The concept of "AI will review AI-authored PRs" seems completely wrong to me. Why didn't the AI write the correct code in the first place?
If it takes 17 rounds of review from 5 different models/harnesses – I don't care. Just spit out the right code the first time. Otherwise I'm wasting my time clicking "review this" over and over until the PR is worth actually having a human look at.
Because the code generated is only as good as the initial description of what you want. It's not too different from "standard" coding where you have a first go at solving it and then iterate and polish as you go along.
I've had multiple situations where things "just worked" and at other times you just have to steer it in the right direction a few times, having another agent doing the review works really well (with the right guardrails), it's like having someone with no other intent or bias review your code.
Unless you're talking about "vibe coding" in which case "correct" doesn't really matter as you're not even looking at what the output is, just let it go back/forth until something that works comes out, I haven't had much success or even enjoyed it as much working this way, took me a couple of months to find the sweet spot (my sweetspot, I think it'll be different for everyone).
I've had tinnitus for almost a decade now. It gets better.
The first six months were hell because I kept focusing on how awful it was. Eventually you stop noticing it. It just becomes part of the background – like how the sky is blue, grass is green, etc.
I strongly recommend reading "Living with Tinnitus" by Laura Cole. Tinnitus is a very poorly-understood condition, but hearing about the experiences of others helps a lot. I hope you feel better soon. https://www.goodreads.com/book/show/37707931-living-with-tin...
I've found Bugbot to be shockingly effective at finding bugs in my PRs. Even when it's wrong, it's usually worth adding a comment, since it's the kind of mistake a human reviewer would make.
Right, but you do avoid worries like "will I have to update this dependency every week and deal with breaking changes?" or "will the author be compromised in a supply-chain attack, or do a deliberate protestware attack?" etc. As for performance, a lot of npm packages don't have proper tree-shaking, so you might be taking on extra bloat (or installation cost). Your point is well-taken, though.
> you do avoid worries like "will I have to update this dependency every week and deal with breaking changes?
This is not a worry with NPM. You can just specify a specific version of a dependency in your package.json, and it'll never be updated ever.
I have noticed for years that the JS community is obsessed with updating every package to the latest version no matter what. It's maddening. If it's not broke, don't fix it!
I remain convinced that RSC and the SSR craze was a result of someone (or multiple) people needing a raise and their friends wanting to start a company selling abstract compute. Statically hydrated, minimal React was pretty great when served over good CDN infrastructure. Then I watched the bundle sizes and lock-in balloon. That second article is a dragon slayer. It really lays out the problem with React. In marrying itself to Next.js and embracing the server, it's betrayed the platform. Meanwhile, the platform itself has matured. React practically built my career, and I just don't have a reason to choose it anymore.
I agree, if there is a death of React it will be killed by Next/Vercel.
I probably shouldn’t care. I’m just not looking forward to the chaos of another full “turn” in JavaScript, akin to query->backbone or backbone->react.
Maybe I shouldn’t fear it. I’ve just yet to see an idea that feels valuable enough to move an entire ecosystem. Svelte, HTMX, etc… where is the “disruptive” idea that could compel everyone to leave React?
That’s interesting. I’ve always held SvelteKit in high regard for greenfield projects because it balances capability, developer experience, and performance, but I’ll have to give Marko a look. I’d love to see a similar deep dive into Electron style desktop frameworks since that space still feels underexplored compared to mobile. I honestly wouldn’t know where to start for a video game interface, and that bothers me.
The TL;DW is: yes, class selectors are slightly more performant than attribute selectors, mostly because only the attribute _names_ are indexed, not the values. But 99% of the time, it's not a big enough deal to justify the premature optimization. I'd recommend measuring your selector performance first: https://developer.chrome.com/docs/devtools/performance/selec...
Author here. I cover this in another post [1], but basically the interop benefits you get on the client just aren't there (yet) on the server. My north star:
> Maybe in the future, when you can render 3 different web component frameworks on the server, and they compose together and hydrate nicely, then I’ll consider this solved.
If it takes 17 rounds of review from 5 different models/harnesses – I don't care. Just spit out the right code the first time. Otherwise I'm wasting my time clicking "review this" over and over until the PR is worth actually having a human look at.
reply