A year-end wrap-up from LWN
A year-end wrap-up from LWN
Posted Dec 19, 2019 7:14 UTC (Thu) by joncb (guest, #128491)Parent article: A year-end wrap-up from LWN
> Chrome to make specific sites work. To those of us who lived through the period of Internet Explorer dominance, much of this looks discouragingly familiar.
I feel this is a little more doom and gloom than reality suggests (to me at any rate). While it is true that Chrome is heavily dominating the browser space, this feels to me nowhere familiar to the old IE dominance days.
1/ When you say "Chrome is heavily dominating the browser space", I think it's fairer to say "Blink is heavily dominating the browser space" since the surrounding UI is mostly irrelevant when compared to the actual HTML Renderer. From that perspective you're basically saying "The open source project behind Chrome, Opera, Vivaldi and (probably) Edge is dominating the browser space". This strikes me as gloomily as the idea that OpenSSL is dominating the cryptography space.
2/ Blink seems (so far) to be pretty compatible with it's parent project. Wikipedia at any rate says that Chromium on MacOS uses WebKit instead of Blink which suggests that it's not a huge operation to switch them out. This implies that you're really saying "The pair of open source projects behind basically every browser other than Firefox is dominating the browser space"...
3/ Even ignoring that, Chrome is pretty heavily available compared to IE which was available on Windows and MacOS.
4/ I use Firefox as my personal browser (Vivaldi at work) and so far i haven't seen any page that didn't work due to being Firefox. I get(vastly) more problems from my javascript blocker (uMatrix) than I do from my browser choice. Do you have any references for "forcing even dedicated users of other browsers to launch Chrome to make specific sites work"? (Not saying you're wrong or lying, it just completely clashes with my experience) I could definitely see needing to use Chrome for things using experimental tech (like PNaCl) just like you needed to use firefox for their own experimental tech (like early ASM.js) but there's a difference between experimentation and "embrace and extend".
Having said all that, it's definitely something to keep an eagle eye on. While i think people sometimes jump on the google bashing bandwagon a little too readily, I certainly can believe that they would push things in a direction that wouldn't have the best interests of users in mind. But suggesting that this is "discouragingly familiar" to the bad old IE days seems nonsensical to me.
Posted Dec 19, 2019 9:27 UTC (Thu)
by rhertzog (subscriber, #4671)
[Link] (1 responses)
Posted Dec 21, 2019 11:53 UTC (Sat)
by joncb (guest, #128491)
[Link]
Posted Dec 19, 2019 12:57 UTC (Thu)
by martin.langhoff (guest, #61417)
[Link] (14 responses)
This is very different from the bad old days when a proprietary engine with insecure "extensions" to the web platform (ActiveX) was dominating.
Naturally, multiple compatible implementations are better, and I want Firefox to remain viable and available. But if either FF or Chromium is to dominate... I would not be worried, as long as licensing remains open and governance is reasonably good. That multiple players are using Chromium and collaborating on its codebase tells me that governance is fine.
Posted Dec 19, 2019 13:25 UTC (Thu)
by pizza (subscriber, #46)
[Link] (4 responses)
Alas, Chrome plugins are rapidly becoming the new ActiveX.
Posted Dec 20, 2019 0:42 UTC (Fri)
by Kamilion (guest, #42576)
[Link] (3 responses)
I only have the pepper flash plugin (which is supposed to go away soon?) and the pepper sandboxed pdf renderer...
Are you sure you're not mixing plugins up with extensions?
Posted Dec 20, 2019 1:34 UTC (Fri)
by pizza (subscriber, #46)
[Link] (2 responses)
I'm not referring to the likes of uMatrix, EFF Privacy Badger, and other reputable F/OSS extensions that users are free to install (or not); Instead I'm referring to corporate-mandated antivirus/data exfiltration extensions that hoover up everything you do, web sites that spam you with "install our chrome app" messages (to ensure you don't block their ads and data collection), and other such practices.
Sure, they're not binary x86 code with (effectively) full system access, but they are still locking you into a specific (browser) platform.
Posted Dec 20, 2019 9:08 UTC (Fri)
by farnz (subscriber, #17727)
[Link]
The interesting development here is WebExtension APIs that allow you to port from Chrome to a cross-browser standard. Assuming they're good enough for the corporate-mandated extensions, that permits a migration to Firefox in future.
The critical question is whether those APIs are good enough to polyfill from Chrome to Firefox in one step, or whether you have to do significant engineering effort to port. If Google started a migration from their Chrome extension APIs to WebExtension, this would definitely break the monopoly - extensions would be cross-browser by default.
Posted Dec 21, 2019 11:50 UTC (Sat)
by joncb (guest, #128491)
[Link]
I feel there's a meaningful distinction to be made between "restricted corporate world" and "general internet". So long as the corporate browser monoculture (and there is solid reasons for the corporate world to push in the direction of a monoculture) remains compatible with the general internet, i'm willing to dismiss what they do as mostly irrelevant.
Posted Dec 19, 2019 14:00 UTC (Thu)
by rahulsundaram (subscriber, #21946)
[Link]
That speaks more to the overwhelming and unhealthy dominance of a single web platform and others sorta giving up more than good governance. Firefox's lack of interest in being easily embedded didn't help either. Yes, the situation a bit better than IE but the parallels are quite clear
Posted Dec 19, 2019 15:48 UTC (Thu)
by mathstuf (subscriber, #69389)
[Link]
Are there any collaborators that couldn't be grouped into a multi-national corporation association?
Posted Dec 21, 2019 2:28 UTC (Sat)
by roc (subscriber, #30627)
[Link] (6 responses)
It tells you no such thing. What it tells you is that wannabe browser vendors prefer to avoid the massive cost of developing their own engine, and they like how Chrome market share means Chromium is the most Web-compatible engine.
Chromium's governance model is simple: Google's in charge. The kerfuffle over extension manifest v3 illustrated this perfectly: Google engineers made a decision, it was very unpopular with certain Chrome extension authors and users, and murmurings of discontent from non-Google Chromium users, but there was never any process for challenging that decision.
Posted Dec 21, 2019 7:45 UTC (Sat)
by flussence (guest, #85566)
[Link] (5 responses)
A bit like how Firefox Focus uses Chromium because no other browser has a competent embedding API.
Posted Dec 21, 2019 8:11 UTC (Sat)
by roc (subscriber, #30627)
[Link] (4 responses)
iOS Firefox Focus used Webkit and still does because Apple gives you no choice on iOS.
Posted Dec 21, 2019 21:29 UTC (Sat)
by flussence (guest, #85566)
[Link] (3 responses)
Then what's WebView? Webkit? Opera??
Posted Dec 22, 2019 1:33 UTC (Sun)
by rahulsundaram (subscriber, #21946)
[Link] (1 responses)
Webview is WebKit. Not Chromium
Posted Dec 27, 2019 20:57 UTC (Fri)
by flussence (guest, #85566)
[Link]
Posted Dec 22, 2019 1:34 UTC (Sun)
by roc (subscriber, #30627)
[Link]
Posted Dec 20, 2019 5:14 UTC (Fri)
by ThinkRob (guest, #64513)
[Link] (1 responses)
I don't know... it's familiar in the sense of it being a (near) monoculture controlled by a company whose goals don't really align with a Free and open Web.
But IE had an even higher market share in the darkest days of IE dominance than Chrome does now, and Firefox was still a viable choice back then. Every once in a while you'd hit some site coded by total boneheads that depended on an IE bug, but it was usually solvable... often with nothing more than a bit of user-agent fibbing.
I guess I'm optimistic that, if/when necessary, the open-source nature of Blink will at least help the FF devs figure out how to implement bug-for-bug compatibility modes.
Or maybe the modern web is too complex for that and we're all doomed to a pro-tracking browser. But I hope not!
Posted Dec 21, 2019 12:22 UTC (Sat)
by joncb (guest, #128491)
[Link]
Is it possible that Google could go rogue with Chrome? Absolutely. If someone opined that it's only a matter of time before they do so, i'd probably agree. No the real thing that makes me optimistic is that if they go rogue in Blink, they have to do it in the open and I have faith that someone will see it. Particularly when you take into account things like Brave who are (theoretically) explicitly trying to place themselves in the space where Google is most likely to go rogue.
I think the only thing i really disagree about is the distrust of monoculture. I don't trust a closed monoculture and if blink ever goes closed source, that's an immediate warning sign to me. I don't think open source is perfect, hearbleed should disabuse anyone of that notion, but it's still vastly better than other options and I think having an open source HTML Renderer as a monoculture is actually better for us as users as it makes it impractical to create a closed source one.
Posted Dec 21, 2019 2:22 UTC (Sat)
by roc (subscriber, #30627)
[Link] (1 responses)
Then Wikipedia is wrong. For sure Chromium on MacOS uses Blink.
Chrome on iOS uses Webkit, because Apple doesn't allow apps to use any engine other than Webkit. But Chrome on iOS is not Chromium.
Posted Dec 21, 2019 11:12 UTC (Sat)
by joncb (guest, #128491)
[Link]
Sorry, mea culpa. You are correct, I mixed up iOS and MacOS.
A year-end wrap-up from LWN
A year-end wrap-up from LWN
I can sympathize with voice/video chat being hard to do cross-browser compatible.
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
Almost all of my extensions are available for firefox too... uMatrix, uBlock Origin, EFF Privacy Badger, singlefile, tampermonkey, visbug/firebug, web of trust, proxyswitchy omega... the only extension that doesn't work is the NaCL OpenSSH extension that used to be a chromeos app and was migrated to an extension, and the mosh extension.
But firefox has a jetpack for mosh, and I think it works for ssh too..?
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
https://support.mozilla.org/mk/kb/geckoview-firefox-focus...
It never used Chromium.
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
While it is true that Chrome is heavily dominating the browser space, this feels to me nowhere familiar to the old IE dominance days.
A year-end wrap-up from LWN
A year-end wrap-up from LWN
A year-end wrap-up from LWN
Having said that, it's explicit about "Chromium on iOS" not "Chrome on iOS" however that's still working with third hand infomation, i have no way to check and the build page for iOS isn't explicit about what engine is being built so i'll bow to your superior info on this.