Opera 9.5 To Fully Support CSS?

Albert Sandberg writes "According to a developer blog, it looks like Opera 9.5 (which has been code-named Kestrel) will be the first browser to fully support the CSS selector test (test is here). Finally! Weekly builds should start being available in a few weeks."

Safari Beta 3

From the 43 selectors 25 have passed, 9 are buggy and 9 are unsupported (Passed 346 out of 578 tests)

Not great, but a lot better than I ever did in school. ;-)

Re:Safari Beta 3

For the record...

Iceweasel 2.0.0.4

From the 43 selectors 26 have passed, 10 are buggy and 7 are unsupported (Passed 357 out of 578 tests)

Konqueror 3.5.7

From the 43 selectors 43 have passed, 0 are buggy and 0 are unsupported (Passed 578 out of 578 tests)

So konqueror (which I thought shared source with safari?) is 100% compliant at least as of version 3.5.7 (I don't have an earlier version to test.).

Re:Internet Explorer 7

Me?? I dont use CSS.. I'm a low-tech-guy

why is it so hard?

is it a problem of CSS spec if nobody can support it easily?

Re:why is it so hard?

Part of the issue arises from the fact that much of browser rendering code is ancient. Much of the basic rendering pieces weren't built to handle some of the CSS properties. For instance, many advanced selectors break when you are dynamically adding content or changing/adding stylesheets.

Expect Internet Explorer to lag again unless they completely replace large parts of their HTML rendering engine for standard-compliant sites. There is simply too much legacy code running against the Internet Explorer control, unfortunately.

Opera allows me to do great things

Really, the Opera web browser [opera.com] has allowed me to do great things throughout the internet, with hundreds of tabs open, and consequently more bookmarking being done, and session management, I do not know how productive I would be with Firefox alone. Commonly, when stranded on Firefox-only systems, I am burdened with odd tab loading impairments and generally limited to acting like I am doing literally one thing and one thing only-- no queuing up content or strands of thought, etc. Even with the hierarchical vertical tabbing enhancements through the TBE extension [sakura.ne.jp] akin to iRider [irider.com], my productivity seems to drop. So, I am glad to see more (good) publicity for Opera.

Not Even Close!

This post is fanboyism at its worst. Opera is going to fully support CSS selectors, not CSS. Selectors are just one structure in the CSS language. There are still many other parts of the CSS standard that are not supported by Opera and are not yet planned for any future release.

The Internet-Age-Old PITA...

...Do I use a fully compliant browser in which half the pages out there won't display properly because they've been coded by lazy, clueless hacks with MCSE...or do I use the shit that is Internet Explorer because almost all pages will display semi-properly, even though the code - and IE - is totally fucked up?

I use Opera exclusively, and I know that one day everybody will create compliant webpages. HAHAHAHAHAHAHA. Sigh...

What does it matter?

I use Opera, which is already known to support existing HTML standards pretty completely and accurately.

I still frequently run into web sites built by clueless authors who feel a need to do a browser check, and finding it's not IE or Firefox (or sometimes Netscape!), think it is their duty to inform me that their sites only work with "modern" or "updated" browsers. Feh. By and large, that immediately sends me to the site of a competitor if it's a commercial site I'm visiting.

When will web authors get a clue, and start coding to standards and not implementations. (fuck it if IE breaks because they don't do things correctly)? A properly written web site should never need to do a browser check.

Re:What does it matter?

Yes, because kneejerk reactionaries of the sort trolling through this thread would be off in their little corner of the web and the rest of us sane people would be able to talk civilly. Telling the majority of your users (including, for example, poor people who don't own a computer and use whatever is installed on the computer in the public library) to fuck off is like a shoe store refusing to sell shoes to anyone with uncool socks. Much better to sell someone a nice pair of shoes, and say, "By the way, you could try these really nice socks to go with them. The don't have holes in them like the ones you're wearing."

Still no icon

Despite Opera showing its superiority as a browser over and over again and on multiple platforms, from desktop to mobile to game systems, ther eis still no Slashdot Icon to mark Opera news stories.

Impressed, because ...

I'm really impressed if they dare to follow the standard.

Because with their good example, pages will render differently in opera than the author wanted it too as the pages are probably tuned for IE/Mozilla/Konqueror.

Go Opera!

Very nice news but somehow not surprising by the constant underdog. It truly is a shame that Opera only has 2% of the market considering how great it is in comparison to its competitors regarding speed, features, innovation and security. Imagine a browser so great that people actually paid for it as late as 2005 (these days, Opera is 100% free).

Test results

Firefox 2.0.0.4 for Windows Results:
From the 43 selectors 26 have passed, 10 are buggy and 7 are unsupported (Passed 357 out of 578 tests)

Internet Explorer 7.0.5730.11:
From the 43 selectors 13 have passed, 4 are buggy and 26 are unsupported (Passed 330 out of 578 tests)

Any other tests to report?

Re:Test results

Konqueror 3.5.6 Results:
From the 43 selectors 43 have passed, 0 are buggy and 0 are unsupported (Passed 578 out of 578 tests)

This release of Konqueror has been their stable release since last January was supplanted by 3.5.7 last week. So Opera isn't the first. A stable released browser has been able to pass this test for at least 6 months. I don't know how 3.5.5 and before would have done on it.

All that said, Firefox tends to do better with the javascript heavy sites and has extensions I can't live without. If I were going to use something else it would probably be Konq though. When KDE4 comes out, Konq will be easily installed on Windows and OS X. That might get a bit more momentum behind it.

Konqueror FTW

I'm using Konqueror 3.5.7 on Kubuntu right now, and it passes completely. I don't know how long it's been able to pass, since I just found out about the test now. Firefox 2.0.0.4 fails pretty badly, but this version of Konqueror says that it passes all the tests. Yet Opera claims that it is the first browser to pass? Objection! At least one browser has passed before it, and that Opera version is not even out yet, it's in the weekly builds. This is the stable version of Konqueror

Re:Konqueror FTW

A quick check of the KDE changelogs shows that Konqueror was brought into compliance with the tests with the release of KDE 3.5.6. Linky. [kde.org]

3.5.6 was released in January.

Re:Konqueror FTW

No, you didn't test Konqueror 3.5.7 unitl AFTER Opera apparently passed. So Konqueror passed after Opera. Sure, it may have been ABLE to pass before, but it never took the initiative.

Re:Konqueror FTW

The claim that Opera is first does not occur in the blog entry. It is probably something the submitter added.

Konqueror

I just ran it in konqueror

From the 43 selectors 43 have passed, 0 are buggy and 0 are unsupported (Passed 578 out of 578 tests)

So I guess opera isn't the first.

But...

I used Opera exclusively on Windows, Linux, and BSD for several years, but recently switched to Konqueror. I finally got fed up with a few things in Opera.

My first complaint is their lack of 64-bit support. I'm running the AMD64 version of Debian, and Opera is (was) the only 32-bit program I had to run, making it a pain to keep a bunch of 32-bit compatibility libraries around for one program. I think 64-bit is popular enough now that it'd be worth the time to compile for it. Given the large number of platforms Opera runs on, it should be pretty easy to port.

The second big complaint was that it doesn't support more than 9 mouse buttons. I spent $100 on a fancy mouse, hoping I could control most of my GUI programs with only the mouse. Much to my surprise, any shortcuts after Button9 simply don't work. This was quite disappointing, because Konqueror (and KDE in general) doesn't support mouse shortcuts, so I had to go with xbindkeys. It works well, but seems like a hack.

Also, since at least the 9.0 release tabs have been broken. I had to stop using it after 9.0 because switching between tabs was incredibly laggy. With more than a couple tabs open, there would be a very noticeable pause switching between them. I'd say it was up to 5 seconds or more on some occasions.

I really hope Opera can fix those issues, because I was really sad to switch. IMHO, Opera is the only browser to do tabs correctly, with a full MDI. Yeah, yeah, Firefox has a plug-in, I'm sure. But when it works, it works 10x better in Opera, right out of the box.

Re:But...

The second big complaint was that it doesn't support more than 9 mouse buttons. I spent $100 on a fancy mouse, hoping I could control most of my GUI programs with only the mouse. Much to my surprise, any shortcuts after Button9 simply don't work. This was quite disappointing

  I'm trying to figure out if that's a joke. Nine mouse buttons?

  Any Mac user will tell you that one mouse button, when used in conjunction with seven funny-looking keyboard keys should be enough for anybody!

Different Browser Ratings

Firefox 2.0.0.4 on Windows Vista:

From the 43 selectors 26 have passed, 10 are buggy and 7 are unsupported (Passed 357 out of 578 tests)

Internet Explorer 7.0.6000.16473 on Windows Vista:

From the 43 selectors 13 have passed, 4 are buggy and 26 are unsupported (Passed 289 out of 534 tests)

Lynx 2.8.3dev17 on Windows Vista:

No JavaScript == No tests. :(

Opera 8.5 on Nintendo DS:

From the 43 selectors 14 have passed, 3 are buggy and 26 are unsupported (Passed 313 out of 578 tests)

Opera 9.1 on Nintendo Wii:

From the 43 selectors 30 have passed, 2 are buggy and 11 are unsupported (Passed 450 out of 578 tests)

Opera 9.21 on Windows Vista:

From the 43 selectors 25 have passed, 3 are buggy and 15 are unsupported (Passed 346 out of 578 tests)

Safari 3.0.1 Beta on Windows Vista:

From the 43 selectors 25 have passed, 9 are buggy and 9 are unsupported (Passed 346 out of 578 tests)

Oddly enough, the Wii with an OLDER Opera wins in the Most Completely Working category, while Firefox wins in the Most They At Least Tried category (least unsupported).

Why not Firefox?

Mark me flamebait if ya like. If there's some way to tag my user agent data, you'd see I'm running Linux and Firefox 1.5.0.12. But I have to wonder why Firefox hasn't been all over the idea of 100% compliance.

It's slick, it's fast, it's effective and it's very compatible. I also love the plugins. But it's not much of a 'selling point' that it's not 100% compliant with whatever standards there are out there. It's especially damning when the same demographic often cite that MSIE isn't compliant with standards and often breaks things to the detriment of the use of standards on the internet. Firefox should be the alternative and the recommended benchmark for all web developers out there.

OH MY GOD, COPYRIGHT VIOLATION!

Opera 9.5 To Fully Support CSS?

Someone phone the AACS-LA! Opera is going to distribute pirated player keys!

Fix streaming media instead!

Having CSS working is great and all, but I still dont understand how streaming media (You Tube, etc) still stops once the window where its playing is in the background. Its the main reason why I switched from Opera to Firefox, it seems so basic to fix that Im still wondering if its only my versions of Opera that dont work...

hmm..

Well, here are my results in Firefox 2.0.0.3 From the 43 selectors 42 have passed, 0 are buggy and 1 are unsupported (Passed 577 out of 578 tests) I'm not entirely sure which the unsupported one is.

Reject bad code!

Maybe it's just me, but I personally have a problem with browsers that are proud to "render badly coded websites correctly."

Why? Because it will just encourage bad code. Granted, Opera doesn't have much of a market share and the problem is with MSIE, still, it is a good start. Just reject badly coded websites. Browsers are compilers and interpreters in a way, and I have never seen any compilers that are proud to compile badly written code.

Explain to me this

I never quite understood Opera's raison d'etre. They've always been the underdog, and they've always boasted better standards compliance, but does the common user care about compliance ? I think not. In fact, even if a web site looks like ass in their browser, as long as the clicky things work, they will put up with all sorts of mediocrity.

I switched over to Firefox around the same time that IE7 was released, mostly because I was royally underwhelmed. Then I fell in love with the many development add-ons that helped me debug my web pages and javascript, and now there is no turning back. We all know IE isn't going anywhere, but it would be nice if Opera and Firefox could pool together and release one uber browser to rule them all.

Full support for... a test suite

Passing a single test suite isn't exactly the same thing as supporting the whole standard perfectly. Test suites, by their very nature, only test select subsets of the standard. A single general test suite cannot expose every possible bug in every feature. On top of that, this test suite only covers the selectors, which is a fairly simple and straight-forward part of the spec. Heck, even Internet Explorer supports a bunch of CSS 3 selectors. It's one thing to claim full support for selectors; it's quite different to claim that pseudo-elements with table display values in nested floats with negative margins always work correctly. It's great to see progress, and Firefox and Opera are both impressively close to full support for the current CSS 2.1 specification, but let's not exaggerate the situation. They both still have a lot of work to do (as does Safari, which was clearly behind overall in version 2.x and is likely still a bit so in version 3).

I can't believe...

...that no-one has mentioned some of the other gems from TFA, especially in relation to the *nix builds:

64bit Linux builds
Qt4 builds
Faster tab switching (my only gripe with the current Opera under Linux)

I've been using Opera since 2001, and on Linux since 2004, and it's great to see a vendor maintaining feature parity across different platforms.

The improvements to CSS et al are always welcome, but as some other users have pointed out it's almost always crappily coded sites that give "alternative" browsers a hard time, so it's also good to see they're apparently factoring in better support for error-ridden sites.

Opera CTO Father of CSS

For the record: The CTO of Opera Software, Håkon "howcome" Wium Lie [opera.com], is the father of CSS.

Re:Who in their mind....

Right. Opera has been completely free since 2005.

Re:Who in their mind....

Opera is free, and has been for a while now.

Re:Who in their mind....

Um, yeah... maybe you didn't get the memo, but Opera's been free of charge and advertisements for like 2 years now.

Re:Who in their mind....

Opera doesn't cost money or contain any ads anymore.

Re:Hey this is great news.

It stands for Cascading Style Sheets [w3schools.com].

(That link was the first hit on google for a search on CSS, incidentally...)

Re:Who in their mind....

OK, so it's a troll then. (see the related user names)

I hope some mod will get it and mod accordingly.

It makes sense too -- I couldn't believe someone would have missed these news.

Re:Hey this is great news.

Cascading Style Sheets.

But if you don't know what it means, you're probably not too affected by it anyway. I don't think that CSS3 compliance will matter for "average users" soon, because far from all browsers will support it at the time Opera 9.5 will. It's a step in the right direction for sure, but it'll only be of interest at first to geeks keeping up to date with the latest web browser developments. It may not matter in reality until a few years ahead, or whenever IE 8/9 or whatever gets this far.

Re:Hey this is great news.

It would be even better knews if I had a clue as to what means "CSS".

Seriously? I mean really? For true?

OK, fine. It stands for Cascading Style Sheets [wikipedia.org]. Welcome to the Internets.

Re:Hey this is great news.

Since 1999 or so, the preferred way of putting style on web pages ("how this part of looks") is not mixed into the content structure ("what kind of information this part contains"), but in a separate place, the style sheet.
The style sheet Selectors say what parts of a page must carry it associated style, e.g. 2nd level headers (selector) must be blue and use a 14 point, bold, sans serif font (style).
The CSS stylesheet standard allows lots of complex kinds of selectors, and so browsers used to support only a small subset of selectors.

Thy Parchment bear Good tidings!

Since the 99 of the great 1900's, ways of preferred stoning, and styling leafs of thy webbing, ("imparts of the face of thy brows") Is not structure forged with contentment? ("What manner of entrails subsume thy tiding parts?")

But in thou'ists separate standings, the sheeted of the stylets suffice.

Upon thy Selectors of the Sheet Stylets' dictate: ("What parts of this beast ought carry thy consorts!") E.G. Archfiend the 2nd, Level of the Headers, Lord of the Blue, and Bold user of the Fourteen-Pointed Seraphim") Indeed, it is but I, Sir Salvor of the Cataclysm. Eternal Barron of Travelers and appointed ruler of his Majesty's canonical archetypes.

GO FORTH IN GODS' GLORY VENERABLE SOLDIERS!

Re:Hey this is great news.

For you, "Confederate States Ship".

Re:Hey this is great news.

>
Not really, I just get tired of some of the acromyn laden crap.
Sending morse code, things like this made sense, this is just lazy writing.

73

NB: It was not a troll. IT WAS A SARCASTIC COMMENT.

Re:Er ... I just ran Konqueror 3.5.7

Yeah for some reason it passes the test, but have you tried running the CSS3 demos... hardly any of them work, while with firefox almost all of them do.

so it passes that test just fails at displaying CSS3 correctly.

Mostly the demos seems to be written specifically to firefox and safari. The -moz-* and the -webkit-* bits are sort of giveaways. Or are there some true css3 demos, that uses css3 ?

Re:Er ... I just ran Konqueror 3.5.7

"From the 43 selectors 43 have passed, 0 are buggy and 0 are unsupported (Passed 578 out of 578 tests)" I assume this is what Opera is suppose to get?

No, Opera 1-ups this with: "From the 43 selectors 65 have passed, -15 are buggy and -7 are unsupported (Passed 614 out of 578 tests)"

Re:Progress is made

Time will clear them , opera offers real good details that helps for a better web experience.

Re:RTFA!

Hello there coward anonymous, OP here. Which other browser do you know to support CSS selectors fully? Exactly.