Hacker News Reader

Comment by DrJokepu | Hacker News Reader

Firefox is not what it used to be. Remember the days when it was called Firebird and Phoenix? When it was actually a fast browser? Decisions at Mozilla now made with politics and not the user in mind. I'm hugely disappointed by Firefox's poor codec support for the HTML5 audio and video tags. This will make either the audio/video tags or Firefox irrelevant but there's one thing it won't achieve; making Ogg Vorbis and Theora popular. Seriously, what they're doing is pretty much an abuse of their market position. Firefox might be popular and mainstream now but if things don't change it will lose its relevancy really quickly.

I think at this stage the best solution would be forking Firefox. Remember when they forked out Phoenix from Mozilla to create a lightweight and fast browser that offers features that users actually want? I think it's time to do it again.

View on HN

I've used Firefox since the very early days (Phoenix 0.3) and I can honestly say it could never be described as a "fast browser" - before Firefox I was using Opera which was much faster and much more lightweight. If anything, I would say Firefox is faster now than it used to be, but perhaps now the difference is it's easier to make an unfavourable comparison to the WebKit browsers which didn't exist back then.

The audio/video tag situation is messy, but I hardly think solely blaming Mozilla for it is fair. Apple won't support Ogg because of submarine patent concerns (although I doubt them already having a license and having millions of devices with H.264 hardware support hurt) and Mozilla won't support H.264 due to known patent concerns. It's hardly the overdramatic "abuse of market position" that you make it out to be, especially given that their intentions are also somewhat altruistic.

I also don't see what a Firefox fork would achieve. I doubt anyone is going to magically be able to make Firefox faster if Mozilla's full-time staff is unable to, and I doubt ignoring patents or doing messy hacks to integrate OS codec support is going to win widespread favour given the current position of the video/audio tag.

> I doubt anyone is going to magically be able to make Firefox faster if Mozilla's full-time staff is unable to

Maybe, maybe not. It depends where their priorities are. Firefox still has memory leaks that need to be plugged, etc. If Mozilla staff is divided between fixing memory leaks and adding features, then a group solely focused on memory leaks might have a better chance.

Isn't part of the issue at hand that the extensibility of Firefox is deep, while that of Chrome is shallow(er)? I understand that deep extensibility isn't necessarily a cause for memory leaks, the ongoing inheritance of a community making use of that deep extensibility can limit

Also, I personally haven't been exposed to any evidence regarding the existence of memory leaks in a Firefox install with no extensions. Many people have anecdotal evidence of their issues with memory leaks, but I never see a corresponding claim for having no extensions installed.

Of course that point is also anecdotal.

IIRC, Firefox extensions are Javascript+XML... so shouldn't there be a garbage collector for 'memory leaks?' Or are we also including 'run-away memory usage' as 'leaks?'

In garbage collected environments, 'memory leaks' usually mean keeping references to objects even when they're not needed anymore.

Any IT enterprise, whether a part of a larger entity or itself the entity, must strike a balance between what its clients demand and what it must do. Mozilla feels that it needs to push Ogg Vorbis and Theora into the mainstream, despite the fact that its clients could not give a rat's ass about that. They failed to write it directly into the standard so now they are using their market position for the questionable tactic. However, what they are doing is still better than what Microsoft had done when they were in Mozilla's position.

Also, isn't FF3.5 faster now than it ever was, even as Firebird or Phoenix?

As far as forking FF: good luck. The code is a bit of a mess and there's a huge amount of it. What organization will want to take this on? I think you would have better chances at cloning WebKit+V8.

Reply on news.ycombinator.com