community seems to have coalesced around them. I think Outlook's pretty positive right now. I'm sure we'll get into specifics more as we dig in here.

There's going to be a lot of chat, but Yeah, it's just been interesting to see, I think a little bit surprising maybe about how much progress and chatter seems to be happening [00:02:00] around server components.

Was there a specific instance that you We're like, Oh, this is clicking for ~web dev, ~the web dev community as a whole.

Ooh, it's hard to point to a specific one. ~I think we've, ~I guess maybe ~like ~we've been redoing a bunch of our internal stuff, like a log rocket website. ~And it was ~you know, we were vetting tools out ~and stuff ~and one of the other devs on the team without too much prompting from me was just like, I think we should do this, like just seemed like the easy way forward.

So maybe that was what kind of made it really pop for me, ~but it's I don't know. ~I think it's primarily anecdotal and just like how much, chatter seems to be around it and how it's generally positive. And just like how much of the kind of headspace it seems to be occupying.

Yeah. It definitely seems like it's now becoming just like more mainstream. Like you were saying, ~like~ you're using it in your everyday development now, which is cool to see. Paul, what about you? What have you seen in the first half of 2024 that you think it's going to explode or is really promising?

I was going to say the same [00:03:00] thing. 

Uh, and then Nolan first. My anecdotal evidence here just to pile onto the bandwagon is when a project manager asked if we could use RSCs for something and I went, whoa, didn't think you would ever say that, but here we are. I think that there's two strata to where something's like mainstream for me.

And this is just my own personal, this has no relevance in the real world, but are the YouTubers talking about it? That's step one. And then are the project managers talking about it? We're already on step two. So to me, it's like a sweeping acceptance of RSCs, and I also like to see some of the Documentation and training out coming around it.

I'm not like an all star web dev. I make apps I enjoy making apps, but when it comes to the nooks and crannies when something goes wrong not having to go all the way Solo often off of some Next. js error is really nice. So there's more documentation and the project managers are turned on to it.

It's feeling like it's getting a lot of traction and the [00:04:00] traction seems to be positive for my own experience is positive. I think not having to deal with the queries from the client side back to the server is it doesn't sound big. But when you're just trying to slap something together, it's so powerful.

You can make something so quickly. I want to spend as little time possible coding, so it feels good to me.

~So what, and maybe ~going into the back half of the year now What do you think that you would want to see from web dev? Obviously RSCs are becoming more mainstream and everything. They're doing really well. Was there anything ~like~ disappointing or didn't live up to your expectations? The first half of 2024 that you're like, hopefully if they do X, Y, and Z, it would become more of a thing in the back half.

I would love to take this one first this time, which is, I want to see Dino and Bun do the things I want them to do. They do a lot of things, and they do a lot of things right, and for some scoped projects, having used Dino, it was [00:05:00] like such a breath of fresh air. And this is going to tie into my hot take at the end of the episode today.

there's like a lot of business processes that like at the time when I first tried out, I just wasn't able to grapple with easily. Maybe there was a solution to do it. But we've talked about here on pod rocket, for example, they're like, Having that lead foot on the gas pedal when it comes to actually porting over different node APIs that you'd want, like essentially now everything is compatible.

So I really want to revisit Dino see what ~you, ~we can do. And I'm sure it's just going to get stronger in the later half of the year. So that's the big thing I'm looking out for.

~I don't know. I'm not, ~this one's a little bit tricky for me. I think that there still is a little bit of confusion or just difficulty. And I think we've talked about this in the last episode as well. Just like the ramp up process for front end devs. Is weird right now. If you've never done this before, I think that the push to have things rendered on the server and like doing things at build time versus runtime and all this [00:06:00] dialogue is still really hard.

And I think it's been hard to even like land on ~what, ~how we prescribe this to people that are not familiar with the space because everything's been so in flux still, like even the tooling, if you know what you're doing, there's still a ton of decisions now on or at least there were on like how to get a project set up.

So I'm hopeful that ~kind of I don't know,~ like the recommendations there the, like the tutorials, the way we tell people to get going can be simplified. I don't know if it's going to happen quite yet. I don't know if we're quite on that point on the curve yet, but I think it'd be cool to be back in that space.

On that topic, Noel, I know we're going to get into some of the new Next. js stuff today, but they just slapped us with, what is it, partial pre rendering where like someone can be, I'm like, oh my gosh, there's another thing now? You're kidding me.

Yep. Yep. Yeah. Server components, without a server with the server. ~It's just ~yeah, like it makes sense if you've been here for a while,

It's really funny because my partner Diana, she's learning web dev right now and she's succeeding at it. I think she's on her third Next. js project right now, but I get to see this [00:07:00] process of somebody learning web dev for the first time and ~it's ~I feel so sorry for her. It's really tough.

Yeah. I'm sure.

Yeah,

I see it all the time.

I just, casually like learning all this stuff. I can't imagine actually doing it. So kudos to your partner. ~All ~I think my next question is obviously, and I hate to bring it up, but AI, I know I was in a marketing meeting yesterday and my VP was like, honestly I don't think any of us saw how quickly AI was going to be so a part of all of our lives.

~Even from a marketing perspective. It's insane how much has changed about our jobs within the past six months to a year. ~So I want to get from your perspective~ as we always do from the web dev perspective~ how do you feel about how AI has so quickly evolved this year? ~How are you seeing it?~

Do Like how it's being implemented to tools and features for web development. And is there anything you don't want to happen in the rest of the year when it comes to AI or worried about when it comes to AI? I~ I~

~interesting. ~Yeah. I don't know. I'm not ~worried~ too worried about it. I don't think, I think that ~we have,~ everyone's got opinions on ~like ~Gemini, like search results, having blurbs. I think that affects everyone from ~a, like ~a [00:08:00] business perspective. Maybe not quite as technical.

There's probably some technical considerations there. Just like, how do you change your projects and stuff to either work with that well, or circumvent it to make sure you keep getting clicks or whatever it depends on. Your market industry. But I dunno, from a tooling level, co pilot still nice. I don't know, saves me some time. But I'm not like reaching for tons of stuff in ~like ~the development process itself. It's changed how I look stuff up, but beyond that, it's a lot of like kind of integrations and stuff that we've tweaked a little bit. So it's like tooling layers to help pull data from things in ways that would have taken a lot of work before.

And I think we'll continue to see that AI. Applied to a specific domain to help with a problem. I think we're still in the phase of those products coming to market. So I think we'll see more of those, but I'm not ~like ~worried about them. I think that those are cool, ~like ~interesting, fun, new dimensions and ways to parse data easily that we've never had before.

mean, we always do come back to the conversation of AI is a great tool [00:09:00] as long as we're using it

correctly and Paul, what are your thoughts?

I think I mostly agree with Noel. I feel excitement Especially if you're willing to play ball you can get ahead of the competition ahead of the curve There's just so many people I see not only in tech especially in art and Emily you might relate to this with art friends But people are super anti AI, but if anything I see this as an opportunity It means if you are AI you're gonna beat them It's a little narcissistic, but that's exciting to me. And yeah, on that personal level, I think more at its core, though, I am scared of the AI revolution. And it's not primarily like software or your turkmerderbs. It's more robotics, for example, like right now, NVIDIA, they have technology that's being developed where they can run a humanoid, the humanoid can try to move an arm up, try to move the other arm up.

Oh, no, I wasn't able to do it like let's readjust the motor movements and the way this character is rigged and they [00:10:00] can run millions of them at the same time for a million years. There is no way that a human being can probably get close to the level of finite detail and just like fine tuning that these robots can express and I've seen some demos of how these new humanoids are going to walk.

That is scary to me that it's more not software and I don't know how those things are going to be used, but I don't feel comfortable with a bunch of AI. Physical. Things. Existing in my space.

~That's fair. ~That's fair. ~I have Very strong opinions about AI art, but this is not the forum to get into that, but I see your point. Alright ~we're going to get into React 19 and Next. js next. But we talked about web components and~ the surge ~the surge of popularity At the end of last year, it seemed like ~everyone was already ~everyone was again, very hyped about Web Components. Did you see that following us into 2024, or do you think it's just mellowed out again?

I think it's mellowed. I think it's like everything where there's like excitement around it, but I feel web components are one of those things where They're probably not going to be a thing that most devs are like thinking about consciously most of the time. Like those are good.

They're going to end up being [00:11:00] implemented. And ~that, that, ~that layer of abstraction is going to be. ~Now ~probably the frameworks domain, right? Like for a lot of how we think about this, at least I suspect. So I think like we talk about it, devs get excited about it. Like it's a cool thing ~tech like ~technologically, ~like ~people aren't going to use it day to day.

People aren't jumping into it just because it's not an abstraction that's super helpful yet. So I think it'll ebb and flow, but I don't think that's ~like ~for lack ~of kind ~of. Viability. I think it's just not a thing that most people are going to be dealing with most of the time. So it's not always at the top of everyone's mind.

Have you played with them, Noel?

Haven't touched them.

yeah, like I've done like my little hello worlds and stuff. I'm like, look, we'll inspect it in the browser and stuff like that, but not like anything.

~it ~How did it feel when you did your Hello World?

Like a little more work than I wanted to be doing. ~It's ~it's, uh, like I was doing more to get, my baseline established and I quickly had this, I guess this is what formed his opinion where it was like, I feel like I shouldn't be the one [00:12:00] that's doing this.

If I'm going to have to be using a framework anyway, like if that's the future of web dev, I feel like I'm like writing assembly. You know what I'm saying? It's not that extreme, but it's this is not the level I want to be developing at. And maybe that's like a hasty.

Opinion, but we'll see.

Any thoughts on Web Components, Paul?

I don't have them because I never tried them. 

Oh, that's right. 

Yeah, we talked about them on the pod here, just learning about them. ~It, ~it was like I don't feel compelled to go build with them. And I feel like a lot of people might end up in my boat or Noel's boat, but that doesn't mean they're bad.

Yeah.

And that's going to also tie into my hot take at the end of the episode, which I can't wait to lay down.

~Very excited for that. Cool. Any other last thoughts on first half of 2024, what you're looking forward to in the rest of 2024?~

~I think we'll cover a lot~

~I think we will too.~

~in our, in the rest of our outline here. So ~

Alright let's get on to our next topic which is It's probably going to be a controversial one or not. Who knows? React 19 and Next. js 15 RCs have recently been released. So Vercel's ship just happened in late May, and as usual, there were some big announcements, including the release of Next.

js 15 RC [00:13:00] feedbacking on the release of React 19, the React 19 RC. So Did any of the features or the new releases that they highlighted at Vercel Ship stand out to you? Do you feel like it's more of the same? What are your initial thoughts of Vercel Ship Week?

My boss is getting aggravated that I keep rewriting the same damn app every 60 days.

Can you go into that more?

This one came fast. They keep coming fast. It's not necessarily breaking changes, but. We talked about dev ramp up friction. I want to throw into the mix maintenance friction. there's value in not the opposite of change. What would be the opposite of change? I'm searching for a word here of stagnation.

There's it's not stagnation though. Like it.

Like we're not improving. We should improve, but we shouldn't change. There we go. Thank you. No. Oh my God.

there's value in stability. Do~ Do~

value in stability,

~Do you feel like ~do you feel like companies like Vercel [00:14:00] are, what's the word I want, are trying to be~ sorry, I just got a weird~

like a little too revolutionary.

I wouldn't even say revolutionary. Do you feel like they're trying to be viral as a way to push their products so it feels like they're releasing things all the time? And this isn't a knock on Vercel, but sometimes I see as a marketing perspective, their marketing is incredible. But from a web perspective, because you're saying, okay, all these constant updates, it's like, I'm constantly doing all this work, ~do you feel like, ~do you feel like web dev companies, products, etc, need to pull back and focus more on, functionality, rather than the glitz and glam of constantly Yeah.

I think that's the next thing we need to work on is releasing things. 

~What is, ~I agree. Yes, Emily, but it also makes me ask the question, what is functionality? Because to them, they could say functionality is being able to serve this extra, 0. 05 percent of the customer base that really needs this. And everybody else could use it and benefit from it. But like this 0.

05 percent really need it. To make ends meet, which is like [00:15:00] a new piece of functionality, but maybe there's another sort of flavor of functionality, which is the functionality of rhythm and discipline and sanity for like long term involvement in a project without having to change team and bring in like new blood all the time.

So I, maybe it's a different flavor of functionality.

That's tough. Yeah. I like from a, what is the appropriate cadence? That's an interesting question. ~I think it's, ~I think it's weird in that we're going through this push where we've never had as much attention really paid to this kind of like glue layer that I think next is ~it's ~a lot of things, but I think it's really trying to smooth out ~the, like ~the rendering on the server relationship and being able to do that in a similar way to how we think about code and the client historically, ~like ~we've been doing this For a little bit, but I don't think it's ever hit critical mass in the same way before.

So I understand why there's like a fervor and there's all these new changes and optimizations and things all the time, but it is exhausting. [00:16:00] So I don't have a good answer. That stuff's going to change a lot. But I think it is a complex space. So I get it. Like I empathize with why ~the, like ~all the features make sense.

For the most part, I'm like, yeah, I get it. Like I understand why it's there. Some of them were cool. This is neat. This is a time saver for sure. But it is I don't know, maybe it's that the term's going to fail me, but it's that same thing. We're like, when do you launch a rocket?

Because we'll probably develop a faster rocket sooner. That'll catch that old rocket that you launched. Like in, we're talking about~ like ~space exploration. So I feel like it's that, like, when do you go all in on a version? Cause it's like, when is it worth my time? Yeah. To say, this is a reasonable feature set to start with.

And I think that is tricky is like with the current cadence, but that's not stopping me from, trying to do the latest thing. So maybe it's fine. ~Yeah.~

I like to think, I love the rocket analogy, Null, by the way, it really like paints a picture in my head. And it also makes me think about another flavor of web dev that is running around today, which is like PHP 5 came out, and it's pretty good. I don't know if you've tried to use it, [00:17:00] but it's actually slightly ergonomic.

They have like functional programming, mapping and reducing and just stuff you would be used to in the JavaScript TypeScript land. And one great thing PHP has done is even with this ground shattering update there and adding features, it's still. Like very old school PHP and you can still make great stuff with it and that it's that Unrestriction that or like the weird cowboy ness of just eating code up there that I miss It's it gives you that creativity as a web dev versus feeling like I'm working in the Pentagon

Oh, ~yeah. ~Yeah. Make like always, am I doing the wrong thing all the time? Is like not a good kind of dev work. It's not a super fun anyway. Like it doesn't feel like it's hard to find joy there. I think so. Yeah, I get it.

~Let's get inbefore~ inbefore we get into some of the new features I do want to touch on what everyone talks about when a new Next version comes out or a new React version. What do we think about the increasingly intertwined relationship that React and [00:18:00] Next slash Vercel have with each other? Do we feel like it just makes sense at this point?

Do we think that there should be a separation? The React 19 RC came out and then within, I don't know, a few weeks to a month Next JS 15 RC came out. So what are your thoughts on that?

is entering the party, too I mean we got react router you can remix your remix with react router now.

Which we will be talking about later.

Yeah.

So love the spread

I think it's both, right? Like they are tightly coupled. ~It's,~ I think there's a lot of benefits there. There's concerns, which we've talked about at length in the past. Find those old episodes. But I don't, I'm not sure if there's much anyone can do at this point, other than if you don't like it, don't use react.

~That's the, ~that's the play. Yeah.

Hey, love to check in on this question, but it seems like it's getting shorter and shorter responses every time. I guess they're together forever now. 

Let's get into some of the features. 

~They included Platform wide feature flags which includes faster load times through EdgeConfig overriding feature flags from the Vercel toolbar and are embedded into Vercel's web analytics. Do you think this was a good integration to add to the entire platform? Does anything stand out to you? Is this helpful for you or is this just I don't know. Another feature.~

~I don't know much about this ~

~Yeah, that's fine. We can also skip this one.~

~sure if Null, you've dug it into the~

~I'm not, we don't have anything at that scale. Like I looked at the, I looked at the release notes and it was like, yeah, seems useful, but I don't know. Yeah. I don't have strong. ~

~Let's skip that one then. Oh Vercel is also continuing to push their VO Platform~

~feature ~

~Is it V0?~

~always be zero in my head. ~

~I always assume. Okay. Yeah, ~

~I'm just, not a tech bro. So I would say, Oh~

~but yes, V zero. ~So Vercel has been pushing V zero Jared Palmer gave [00:19:00] a short talk about it at the Vercel ship keynote. They're continuing to grow it which V0, for those who don't know, is an AI, do one of you want to kind of give an overview of what V0 is?~ actually, do one of you want to give an overview of what V0 is? Just so I don't say the wrong words.~

~got it, Paul.~

So V0 is a LLM powered ground zero. Meaning very beginning. React code generator and ~it~ you can give it a prompt kind of tell what you need Of course the more descriptive you are the better It's gonna just spit out three different versions of like you might want to start with this layout It actually is pretty basic It doesn't do that much but it kind of highlights how difficult it can be sometimes to start Just starting a project can be hard just getting something on the page and then oh look I already have this Let me like build on that v0 aims to get you there helping you start it does a fairly good job at that, too

Which is a really good example of how AI can be a great tool for us to use and not going to take our jobs. Jared Palmer [00:20:00] also said that Continuing to grow v0 will bring everyone quote one step closer to industry standards for inclusive user interfaces. What do you think of that statement?

I totally agree with that and I love that statement because in my experience using V0 as a mediocre like web dev who just likes to make projects time to time, there are things I have picked up and learned about the proper structuring and semantic use of HTML elements that I would not have otherwise reached for.

Or maybe I knew about them, but V0 says, Hey, I'm going to use these IDs and spans and structure it this way. And then I'm like, Oh I guess I will continue because.~ It already~ the app has started that way. I don't want to ruin the nice layouts. So it has taught me things and I am grateful for that.

And it has definitely set forth some standards for me that I would not have otherwise put emphasis on

Yeah, I think it is nice. It's always nice having a good example and using that as a starting point, right? It's oh, okay. I know the things I need to be doing. [00:21:00] If I'm introducing a new component, I can quickly go figure out how to merge that into how I'm doing the rest of this but still be, have all of the formatting and attributes we need for accessibility and stuff like that.

~I'm not sure if this is the ~It feels kind of band aid y, in a way it's yeah, we can keep pushing devs to do it. I think that there's, still a large place for Linting tools, and tools that run in the browser, and look for these kinds of things, like extensions and stuff like that.

But, I think that this is another start from a good point and then have less work down the road, set yourself up for success. so yeah, I think I agree. I think it's a step closer.

the great educational tool as well. Like I'll refer back to my partner who's getting into web dev right now, supercharges that learning experience because v zero places ~you puts ~you in this place where it's like what you get isn't usable. It's just not. And so you have to learn how to read that code, dissect the code, and then make it usable.

And that's a great learning experience for people just trying to get their hands wet.

~it's a really good take. Awesome. ~Last thing for Vercel Ship Week Vercel claimed that [00:22:00] their AI SDK does the same thing for LLMs as Next does for React. ~Do you see the AI SDK becoming the standard aggregator of LLMs? And actually, hold on, let me rephrase that question. First question. ~Do you see the AI SDK becoming the standard aggregator of LLMs? Just as Next is for React in this analogy.

Yeah, no.

no, I don't. ~Like~ it just seems like it's not even the same scope, but

hmm.

Maybe I'm just pretty 

Maybe someday. 

~Yeah.~

~Yeah, go ahead.~

I also want to add GPUs are getting good, they're getting bigger and fatter, the VRAM is going up. The fact that I can already run Llama 8B on my MacBook Pro Max is amazing. Blows my mind and it's good. These models are just going to get better. And I sincerely believe that there's a majority of tasks that we're going to want to do as devs and creators, that the open market for these open source models is going to explode and it's going to fragment.

This like current monolithic AI, like mantra that we have going on. The fact that, no, I don't [00:23:00] know if you've heard about this. I think the twin, the 50, 90 TI, which is still a consumer car that's coming out is going to have over 30 gigabytes of VRAM per card, which is nuts.

Yeah. Wild. ~Yeah. Yeah, that is.~ Yeah. Yeah. Apple's doing a ton of work. Like all of their recent patents, it seems are all like, little AI oriented chips for ~like ~local processing and stuff. Like I just, I think it's too early. Like we don't know the abstractions yet with which these things are going to most meaningfully operate over.

Like it'd be like, I don't know, J query like was for web. You know what I'm like? I feel like we're still in that era. Like we just don't know the way in which ~we're going to, ~devs are going to want to interface with these systems. So this is going to change. I just ~can't, I ~can't imagine ~these, ~this current version of SDKs is ~like ~going to be the interface.

That we're going to use forever. ~Like ~maybe, but ~I'm ~I'm skeptical. ~There's just, I'm like~ it seems like there's going to be major iteration here and just how these are even, yeah. Are they always in the web? Are we going to start doing ~like ~hybrid models? Or is that going to become the norm?

Maybe the SDKs [00:24:00] will adapt, but it's just like, there's so many unknowns still.

And then you can look at some of the work that's been done to abstract and build these SDKs on top of the LLM layers, such as Lang chain. And you go, I've dug into the source code. I've written my own modules and stuff. And ~it's ~that is a horrifying planet to land on is the lang chain source code.

Those people know LLMs. They don't know software engineering and So like you read that and you're like, wait, this is what people are using, man. We have, there's ground to gain.

~Awesome. Awesome. ~Last question. What do you think of quote generative UI? And the claim that react server components and the AI SDK will be the ideal architecture for streaming these next gen user experiences, or is this all buzzwords?

Maybe it is a clear path to a pretty optimized. Configuration right now, right? Like with what, like with what's available and ~how. ~How most products are probably leveraging LLMs right now. Like it is a clean Oh yeah, this is pretty performant. This is [00:25:00] a logical architecture, but again, like it's the same kind of problem.

I think this is all going to be dynamic and we'll be changing for a while.~ So I don't think it's bad, like like bad like use ~using next with generative UI and letting the LLMs. Run however, wherever you want them to be running on the server. ~I, it'll be, ~I think it will be pretty well optimized compared to other solutions, but is it going to be the best forever?

Probably not.

~Any other closing thoughts on Brazil ship week? Cool. Awesome. All right. ~Let's move on to remixing. React router again. At react conf in May ~Remix announced that they plan~ remix announced that the planned release of remix v3 is now going to be released as react router 7 ~Ryan Shit, sorry. Who is it? Ryan, not Karniato, Florence. Hold on. Let me just, I didn't write it down. Yeah. Okay. ~Ryan Florence wrote, if we just ship a VEET plugin for React Router, the two projects could be merged. So that's what we're gonna do. What are your thoughts on how Remix is now ~officially emerging? Excuse me, ~officially emerging into React Router, and do you think it was a good move to merge?

I think if the remix team thinks it was a good move, it probably is a sign that it was like I'm not a huge remix user. I've written a couple remix apps in prep for [00:26:00] episodes covering remix, but that's it, but I, this makes sense to me. I understand why this is like a logical step. ~That's not, I'm not, I don't, ~I'm sure there are some out there that are like.

Lamenting more unification centralization of these tools, but I get it. I understand this move.

I guess that also goes into we talked a little bit about it with next and react, but like the unification of these tools. Are there any downsides to this? Cause obviously there's a lot of upsides. It means you probably get more funding. It means you get more capacity to work on these tools, to get them to wider audiences. Do you find any downsides that might come for this consistent. Merging, it seems like obviously React is a huge language platform framework, whatever you want to call it. But what are your thoughts on that?

I think remix was interesting in that it was in that swath of tech that was trying to solve the like framework platform problem without necessarily being inherently [00:27:00] tied to a given like technology, like a given, framework. ~So like there, there's a~ I think that's a little bit of a loss.

I don't know if there were many people doing remix and like trying to make it work with, view or something like, I'm just not sure, but I know that was a goal originally. And I just think that is increasingly becoming. Not super pragmatic to try to do as a developer anymore.

That's just got to buy into a framework and a platform, which is not, doesn't feel super great. In that, I guess we're getting back into that prior question. We didn't want to get into, but in that like open source, I don't want to tie my project to a specific provider or, like corporate entity that I am then beholden to.

So I think that's the. That's the loss. 

~Do we feel like this is not like an indicator of the what's the word I want? Not undemocratizing. The word I want is sorry guys,~

~hegemonization. That's all right. That's all ~

I definitely think it's a good thing. Actually, one of the enterprise apps that I'm helping build right now, and it's huge. It's a huge app. It's all in Remix and the Remix router is used pretty extensively. And just nothing [00:28:00] special about it, and if let's go back to when humans were prehistoric.

We accelerated because we did specialization and we got artisans and stuff. And I think in a similar manner, Remix should specialize in what it's good at. which is using web standards very effectively to take a lot of the burden off of state management data freshness, and all these sort of like benefits that if you're reaching for Remix, that's why you reach for it.

So let Remix be good at that. And then don't put the burden on me to get then go learn another router. It's just a common language thing. They're basically the same thing. Like they do all the same stuff. So let's speak the same language and focus, let them focus on what they're good at.

Awesome. What are your thoughts on React Router being the replacement for CRA? ~Which ~CRA has been ~off, ~ a little while, correct? ~Obviously, ~what do you think of this change? Do you think this is a good next step [00:29:00] for replacing CRA or does it matter?

Again, like I think now in, in this era where ~it's ~you are tied to a platform, right? Like ~it's man, ~it's mandated by the machinery of the thing you're building, right? Like the templates are need to know what they are and where they're going. I think like it is, it's inevitable, like something, there needs to be some tool there.

I don't know if I've got like super strong emotional feelings or anything, but like there, it's, is the correct spot for a tool like this.

I also love seeing what the open source community is putting out for app starters. ~I personally have used So, I've, let me rephrase this, I said so weird. ~I have personally used so many different like app starters other than CRA that have brought me great success and I'm in particularly thinking of the T3 stack. ~Does that, ~maybe that uses CRA under the hood, but it does a lot of other special stuff and that just makes me think of, there are folks out there who have an opinion about how they want to set this thing up and they're going to put out a tool to use it. Once you find one [00:30:00] that works for you that's just the end of the convo.

Go use it.

Yeah. It's a tool. No, one's holding anyone over the fire here,~ feet over the fire. It's yeah, like ~it's handy. It's nice. If you need this, it's there.

And it'll probably still work. ~So~

~What do you think last question. And if, again, if you feel like we have nothing to really talk about here, that's fine, but ~do you think that this consolidation of react router and remix bridges us to the quote, future of react like Ryan Florence says during his react conf talk And what do you think of the incremental nature of this phase of merging Remix and React? It seems like they're doing this rather than doing full rewrite. Do you think this is a better I guess this kind of goes back to our discussion about, ~like,~ how Brazil is constantly turning out. Updates. Do you think that Remix going about this in a more incremental way is more helpful for developers?

Doesn't that matter? What are your thoughts on that?

my general opinion is if you speak the same language, it's helpful for developers. This of course comes with the caveat of the heavy hand of a company like Bristle, but There's like [00:31:00] detail to discern there, which is is this a truly community funded and open source kind of push and mental model that we're trying to spread the love of for everybody?

More adaptability with less language drift is the direction, the two balances that you want to hold on both of your shoulders. So yes, I think it is good, but with those caveats in mind,

Yeah. I don't have too much to add. I feel like I'm not enough of a remix expert. 

That's fair.

yeah. I don't know if this is bridging the future. React. Like that kind of just seems like some, trying to placate. Yeah. Make people not upset about something, but yeah. I dunno.

~Who knows? Any other thoughts about this change for Remix? All right. Okay. ~

Oh, I'll add Kent C. Dodds is happy about it. So it must be a good thing,

that must be a

good thing. I do trust Ken. 

Yeah. 

Awesome. All right. ~Let's, we're running up at the end of our recording time. ~Let's hop into our hot takes. Paul, you've been very excited about your hot take.

So do you want to start it off?

I'm excited because I've been passionate about this in the past like six months of my dev [00:32:00] career, which is just like ~I'm so sick of JavaScript or node.~ I'm so sick of node. I'm like tired of it. I'm going to continue to use it for the rest of my life because I can build things faster. From zero to one than any other language, hands down.

I think a lot of people can share this, but I am so sick of it like it is, it's like going on a bad psychedelic trip and you don't know what's coming next, there's interesting things happening all around you and inside your head. ~I. ~I've coded in Go, Rust, PHP, like C, ~we, I've got gone down to the, ~I've had jobs where I've worked as a server monkey.

I've had jobs where I've worked as a full stack engineer as a tech lead. And having gone through these experiences now, I'm not, I don't have decades of experience. I've been in the field for six or seven years. That's it. But after those six or seven years, I look back and I go, Oh, my gosh, like the quality, the durability of things I built that are the highest have never been in node. They just haven't. I can build shiny features and I can impress stakeholders and make the PM [00:33:00] happy, but I'm not making the long term maintenance and survival efficability of the org sound by using Node. And I think a lot of people are starting to talk about this like a little more seriously. And I say seriously because it's like a solemn realization because we love Node in terms of like how fast you can build things.

But it's solemn to realize that. You're huffing on hopium when you're putting out features, you're like, look what we have now. Do you really, does the business really have this? And if the answer is yes, but we need to like then update when a new framework comes out constantly.

Oh, yes. But it has difficulty running and deploying and building. And then I would argue the business doesn't own that product, they own an opportunity. And now they have to hire additional staffing to So yeah, my hot take is other languages have weight. There's a reason they have weight. And personally, I am starting [00:34:00] to really be like that girlfriend meme where you look behind and you're like, Whoa,

nice.

What do you want to see from Node to change your mind or do you just Not want to have to deal with it. I know you say you're going to have to use it for the rest of your life, because obviously it's a mainstay in web development, but if you had your way, what would you want to see from node and, or would you just give it up?

There's already so many transgressions of sanity that have happened that it's hard to ~tell, ~say what I would want. Recently, I tried to make a custom build for Node RED, which is a drag and drop workflow style editor. I wanted to put in a monorepo so that I could then ~like ~import from my ~You know, ~giant plethora of packages.

And then I ran ~up ~upon the realization that ~node is in common, ~node red is in common JS, and ~every, ~everything should be in common js. And of course, I'm in a monorepo, so it's all ES modules and they're one a week of my life.

~40 hours, let, ~a week is 40 hours. And if a dev gets paid, I'm just gonna throw 50 bucks an hour out there.

What's 40 times 50? We're [00:35:00] talking $2,000. This is a $2,000. Mistake. Throw another engineer in a meeting. Now you have a 4, 000 mistake. So the organizational cost is unrealized and I think we're starting to realize it.

Yeah. I think if I can weigh in on your hot take, I think that ~the like com, like the common. The CommonJS, ~the module ecosystem is a pain, like for sure. I think that may be the worst thing here that like, as a specific example that you brought up, it's such a pain. I will say that I have little teeny node projects that are like server only.

That ~have ~I've not updated the node version in a long time. Like they're very like clean data in data out. And I like, I don't know, they work fine. They hum away like it's code running, doing its thing. But yeah, I think node is often reached for as this we can get this done so quick out the door, very little testing, yeah, I don't know. I don't know how much of it's like language versus the just like coral, like the correlation of where node finds itself is like in these kinds of things that tend to be a little more [00:36:00] brittle and super dependency, 

I also feel like as we get more folks into the tech field really, Having solid foundations is more important than ever because you have very flexible languages like Node. Like you go in there, it's like a sandbox, like you can make a sandcastle anywhere. But then you use a language like, I'll throw Go into the mix here, you can't return from a function unless you handle every error.

There is a value and then there's an error and it's a pain in the ass. But oh my gosh, when you're working with a typical setup on a team where you have a principal engineer and then you have maybe a lead or two and then you have a bunch of junior engineers, it provides very strong strata to make sure the software is developed responsibly.

And that's a great guarantee to have built into a language.

Agreed.

~Awesome. ~All right. No, what's your hot take.

Mine's not exactly the same, but I feel like there's been a trend. ~I'm not gonna, ~I'm not gonna drop any names here, but I think this has happened with ~a couple of big, ~a couple of the big hosting providers recently ~where like these, I think we are, we're getting into the era ~where like, the business is now starting to [00:37:00] measurably impacts how devs feel about these platforms~ like ~where one can run.

~Like ~code on the edge or like your web hosting provider. ~And ~it seems like the norm is that these sales teams become ~like ~super aggressive and gnarly and duplicitous. And I just really hope that stops because I feel like we've got enough tricky decisions happening here. ~And, I think ~this is ~like the, ~like back to our previous discussion about ~like ~the conglomeration of everything and the fear there.

I think we may be starting to feel it. ~And ~this has me worried. ~Like ~I am worried about this and I don't know, I just ~feel like it doesn't need to be this way. I ~feel like it doesn't need to be this way. ~Like ~there's enough business here that this should be very doable without these ~like ~crazy valuations and ~like ~all this money flowing around.

~It's man, these, like ~these businesses are reasonable businesses. Everything would be fine if we weren't, ~creating like insane, like~ propping these companies up at just. Asininely high market caps. ~And it's just ~we just gotta slow down. ~Like I, I don't, ~I can't fix this. No one can, but it's just, ~it's ~very tiring.

I

I totally agree with that. ~No, that's a great hot take.~

That is great.

I also like it [00:38:00] because it extends beyond our bubble of software speak and more into the meta of what are these groups that we're building and how are they interacting?

Yeah. Like to what end? Why are we doing this? We're just going to burn it all to the ground. ~Yeah.~

There was something that JLo I remember said to me when I, when we used to work together and he said who thought it was a good idea to teach a rock to think? Not me.

Yeah. Well said.

All right. Any other thoughts before we close out today? ~So both great hot takes.~

This was a fun panel.

Yeah.

~good. It was good. Yeah. ~Thanks for coming back you guys. Thank you, Noel, Paul, as always for coming on, doing panels. Thank you everyone. Who's just been like super patient as we get some of these episodes out. And thank you everyone who listens. And. If you have anything you want us to talk about on our next panel, if you have any questions you want to ask web development experts we should be having a new Mailbag episode coming up in the next month or so [00:39:00] and we will always have our panel episodes every month.

~You can get in contact with me at, I don't even know my Twitter name anymore, hold on. I had to change it.~

~is Twitter? ~

~what is, Twitter? X. ~You can find me on Twitter at Emily underscore Kohanic and that will be in the show notes. Thank you again, Noel and Paul. And we will see you at whatever episode you were on next.

Thanks,

Thanks, Emily.

~All right. Thanks guys. Bye.~