> A browser (or anyone) comes up with a spec, a browser can ship it (to test the waters in an origin-trial, to gain traction if they believe in it), and the standard (often) comes after the fact:
1. Google often doesn't bother even with a spec. Or it creates a semblance of a spec, throws it up on a googler's Github account, ships it and advertises it as "emergin standard" on web.dev
I mean, the status of many (if not most) of the APIs that these sites push are literally "napkin scribble, not on any standards track".
2. Google pushes a lot of APIs quickly into production even if there's a very explicit open objection from other browser vendors (any objections are routinely ignored: from general objections to the shape of APIs to whether it can even be implemented outside Chrome).
3. I wouldn't really quote Alex Russel on anything related to standards, as he is responsible (directly or indirectly) for quite a few of those because of his work on Web Components. E.g. Constructable Stylesheets were shipped in Chrome because Google's own lit project needed them. They shipped it in production when the design contained a trivially triggered race condition, it was called out, and Google completely ignored it because "users want it" or something.
4. Browser vendors quite literally agreed not push incompatible only-exists-in-one-browser shit after the browser wars. The whole standards process is designed to minimize this. Well, Chrome is the dominant browser, so of course they shit all over the process, and quite a few people cheer them for that.
Internet Explorer in the 2000s: shits out a bunch of own non-standard crap, people boo them
Chrome in the 2010s-2020s: shits out a bunch of own non-standard crap, people cheer and blame other browsers for not implementing this crap because... Google is "the champion of open web" or some such bullshit.
If a web app doesn't work on iOS, a business builds a native app instead. iOS is too important. So that fantastic native-like PWA never gets built in the first place.
Apple is not just holding back PWA on iOS, they're holding back the entire web everywhere.
Compare that with desktop, where web apps (maybe not PWAs, strictly speaking) are dominating: Gmail, Office/Docs, GitHub, Figma, you basically do everything in web apps.
And if you count Electron [1]: VSCode, Slack, Spotify, etc, etc.
[1] Importantly, Electron lets you bring your own (browser) engine. You can build a native app on iOS that is just a wrapper around a web app, but it has to run on iOS' WebKit, and is thus limited by what Apple deems worthy
Did you try opening the page? Each feature says which spec it's part of (e.g. "W3C Draft", "W3C Candidate") with a link to it. It also shows which browser implemented a feature first. Often it's Chrome, but sometimes it's Firefox, Opera, or even desktop Safari!
Likewise, you can click on the Chrome icon to change comparison browser. Here's a list of features implemented in Firefox on Android but not in Safari on iOS (and therefore, not in Firefox on iOS either):
Fwiw, I've been a Firefox user for about 9 years. I would love to see Firefox be able to ship their engine on iOS. The main reason Firefox haven't implemented as many features as Chrome is that they lack the resources to. Anti-competitive behaviour has hurt them a lot, and being forced to use a sub-par, undifferentiated browser engine on iOS - the world's most valuable and influential OS, has played a big part in this.
Hi! Creator here (of iOS404) - you can filter level of standard and compare to FF Android (or compare to Safari Desktop, or any mix) instead if you'd like.
First of all, features can be a standard without (full) FF and/or Safari support.
Second, Safari has a monopoly on iOS and controls what other browsers can support on the platform (that also usually means "less than Safari", because SF gets to support things first). They are in a unique position to hold back the entire web, even on other platforms. They're holding the standards hostage by not allowing the market to decide which features are important to them (and put pressure on Safari and FF to implement them)
No, because any browser can decide to ship a feature that it thinks is worthwhile. Users can decide which browser they trust to be their User Agent. The distribution model is open. You type a URL, you click a link. No single company in control.
> No, because any browser can decide to ship a feature that it thinks is worthwhile.
Yes, yes they can. They don't get to call it standard or essential. And Chrome-shilling sites like the pwa.gripe and a slew of others don't get to call those features "essential standards of the web".
> No single company in control.
That is literally not how standards work in the browser world by literal agreement of all browser vendors.
We literally lived through this with IE pushing its own non-standard features and calling it a day. Hence the whole "let's reach a consensus, and have several independent implementations of a feature before calling it a standard".
And if "no single company is in control", why then you're so enthusiastically pushing for a Google's full control of the web?
yeah I'm using mobile Firefox and it has an awfully high overlap with Safari. Almost like a bunch of the stuff Chrome supports isn't actually a standard at all yet...
1. Google often doesn't bother even with a spec. Or it creates a semblance of a spec, throws it up on a googler's Github account, ships it and advertises it as "emergin standard" on web.dev
I mean, the status of many (if not most) of the APIs that these sites push are literally "napkin scribble, not on any standards track".
2. Google pushes a lot of APIs quickly into production even if there's a very explicit open objection from other browser vendors (any objections are routinely ignored: from general objections to the shape of APIs to whether it can even be implemented outside Chrome).
3. I wouldn't really quote Alex Russel on anything related to standards, as he is responsible (directly or indirectly) for quite a few of those because of his work on Web Components. E.g. Constructable Stylesheets were shipped in Chrome because Google's own lit project needed them. They shipped it in production when the design contained a trivially triggered race condition, it was called out, and Google completely ignored it because "users want it" or something.
4. Browser vendors quite literally agreed not push incompatible only-exists-in-one-browser shit after the browser wars. The whole standards process is designed to minimize this. Well, Chrome is the dominant browser, so of course they shit all over the process, and quite a few people cheer them for that.
Internet Explorer in the 2000s: shits out a bunch of own non-standard crap, people boo them
Chrome in the 2010s-2020s: shits out a bunch of own non-standard crap, people cheer and blame other browsers for not implementing this crap because... Google is "the champion of open web" or some such bullshit.
reply