/tech/ - Technology★

>>942937 (OP)

Monopoly and simply because it's popular.

Firefox mobile is simply shit. Unusable, pile of garbage code that wastes every CPU clock it consumes.

Firefox Focus will moves from Webkit to Gecko. What do you think will make it better or worse?

Why would they not? Which part of the Chromium stack would you like them to change?

>>942961

nothing

>>943045

What about servo? Is that viable?

>>943045

Palememe is working on restoring firefox's old embedding functionality. Except this is like 2-24 months down the road though. Would work to replace chorme/webkit botnet though. As they forked before the pozz got into mozilla's source tree.

I've been using Dolphin on Android (Webkit based) and it works great).

>>943045

at least we now have qt5 webkit engine

>>942942

how is it in 2016?

>>943045

>once Mozilla completes their transformation into Chrome

FUD. Firefox is moving to Servo and adding a bunch of new features.

>>942952

Why would they? They can't because iOS doesn't support gecko. They'd have to drop iOS support then.

>>942937 (OP)

Because android is optimized for webview (chrome), so it's twice as fast as gecko. However, gecko is superior in every other way and so is firefox.

>>943211

Also;

>chink botnet

>>943178

A bunch of new features? Or a bunch of new "features"? Hello, ui redesign, pocket, etc.

>>943045

>WHATWG declaring "HTML"5 defunct in favor of Chrome's source tree

Source on that one?

Fuggin Lightning

>>943048

>>943178

make politics, not browser

~mozilla

>>943211

i don't give a shit about free software. i just want functionality

>>942942

>try to open page in android firefox

>it loads forever

>never stops

>I have to close every single tab aswell as the empty placeholder tab from once you have closed every tab then type in the url to get it to load a page

>every time I want to see a page

No wonder nobody uses firefox anymore.

As botnet as chromium, and while chromium works 10% of the time, firefox doesn't work at all.

>>944082

When you have freedom, you have the freedom to make it as functional or functionless as you desire. If you desire to do nothing but whinge to other people that they're not working for you, then you get exactly what you deserve.

>>943178

>>943048

Fingers crossed on Servo not turning into another HURD, but my hopes are not high. And in any case, that would only be one more engine.

>>943817

Not something people are openly declaring, but there's a lot of "well, even if WebKit/Blink were to become a monopoly that pushed nonstandard behaviors, it's free and open source, so it wouldn't be as bad as IE6!" apologia being thrown around whenever the subject comes up, e.g.:

https://developers.slashdot.org/story/12/11/18/1421204/microsoft-complains-that-webkit-breaks-web-standards

https://www.reddit.com/r/web_design/comments/1crhb3/why_doesnt_microsoft_just_throw_in_the_towel_and/

>>945162

>>943048

When the Firefox team want to try some kind of idea for Firefox, they don't try out that idea with an experimental version of Firefox. What they do is they try it out on Servo. Servo was always designed to be an experimental platform, it will never be intended for use in a proper release of Firefox. So when their new idea is proven viable by successfully working inside Servo, that's the time when they start modifying Firefox to bring in the new idea.

>>945164

A lot of Servo's core ideas, like breaking every speck of the browser into tiny safe threads/fibers that can be decoupled and isolated from each other, are crucial to making a browser that doesn't choke and crash constantly on the steady diet of shitcode webdevs produce. There isn't any possible way that could be retrofitted into Gecko without a ground-up rewrite.

>>945229

That's correct. XUL and Gecko were originally using a single threaded programming model and had limited security and no sandboxing features. This means all extensions that relied on the XUL platform had limited security and relied upon the single threaded programming model. Improving Firefox's long term as a software platform inherently meant changing the Gecko programming model so that it would feature a multi-threaded programming model and that each tab would work within its own sandbox. Changing that fundamental model fundamentally means that all extensions that depended on that model break their assumptions about how the platform worked. This also means that it would be impossible to be backwards compatible with the XUL system.