Really loved Steve witten's recent broad look at web tech. Covers a ton of ground. Almost all the anti-web "bloat" stuff feels unserious as fuck, just vague general whining without the faintest leadership towards actual meaningful better. It's so unclear to me, there's so little evidence to me the legacy really drags us down. Steve nicely reviews real actual points and topics, has actual concerns to bring up. I'm still wholy unconvinced this online endless whinge-fest has anything serious to say, has even a modicum of a points and there's so little effort to point to or steer to better. But God damn does it feel so good to escape totally fucking useless and at least see some good discussion of inside out vs outside in paint models, that at least has specific concerns about core basics of the rendering paradigm. Acko.net is the greatest.
https://acko.net/blog/html-is-dead-long-live-html/The web platform has just done exceedingly fantastically well with incredibly respectable people working really hard to steer things forwards. Having multiple layers of security and technical architecture groups to review, having other browsers review proposals: it's inconceivable to me to imagine spending 1/20th as much effort building an alternative of any even modest caliber. They sure tried with Dart, Flutter, Fuschia, but it's just not a significant enough lift, it's just another app dev platform, and it's desire for pixel perfect graphics made it absolutely ass and terrible at being a good reactive malleable platform. And it being Flash 2.0 made it vastly inferior to what users got from the web.
The web is big and dense. Even ignoring the sprawling amazing magical web platform and focusing on the robust DOM. And that pisses a lot of people off. There is much clutching for certainty, wanting to be steered. Devs crave having the validation of Modernism, of solutions that when written fit expected shape and parameters. 'Another fine Cathedral!'; you know it when you see it. But the web doesn't give us that certainty at all. There's many ways to get there, there's overlapping options and tools, dozens of paradigms to pick from for how you want to organize your css rules, different layout models to swizzle together. The dirty rough open possibility of the web Bazaar drives people frelling off the wall.
I can't help but see those people clutching for certainty as hopelessly out of touch with the universe, obsessed with a narrow engineering paradigm of control & missing the organic beauty of life the universe & everything. Wittens here at least has some core points (amid lots of buckshot, much that various tensions are unresolved).
It's fantastic that we have so many different ways to do web apps. The basics have proven so malleable, such endless ability to be interpreted and governed into forms of whatever we might imagine. The layout of the DOM has gotten so so much better, had so many wonderful additions and fixed in the past decade. The Web Platform Tests have put to the sword so much uncertainty, become a massive way for the web to remain one thing even across many browsers. New browsers like Ladybird and Servo show promise of new engines, new life. Web Components will or won't they questions dog us us still, we all consult our team leaves to guess their future, but meanwhile a similar structuring of custom comments is endemic to React. Invoker/commands are spreading id-ref dogma, for the page linking to other parts of the page declaratively. It's all incredibly exciting. I'd be very happy to have anything else that shows life, that can compete as an equally radiant open question we can continue to improve. But that's not what the DOM haters are after. And the narrow vision view: I think it bereft & sad, and the lack of material they have to point to to steer us, to offer spiritual council upwards rather than just condemning downwards, throwing faint "bloat" condemnations: it makes me sad we critics so obviously up to no good, who just standby and have such vacuous aspersions to cast, on endless repeat.