text stringlengths 0 1.36k |
|---|
**Alex Sexton:** As a second follow-up question - is it better? Does it mean we have revolutions more often, that cause better things to occur, and we move faster? Or is it worse, in that everyone's constantly learning new tools, and the quality of output doesn't get any better? Those are my two questions... You have 2... |
**Mikeal Rogers:** \[laughs\] Okay, so a couple notable things. One is that we used to just call this 'framework fatigue' before npm took off, and this problem actually predates even the npm ecosystem. Obviously, you could think of it as being a little bit accelerated into the npm ecosystem, but we've always had this i... |
**Alex Sexton:** There were the framework wars before JavaScript fatigue, and the framework wars were like five frameworks, not 500 frameworks. |
**Mikeal Rogers:** Well, no... It was five frameworks until jQuery won, and then people argued about what framework on top of jQuery to use jQuery that you were gonna build onto... And it's all the same thing, which is that... |
**Alex Sexton:** I still feel like it's exponentially exploded since... I think you nailed it with npm... |
**Mikeal Rogers:** \[28:05\] Yes... So it has exploded, but here's the thing... One of the reasons why you get this more in this space than in the cloud space, or enterprise, is because people build more new web applications than they maintain web applications. People build new stuff all the time, and so when you have ... |
**Alex Sexton:** But isn't Python primarily -- like, what about... People build websites with Python constantly too, but I guess it's because JavaScript is the common language among every Python, PHP, Ruby or whatever website, that it's multiplied times all the other languages? |
**Mikeal Rogers:** I mean, I know people that have Python backends, and they've swapped out -- they built three completely different web apps on top of the same backend... So they actually haven't swapped out the backend, and it would cost them a lot to swap it out, and they'd have to write a lot of new code again... W... |
**Alex Sexton:** Well, you can... |
**Mikeal Rogers:** You can with npm, but think about it from the top-level framework perspective - you can't. So there's an opportunity for people to do this more often, which is why we have more of these frameworks. But I think a lot of the fatigue really comes down to higher order libraries - particularly frameworks ... |
Also, operating under so many layers of semantics, it actually gets kind of hard to just think about applications and build applications. You start having to think in terms of frameworks, rather than in terms of problems that you're solving, and that makes it much harder for you to switch to the new thing when everybod... |
**Alex Sexton:** Yeah. Specifically, I definitely find that interviewing people, even people that I work with from time to time - and this was true in the jQuery days; there were people who knew jQuery and nothing else, but I can talk to them and they know so much about... Now it's pure functions and functional program... |
You get sucked into this specific world where it feels like you can solve everything, because... I don't know. I see people who are brilliant programmers who have such a narrow worldview that it's a little bit sad, but not in like a sad-sad way; they're all smart enough to learn, it's just weird that they haven't notic... |
**Rachel White:** Do you mean a narrow worldview in the sense that they're stuck in the way that they've been programming for so long that they're not open to new ways to do it? |
**Alex Sexton:** It's usually the opposite. It's usually extremely smart, brand new developers, started two years ago when Redux and React were coming up, and they've come up in that world, and that's how they solve anything. I don't wanna be the "get off my lawn" person who's like "You gotta learn the DOM API before y... |
**Rachel White:** \[32:00\] I see what you're saying. Well, it is the world that you have to live in, right? I think it's notable that Substack doesn't have this problem, and I'll start off by saying "We can't all be Substack." This isn't a solution for everyone. But if you are really diligent about not using framework... |
**Alex Sexton:** For one person. |
**Mikeal Rogers:** What? |
**Alex Sexton:** For one person. And I don't even think just Substack. I think any application that can be written by a single person could very easily adopt that strategy -- not easily, but could adapt that strategy with success. I find as soon as you add a team of 20 people, all working on an application, you can't h... |
**Mikeal Rogers:** I think it's a false dichotomy. A lot of people do write applications that way, with larger teams than just one person, and even a lot of the Substack builds he built with other people... But a really large team likes to have some of these larger frameworks because it standardizes the way in which pe... |
A lot of the problems that those frameworks solve, like redefining how the event system works, or how mutations work to the DOM... There hasn't been a ton of small modules that solve those particular problems, so that you can standardize on "Oh, we'll use this module to solve that problem." But as we move along into th... |
**Alex Sexton:** I'm with you, but I think all of the conversations we're having apply to a backend application written in a different language. The idea of small modules versus a coherent framework is not specific to the web platform being messy. To quote Tom or Yehuda on the Ember team, it's like Ember is not just li... |
**Mikeal Rogers:** Well, so was React and so was Angular... I think they're all kind of built that way now. |
**Alex Sexton:** I didn't mean to say that Ember was the only one, I'm saying that I think the false dichotomy is that you don't have this option of small modules that get expected input and output, it's the fact that you have to choose them yourself, and I think that's actually what plays into a lot of the fatigue... ... |
So I think frameworks actually help to prevent the fatigue that a lot of people feel by making a lot of choices, by default. |
**Rachel White:** \[36:12\] So if the problem doesn't lie with the frameworks, and ideally the groups of people that are working on these frameworks are making the best decisions for modules to implement into those libraries, could the JavaScript fatigue then just extend to the module makers that are making so many dif... |
**Mikeal Rogers:** I used to hope for a better outcome for this kind of stuff. When I started to see all these frameworks be built out of more small components, I thought that we would have a lot more longevity and a lot more sanity around them, but what we've actually seen is weird consolidation and plugin patterns, a... |
**Alex Sexton:** Yeah, I agree. |
**Mikeal Rogers:** Babel has a crazy plugin system, and Webpack has a crazy plugin system, and React uses all of these together to make its crazy thing. So even when you move on to the next framework that you built on top of Webpack and Babel, you're still gonna be locked into these weird - for lack of a better term - ... |
**Alex Sexton:** Sure. |
**Mikeal Rogers:** But they're not proprietary. |
**Alex Sexton:** Democratized in the authorship of those... But yeah, so to kind of jump back to the other question, since I think we could talk about small modules versus large modules (and I'm sure we will again), is the pain that JS fatigue causes - does it pay off? Theoretically, if we can go through more iteration... |
So I feel like maybe the web and JavaScript tooling can move forward more quickly because of this, and is it worth the tradeoff in that thing...? |
**Rachel White:** I agree with that, plus if there's so many options and people are trying them out, the ones that don't work aren't going to get widely adopted anyway, hopefully... |
**Mikeal Rogers:** I think that there's two competing ideas about where innovation in this space is driven... So is it driven by new capabilities being opened up in the web platform, so we need new frameworks and tools to take advantage of them, or is actually driven by completely new use cases? I tend to not buy into ... |
When mobile was on the rise, we saw a new slew of frameworks that solved mobile; it's not that the old frameworks were bad at solving older problems, they just weren't particularly situated well for mobile. |
**Alex Sexton:** Right. I think specifically that Carrot in this case is like what are native apps doing... It's almost entirely the generator of new ideas... How do we compete with more or less a better experience in almost every default case from the native apps and the web; by kind of the way it works it's always go... |
**Mikeal Rogers:** \[40:20\] No, that's notable. I don't know what the next thing is going to do that React can't adapt itself enough to handle, and so we'll get a new framework... But I do think it's notable that one of the problems that React solved was creating this componentized model, and in order to do that, they... |
**Alex Sexton:** Styled Components I think does a pretty good job with CSS in-line, for what it's worth. |
**Mikeal Rogers:** Again though, but Styled Components feel like they're going after some newer use cases as well... Or at least they're being adopted there. I feel like a lot of even the underlying platform features that we've gotten that are gonna make this nicer are actually gonna play out once we know what the next... |
**Alex Sexton:** Yeah. I think it's time for a break. |
**Mikeal Rogers:** Yeah, and when we come back, we'll get into the project of the week. |
**Break:** \[41:51\] |
**Mikeal Rogers:** And we're back. Now we're gonna get into the project of the week, Paths.js. This is a library, it's pretty sweet... It's for doing SVG paths and stuff like that. You've spent some time looking into this, Rachel, why don't you tell us a little more about it? |
**Rachel White:** No, I didn't... You did! \[laughter\] |
**Mikeal Rogers:** We're gonna now finger point... |
**Rachel White:** Well, I mean... Yeah, I'm looking at it right now. You submitted it, you talk about it. |
**Mikeal Rogers:** It's an interesting library... It has a low-level API, mid-level API and a high-level API. But even the high-level API, you need to have a bit more of an understanding about curbs, and stuff like that. It is sort of designed for people that maybe know a little bit about how SVG works, like a better l... |
**Alex Sexton:** Yeah. |
**Rachel White:** Yeah! |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.