Firefox 3.5 Hits Release Candidate Milestone

macupdate writes "Firefox 3.5rc1 has started trickling to users (mirrors and appropriate pages should all be updated soon). You can read the release notes. RC1 still scores a 93/100 on the Acid3 test."

Comments

93/100...

[mejogid]

I still don't understand the obsession with Acid tests - they measure performance in incredibly obscure areas and have a comparatively small bearing on real world performance. Webkit and Opera in particular have designed to the test to an extent, resulting in good scores but not necessarily comparable general compliance. I'm also slightly confused by the use of the word "still" - none of these bugs are severe enough to risk breakage leading up to a release candidate. I believe far more relevant are performance, bug fixes, features and HTML5/CSS3 support (which make far more of a contribution to moving the web on that Acid Test scores do) - areas in which Firefox 3.5 has improved dramatically. Talk about focusing on the negatives...

Re:93/100...

[glop]

Why don't I have mod points today?
The parent is so right. The video tag means that youtube and all the web streaming websites can work without Flash. And since Firefox users update quickly, this means 20% of Internet users will be able to do that within 6 months. That's pretty big when you think some people try to make us believe that HTML5 is 10 years away...

Re:93/100...

[Millennium]

The thing about Acid tests is that specs are ambiguous: there are often multiple possible meanings to a given section, and inevitable different people will implement them in different ways. Some of those will be incompatible, yet both can claim compliance, so this helps no one. Tests, on the other hand, are unambiguous: either you pass or you do not.

This is why Web platforms of the future will not be based on specifications, but on the test suites. Acid tests are not perfect at this, but they are light-years better than previous practice. If Mozilla wants to be seen as taking standards seriously again, they are going to have to start taking these tests seriously, and that means 100% as soon as possible.

They've improved over Acid2, at least, when even iCab -a browser developed by one person- beat them to full compliance by months. But they still have a long way to go. When major tests like this are developed, 100% needs to be a dealbreaker goal for the next major release, not something put off until 2-3 big releases in the future. Opera gets this, and the WebKit folks get this. Mozilla used to, back in the days of the first CSS Acid Test. But somewhere along the way, they lost sight of it, and they need to be reminded.

Re:A little anti clamantic...

[ArsonSmith]

yet don't work 100% in real world webpages. Standards compliant be damned if you can't render real pages.

Re:A little anti clamantic...

[Just Some Guy]

Such as...? I use Safari (at home) and Konqueror (at work) nearly exclusively and haven't had problems with these mythical IE-and-Firefox-only pages.

Re:Still the slowest browser.

[mejogid]

I'm sorry, but without any benchmark or other reasonable test there's no way of ranking the browsers in terms of general speed. I personally find firefox 3.5 faster than IE8 - I don't know if that's because I'm using XP, because that's the result I want to see or any other reason; the point is it's a subjective evaluation. Furthermore, adblock with a reasonable filter list and flasblock improve page load times and responsiveness substantially. The private browsing issue is reasonable, although I've not personally been troubled by it having no real need to combine "private" and less private browsing. I also feel the clear recent history function mitigates that to some extent.

Re:H.264 or Theora?

[armanox]

Since when is H.264 industry standard?