iiinova — where your digital self keeps evolving

We can quibble over whether or not Solid2 does that. I think as said above it solves categories of problems that plague React, for example. Ultimately I want the web to succeed and am enjoying Datastar. I think the communities can learn from each other.

000

I think, as we optimize against marginal devices, the general case should cater to marginal developers (or at least the mean fullstack developer trying to ship at Acme Corp.). Not to say that there can't be advanced solutions for advanced developers. But IMO we should try to raise the floor.

The complexity feels like a natural consequence of the scale of problems we're trying to solve. But if we want the web to succeed, we should continuously ask ourselves if our solutions are actually helping developers succeed.

This comes back to the reason I have, in the past year, started gravitating away from modern TS frameworks. It is hard to see how complex it is when you're deep in it every day. When you have to take a step back and work with developers trying to build stuff on undisciplined teams, it falls down.

My goal right now is to help more intermediate developers (across the stack) build better software on the web. Much is education: folks not understanding the tradeoffs and pitfalls, using abstractions they read about on LinkedIn for the wrong problem, not understanding the web as a platform, etc.

120

I think this has less to do with the architecture and more to do with the documentation. I think the language is (or was, might be better now) obtuse for anyone not already versed in the challenges. And maybe that's correct given the target audience.

100

Dev Wells @wlls.dev · 18h

IMO it is in the best of current TS frameworks. I raved on release about the consolidation of the metaframework into the framework. My main complaint is abstraction complexity and onboarding. Solid2 feels like a framework targeted at people who are already experts in modern JS frameworks.

220

As @drk.wtf said, after over a decade of having my teams fumble with SPA architecture, D* has been refreshingly simple (in the Rich Hickey sense of the word) while still giving me the tools to build high-fidelity apps. I appreciate Solid2 (and use it on an app in prod!).

Dev Wells @wlls.dev · 18h

This is a false dichotomy that I don't think you intended to sound as sensational given your other responses below, re: nuance. I don't really want to keep going back and forth here though. I feel like the original intent of this thread has been misinterpreted (or miscommunicated or both).

101

why do dentist offices all seem to have crazy tech? dentist office: 1s i'ma to stick this wand in your mouth that lets me build a 3D rendering of your entire mouth in real time PCP office: please step on this scale from the 90s and let me tap on your back three times

000

At the end of the day, test and measure. I'll leave it again at that, after building and leading teams building apps for years, I've been pleasantly surprised at both the developer ergonomics and resulting software I'm building with Datastar.

000

I'd also remind that most hypermedia apps are progressively enhanced and *do fall back* to effectively being MPAs while SPAs do not without explicitly handling for it.

101

I agree we're speaking in broad strokes on nuance, and am not sure I'm making my claims any clearer without a clearer example. I'll put that together when I have some time and share out.

100

"degrade rapidly without an internet connection" is funny to me as that's how I'd describe my experience with almost all SPAs I use today in that they are completely unusable. The error message is a result of this not being a real app.

I'm not sure I'm following--there's entire books on data-driven applications? I'm also not sure that's the takeaway that case study is trying to portray given. The point is you often need data to render (whether on server or client). You can be smart about how you do that, agree.

000

SPA vs. Hypermedia: Real-World Performance Under Load | zwei und eins gmbh

For some reason every time I discuss this people assume that hypermedia means full page reloads or replacing the entire page. You can do that, but that is not a fundamental assumption. A relevant interesting read: zweiundeins.gmbh/en/blog/spa-...

zweiundeins.gmbh

We built the same AI chat application as an SPA (Next.js) and as a hypermedia app (PHP/Swoole/Datastar), then benchmarked both under real-world conditions—Slow 4G, mobile device, 4× CPU throttling. Th...

We can discuss ergonomics plenty. What I'm saying is that this *has been measured* and surprised me (as someone who has built SPA for 10+ years). The rhetoric has been that for performance critical apps, data-driven apps, SPA is the way to go. I am challenging that.

100

As @drk.wtf said I'm not talking about in memory state vs over the network. My claim was that the fastest way to get an update to a user (fresh data over network) is HTML + domparser. So unless you're okay with working with a lot of stale data (many apps are), then this is not a red herring.

100

My original point though was that, unless I'm misunderstanding in which case I'd love to learn more, it's not as simple as saying that hypermedia isn't suited for apps (at scale, in case that wasn't clear). I recognize I risk being pedantic here. Thanks again!

100

Design affordances can be handled in both hypermedia (or at least Datastar) stacks and SPAs. The ergonomics of doing so might be easier depending on the app in one architecture or another, though each has their weaknesses too.

100

I think in part where we're speaking past each other is in the difference between, as you said on stream, design affordances (what do I show) and dynamic data latency (how fresh is what I am showing). The water is further muddied by this ethereal q of "who owns the state", which we can debate.

100

But in my experience there are many apps where that is simply not the case. A realtime dashboard for example, the app I build for a living, benefits from the server-authoritative model, regardless of network latency.

100

This might be besides the reality of the situation: building web apps is fundamentally hard, and your choice of architecture may allow you to move the amount of fresh data you can easily manage up and down the scale. For some apps reasoning about local data models might be preferable, I agree.

100

From a DX standpoint, I sympathize. From a more theoretical one, I'm confused, given that my understanding is the fastest way to get fresh data shown in a browser is zstd/brotli compressed HTML morphed into the DOM. To wit I'd expect hypermedia to perform materially better in throttled conditions?

200

To make sure I'm following: you're saying modern SPA frameworks account for slow networks in their design. I assume you're referring to optimistic intent and/or the fact that SPAs tend to localize more state than typical hypermedia stacks.