Talk:Acid2

From Wikipedia, the free encyclopedia

Acid2 is a former featured article candidate. Please view the links under Article milestones below to see why the nomination failed. For older candidates, please check the archive.
May 17, 2008 Featured article candidate Not promoted
Internet Portal
This article is within the scope of WikiProject Internet, an attempt to better organise information in articles related to the Internet. For more information, visit the project page.
A This article has been rated as A-class on the class scale.
??? This article has not yet received an importance rating on the importance scale.

Contents

[edit] Archive

Talk:Acid2/Archive 1
Discussions from 2006
Discussions about Opera 9.25, FFox 2 and 3
Discussions about the brokenness of the test (two days that caused a lot of confusion!)
Etc

--itpastorn (talk) 10:01, 9 March 2008 (UTC)

[edit] Amaya Fails the Test

It's ironic that the W3C developed browser failed the test, worth to add it? —The preceding unsigned comment was added by Juanchito2006 (talk • contribs) 21:57, 3 January 2007 (UTC).

Sure, go ahead. Keep in mind though that Amaya is primarily meant as a testbed for new technologies, not actually as reference software (see the Amaya article), so passing the test is probably not of high (if any) importance to them. --ADeveria 18:18, 4 January 2007 (UTC)

[edit] 'Timeline of success table'?

What do the colours mean in the timeline of success table? I think they should be keyed.

[edit] Guidelines for inclusion of timeline table

To keep the table helpful, I believe it should only contain releases relevant to the acid2 test. In my opinion these include:

  • The first announcement that the browser (or rendering engine) is able to pass the test
  • The first available public build (so anyone can verify that it works)
  • The first available "final" build (noting that it passes in a stable build)

It should not include:

  • Any release between the first public build and the final build
  • Any future builds after the final, assuming they pass
  • Offshoot browsers or other OS versions based on the exact same rendering engine

Browser milestones have their place on their own articles, not this one.

Based on these guidelines, I have removed Safari 3 and the Mozilla offshoots, and probably others should be removed as well. --ADeveria 22:41, 20 June 2007 (UTC)

[edit] Trivia

I'm not current at all on editing guidelines or anything, but given that the Trivia seems to be discouraged, it seems to me that the second bullet point, namely: "Because Acid2 also tests how web browsers deal with faulty code, it will fail W3C CSS validation. This is expected and was the intention of its designers" could be fit in the introduction with some slight rewording, perhaps to the effect of, "It is interesting to note that the Acid2 test page will fail W3C CSS validation. This is due to the portion of Acid2 that tests how web browsers deal with faulty code, is expected, and was the intention of its designers." Isarl 18:24, 21 September 2007 (UTC)

A change essentially the same as that suggested by Isarl has been made. The Trivia section has now (November 2007) been flagged for 5 months. It seems to me that all the parts worth keeping have been put elsewhere. Any objection to deleting the remaining part of this section? JamesBWatson 9 November 2007

I took out the trivia section, there's nothing relevent there anymore. If anyone wants to keep the Opera trivia, I think it'd be more at home under an "Easter Eggs" heading in the Opera article Psym 04:42, 15 November 2007 (UTC)

[edit] IE8 passes ACID2

http://blogs.msdn.com/ie/archive/2007/12/19/internet-explorer-8-and-acid2-a-milestone.aspx —Preceding unsigned comment added by Gniw (talk • contribs) 21:12, 19 December 2007 (UTC)

I strongly disagree with the inclusion of IE8 at this early stage in its development. There have been no independent tests or even a download so that those outside the development team (and possibly expensive msdn agreements) can confirm it passing. Furthermore, it is a beta version of IE8 that is said will pass, suggesting that IE8 will always pass during its development and on release is unrealistic when you look at the history of other browsers attempting to maintain compliance. The quick modification of this article by fan boys really is jumping the gun and I strongly suggest it is cleaned or more accurately described to avoid future embarrassment. —Preceding unsigned comment added by 210.18.218.5 (talk) 01:39, 20 December 2007 (UTC)

[edit] IE just Passed

http://channel9.msdn.com/ShowPost.aspx?PostID=367214#367214 -n3k —Preceding unsigned comment added by 209.139.213.73 (talk) 01:04, 20 December 2007 (UTC)

[edit] broken in all browsers

Actually Hv3 still passes the test as served - http://tkhtml.tcl.tk/hv3.html 150.101.115.231 (talk) 02:02, 20 December 2007 (UTC)

odd its broken in opera 9.5 which did pas the test yesterday —Preceding unsigned comment added by 24.151.199.107 (talk) 02:17, 20 December 2007 (UTC)

[edit] Netscape (classic) test picture

Hi, I made a picture of Netscape Communicator 4.8 taking the test and was wondering if I am allowed to submit it. (In case you were wondering, it fails the test miserably) :) Bubble 94 is bringin' the JE££¥! (talk) 19:23, 22 December 2007 (UTC)

Add it to Wikimedia commons, choose a license and it's ok.--itpastorn (talk) 19:27, 3 February 2008 (UTC)

[edit] Timeline inclusion

IE8 seems worth including since its authors have made the announcement that their browser passes the test. According to the guidelines I wrote, having yet another entry for Firefox seems rather unnecessary. If we were to include FF3b1 just because it's "more stable", we should add Opera 9 betas and probably other builds of other browsers as well. In my opinion, however, we should stick to more specific milestones. --ADeveria (talk) 20:41, 26 December 2007 (UTC)

IE8 did not pass the test. Its like rewriting the questions of an exam to make them ones you actually know the answers to.--Oni Ookami AlfadorTalk|@ 20:24, 27 December 2007 (UTC)

That's a little harsh, I think a better analogy would be like wearing certain glasses first in order to pass the test. Anyway, the current entry is merely there to point out their claim (hence the red background). Once they actually give us a public build to look at, we can show more scrutiny. --ADeveria (talk) 13:11, 28 December 2007 (UTC)
Wearing glasses would be more akin to changing the settings of the browser to make it work. In order to pass, they had to phsyisically add to the test. Your point about a public build is exactly why it shouldn't be included. I guess since I didn't get the score I wanted on the SAT I cross out one of the questions to make it go up, eh?--Oni Ookami AlfadorTalk|@ 14:35, 28 December 2007 (UTC)
How do you know how they implement the standard mode switch ? Isn't the meta tag a speculation ? How do you know they didn't implemented a client side switch ? The original blog post states it's an internal build. If it was me, I'd even simply hardcode it, so build is always in standards mode. The problem is we can't verify it. If we keep the table as is, then IE is worth mentioning (like we mention iCab and Konqueror with their scrollbar problems). If we want to tidy the table (see the very contructive complaint below), then leave only the official release rows, or at least remove all non-public builds from the table.
I have more questions :
  • If OmniWeb and InternetChannel are compliant, why can't we find them in the table ?
  • Prism is using Gecko 1.9 which is in development, and is called a prototype. Shouldn't we change the cell background ?
  • iCab is four times in the table. Isn't it a bit too much, as the first three don't seem to be truly compliant ?--Fenring (talk) 17:35, 28 December 2007 (UTC)
  • The Internet Channel is a derivative of Opera, and OmniWeb is a derivative of Safari.
Just like Shiira is a derivative of Safari. They deserve a row in the table. Do you know the release date of the compliant OmniWeb ?--Fenring (talk) 20:28, 28 December 2007 (UTC)
  • I've fixed the background color for Prism.
  • If I understand correctly, it's debatable whether or not iCab passed the first three times.
  • The material available on IE8 clearly indicates that IE8 will not truly pass. It's not like iCab where it was debatable.
—Remember the dot (talk) 19:04, 28 December 2007 (UTC)
Thank you. Aside from RTD's explanation of it, also keep in mind that when making a claim such as it passing, the burden of verifiability is on the assertion that it passes, not the omission of it.--Oni Ookami AlfadorTalk|@ 19:08, 28 December 2007 (UTC)
I agree that it's lacking verifiability. But we have to choose. Either mention the informative non-public (not verifiable) and nearly-compliant (lack of overflow support,...) builds. Or remove them all.--Fenring (talk) 20:28, 28 December 2007 (UTC)
You seem to be forgetting that the core problem here is that the reason the browser is not passing the test is because it has to change the test itself. Its not a matter of the browser's support, recognition, public status, etc. Its a matter of the fact that however slight the change, it is taking a different test than the rest.--Oni Ookami AlfadorTalk|@ 20:52, 28 December 2007 (UTC)
Where do you see that ? The sources says that Microsoft plans to introduce the opt-in scheme based on the <meta> tag, and will not be in standards mode by default, for compatibility of course. They don't claim the internal build is not in standards mode by default. Am I missing the source that says MS had to change the test to pass it ? Håkon and Howtocreate use conditionnal tense. It's only guesses. Though they all say the internal build of IE does pass the test. Face it, no official release of IE will ever pass Acid2 test by default. That would break compatibility. However there will be a rendering mode that will support standards. That fact is important in the timeline. More important than the five (i miscounted !) builds of iCab, I think. The only question that remains is wether or not we should keep the non-public rows, and wait for a public verifiable build of IE8. --Fenring (talk) 21:49, 28 December 2007 (UTC)
I would be fine with including it provided that the table clearly states that IE8 does not pass with the default settings. —Remember the dot (talk) 01:32, 29 December 2007 (UTC)

"The material available on IE8 clearly indicates that IE8 does not truly pass." -- where can I find this material? CobraA1 (talk) 05:36, 30 December 2007 (UTC)

The most direct source is this video. About 19 minutes and 15 seconds through the video, Alex Mogilevsky, a member of the IE team, points at a picture of the Acid2 page improperly rendered and states that that is how the page will look in the final version of IE8.
There are a number of additional sources that discuss IE8's new opt-in "IE8 standards mode" (which does render Acid2 correctly, but is not enabled by default) and I can provide you with a list of those if you want. —Remember the dot (talk) 06:29, 30 December 2007 (UTC)
Yes, please do provide those sources - not just to make me happy, but to justify the exclusion of IE from the list in the article. Quite frankly, I'm not in favor of the Wikipedia becoming a rumor mill, and I don't like the article being modified just because of some rumor about meta tags. CobraA1 (talk) 12:03, 30 December 2007 (UTC)
The video you gave me is a few months old, BTW, and as such should be given less weight than the more recent video showing IE8 passing the test. There's no law that says Microsoft can't change their minds about passing the Acid2 test. CobraA1 (talk) 12:03, 30 December 2007 (UTC)
The video I gave you is the video showcasing IE8 and Acid2. And yes, it does say that the final version of IE8 does not truly pass. —Remember the dot (talk) 19:40, 30 December 2007 (UTC)
I watched the video for the third time, and all they say is that quirks mode still exists and that the test will obviously not work in quirks mode. Many other browsers, including the existing Firefox and the upcoming Firefox 3, exhibit this behavior. CobraA1 (talk) 15:01, 31 December 2007 (UTC)
About 19 minutes and 15 seconds through the video, Alex Mogilevsky, a member of the IE team, points at a picture of the Acid2 test improperly rendered and states "The video in the bottom is a IE7 version of smiley face...What you're looking at is actually IE8. It is what it looks currently in IE8 and it will look exactly like this when we ship IE8 because we are not breaking any compatibility, and this is a compatible mode of IE8. And, uh, most of the web relies on particular behavior including particular incorrect behavior, so the incorrect behavior will still be there unless the new content wants IE to be in standards-compliant mode, and then they will ask us, and then we will show perfectly standard picture."
Other sources to consider:
  • IE8 now renders the “Acid2 Face” correctly in IE8 standards mode. [1]
  • IE8 now correctly renders the Acid2 smiley face in IE8 standards mode [2]
  • What is “IE8 standards mode”? Developers can now write sites based on standards, insert a flag that tells IE to render in IE8 standards mode, and IE will then switch its rendering engine to use this new mode...For compatibility purposes IE8’s rendering engine defaults to “quirks” or “standards” mode. Site developers will need to insert a new opt-in flag to request the page to render using “IE8 standards mode.” [3]
  • The adoption of any such opt-in switch today is zero. [4]
(emphasis mine) —Remember the dot (talk) 19:38, 31 December 2007 (UTC)
All you needed to do was to link to the site that explicitly states that "IE8 standards" mode is a new mode different from the regular "standards modes" - the other sources weren't that clear that this is an entirely new mode. —Preceding unsigned comment added by CobraA1 (talk • contribs) 20:09, 31 December 2007 (UTC)
It's unlikely that Microsoft will change its mind about the opt-in and default rendering mode. No one hopes IE will ever pass acid2 test by default, the web relies on its non-compliance with so many hacks. It would be shooting in its own feet, by breaking compatibility with nearly every websites currently existing. So in the past, they always said they weren't interested into passing acid2. Now is a turning point : they plan to implement a rendering mode that does fully pass the test. Authors will finally have a chance to develop cross-browsers websites. That is the important fact. Why ignoring that important fact in the timeline ? Add as many "not truly" as you want. --Fenring (talk) 13:42, 30 December 2007 (UTC)

[edit] This page is a f*****g mess

This is all a mess. That table has problems. Someone needs to fix it. Probably get rid of that table altogether. —Preceding unsigned comment added by 200.33.249.135 (talk) 00:02, 28 December 2007 (UTC)

[edit] n810 seems to pass Acid2

I just tried it and it looked good. Firmware is the latest. I don't know what the browser is. —Preceding unsigned comment added by 71.181.39.169 (talk) 05:41, 28 December 2007 (UTC)

Apparently it's Gecko-based, just like Firefox: [5]. —Remember the dot (talk) 05:57, 28 December 2007 (UTC)

[edit] Flock

Anyone know if Flock passes?The freddinator (talk) 18:36, 31 December 2007 (UTC)

I don't think so, since it uses a firefox build of gecko from before it was built to pass.--Oni Ookami AlfadorTalk|@ 20:01, 31 December 2007 (UTC)

[edit] Letter of the law vs Spirit of the law

Geez, this is becoming an edit war. Look, the Acid2 test is not meant to tell browser developers how their different modes (quirks, standards, "IE8 standards") are implemented. It's meant to demonstrate features and bug fixes that would make the lives of web developers easier - which is what IE8 does. Does it really matter that IE8 has to enter a new mode to enable all of this? Web developers can still enable the mode and take advantage of it, so the spirit of the test is still fulfilled even if it isn't fulfiulled tecnically. In the end, Acid2 still did the job it was meant to do. I don't see why people have to be so dogmatic about it.  —CobraA1 21:17, 31 December 2007 (UTC)

According to Håkon Wium Lie, who proposed the test in the first place, if IE8 does not pass the test by default then it doesn't truly pass:
It seems that IE8 will not display Acid2 correctly by default. Instead of following established conventions for how to switch between quirks and standards mode, it seems that Microsoft plans to introduce a new opt-in scheme based on the <meta> tag. And, since we cannot change the Acid2 test at this stage, it will not trigger IE8 standards mode. This issue must be addressed if IE8 is to be considered to pass the test. And hard-coding URLs is not allowed :-)
This is reinforced by Mark "Tarquin" Wilton-Jones:
Note, however, that Microsoft have said that they will require authors to opt-in to use the IE 8 standards mode. How they will do this remains unknown, but it is possible that the Acid 2 test will not qualify. This means that IE 8 may not actually pass the Acid 2 test unless it is changed to include the new trigger, whatever that may be. This would mean that it fails the test. Even if they add a special hack to allow the Acid 2 test to pass when hosted on its current server, that will also constitute a fail, since the idea is that the test will pass no matter where it is hosted. If the real test does not produce a pass when hosted on any server or stored locally, then this page will treat it as a fail.
In short, if a browser does not pass with the default settings, it does not truly pass. Since by Microsoft's own admission IE8 will not pass with the default settings, IE8 will not truly pass. —Remember the dot (talk) 21:33, 31 December 2007 (UTC)
Strange, I don't see where Håkon Wium Lie claims to have "invented" Acid2, nor do I see how it's relevant, since he's not a member of WaSP. And again, I ask: Who cares if it fails on this slight technicality? It's not as if this new mode is going to affect real web development! I'd be more than happy to add a single line of code if it means not having to write hundreds of lines of code to deal with the incompatibilities of older browsers.  —CobraA1 07:43, 1 January 2008 (UTC)
That's your own analysis. Yes, it's great that IE8 can be made to pass Acid2. It would (arguably) be better if it passed by default. We can talk all about this issue, just don't list IE8 in the list of browsers that pass because it does not truly pass. It will really seem foolish to list it as compliant after public builds of IE8 are out and excited web developers visit the Acid2 page only to find out that they didn't read the fine print.
"It would (arguably) be better if it passed by default." -- That would mean a lot of compatibility issues with older versions of the browser. I don't see why compatibility would not be considered a problem, so I fail to see how passing by default is really better.  —CobraA1 07:43, 1 January 2008 (UTC)
You can argue all you want about whether Acid2 is a good, valid test or not. That doesn't change that IE8 doesn't pass. —Remember the dot (talk) 17:55, 1 January 2008 (UTC)
I don't know if Håkon Wium Lie worked on the test or not. He was the one who originally proposed it. —Remember the dot (talk) 07:16, 1 January 2008 (UTC)
Got a reference?  —CobraA1 07:43, 1 January 2008 (UTC)
I already gave it to you above, but here it is again. —Remember the dot (talk) 17:53, 1 January 2008 (UTC)
That link again fails to tell me anything. Yes, the guy talks about Acid2, but it doesn't say he invented it:
  • Answer 1 - he plays with legoes
  • Answer 2 - he makes a thesis on CSS
  • Answer 3 - Opera now works fine with Acid2
  • Answer 4 - He talks about a "naked day"
  • Answer 5 - More about his thesis, and some mentions about different software
  • Answer 6 - Talking about brand names
  • Answer 7 - Talking about downloadable fonts
  • Answer 8 - Talking about Microsoft's dominance
  • Answer 9 - Talking about Opera as a pasttime
  • Answer 10 - Talking about botched up CSS on social networking sites
  • Answer 11 - Talking about Microsoft's lack of concern about Acid2 and IE7 beta
  • Answer 12 - Talking about Acid3
  • Answer 13 - Talking about not including constants in CSS
I'm sorry, I'm just not seeing it . . .  —CobraA1 22:49, 1 January 2008 (UTC)
Okay, jut read the grey stuff, and apparently it's the Opera Bits blog making the claim. Dunno how reliable they are, though, because I don't frequent it, and there's no additional information available to confirm the claim. —CobraA1 22:55, 1 January 2008 (UTC)
There's also [6] which states "Recently, he contacted WaSP to ask whether we could host the Acid2 test, which we agreed to do." At [7], he says "We will produce a test page, code-named Acid2, that will actively use features Web designers crave, such as fixed positioning of elements" (empahsis added). He has definitely been involved with Acid2. —Remember the dot (talk) 23:45, 1 January 2008 (UTC)
Howcome came up with the idea of the test. I wrote most of the actual test itself. Hope that helps. --Hixie (talk) 00:03, 31 January 2008 (UTC)
Good to know, thanks! —Remember the dot (talk) 04:13, 31 January 2008 (UTC)

Requiring a browser to pass with default settings in order for it to be deemed compliant is only logical. Consider the opposite example - Opera passes with default settings, but if you change certain settings, it does not pass. Does this mean that Opera is non-compliant based on a user-generated instance of a failed test? No. If you believe that IE8 should pass becuase it complies in a user-generated instance, then you enter a grey area of "sometimes it works, but you have to do a,b,c to achieve that." If this justification does not make sense, please tell me and I will try again. I am running on two hours of sleep :D Happy New Year, all The freddinator (talk) 16:43, 1 January 2008 (UTC)

"If you believe that IE8 should pass becuase . . ." -- Yes, I know it doesn't technically pass. That does not mean, however, that the web developers won't benefit.  —CobraA1 22:50, 1 January 2008 (UTC)
"then you enter a grey area of "sometimes it works, but you have to do a,b,c to achieve that." -- which is what we currently have with all of these "hacks" floating around. In fact, most current pages will break if IE suddenly becomes compliant by default, because they expect it to be broken. Microsoft is trying hard to be able to pass the Acid2 test and be more compliant with standards without current pages (which expect IE6 and IE7 behavior) breaking all over the place. With the huge legacy of broken browsers, backwards compatibility with existing web pages written to the broken browsers is tremendously difficult.  —CobraA1 22:50, 1 January 2008 (UTC)
No one is denying that this is a big step forward. I for one would gladly trade the many kilobytes of IE hacks for a single special flag. Still, since Microsoft is not improving compliance by default, IE8 will not actually pass Acid2. Microsoft could have chosen, for example, to make a special flag to trigger IE7 standards mode and use IE8 standards mode by default. Should they have? That's a different debate. —Remember the dot (talk) 23:41, 1 January 2008 (UTC)
I've said it before, and it appears I'll say it again a million times - I'm aware that it technically does not pass the test. However, I titled this section "Letter of the law vs Spirit of the law" for a reason - by giving the authors the choice of allowing the improved compatibility, I think the spirit of the test is still fulfilled, even if it technically doesn't pass. As for the question of what the default should be, that's simple: I don't think there's any way Microsoft is going to risk breaking millions of websites just so they can say they're in total compliance with Acid2.  —CobraA1 03:28, 4 January 2008 (UTC)
Wikipedia's job is not to interpret these things to decide what the "spirit" or intention of them was. Unless you want to start loading the article up with useless original research, if it does not pass the test, IT DOES NOT PASS THE TEST. It's that simple. This whole section is a complete waste of time as its just an attempt to shuffle some original research into the article by evaluating the merits of the test's criteria.--Oni Ookami AlfadorTalk|@ 13:35, 4 January 2008 (UTC)
I agree — if it doesn't pass it with default settings, it doesn't pass, period. Besides, anything Microsoft says about unreleased versions must be taken with a grain of salt anyway (see MS-DOS 5.0). - Sikon (talk) 15:43, 4 January 2008 (UTC)
I'm sorry - what I say is "original research" if a tiny minority of people are saying it doesn't pass the test, and I happen to side with the majority? WaSP is still insisting that IE8 passes the test. The exact mechanism used to determine which mode to use is still unknown, and it's possible that Microsoft may change their minds before releasing IE8. Yes, the idea that the page must be modified for IE8 to pass the test is still speculation, not a confirmed fact.  —CobraA1 12:02, 6 January 2008 (UTC)
In addition, I count WaSP as the final authority on what passes the test and what doesn't - not some random guy who might have been involved who talked to Opera.  —CobraA1 12:11, 6 January 2008 (UTC)
WP's purpose is to publish informative articles. It's not the place to advertise a product. So the timeline is not there for anyone to say "look, my browser truly passed acid2 before yours". It's there to show the important events and evolution related to acid2. You did right by renaming the section. I'd even change it to "History of compliance" or "History of implementations" or "History of layout engines". The rest of the article is fine as is. It clearly states IE8 will not pass acid2 by default, and thus is not among the compliant browsers. --Fenring (talk) 11:50, 5 January 2008 (UTC)
"WP's purpose is to publish informative articles." -- Precisely, and just because there are a few rumors from Opera people about IE8 not passing the test doesn't mean that it doesn't. The exact mode switching mechanism is still unknown, and IE8 is not released software and is subject to change in the future. I'm amazed at how much we're taking speculation as fact.  —CobraA1 12:07, 6 January 2008 (UTC)
Have you forgotten that Alex Mogilevsky, a member of the IE team, pointed at a picture of the Acid2 test improperly rendered and stated "The video in the bottom is a IE7 version of smiley face...What you're looking at is actually IE8. It is what it looks currently in IE8 and it will look exactly like this when we ship IE8 because we are not breaking any compatibility, and this is a compatible mode of IE8. And, uh, most of the web relies on particular behavior including particular incorrect behavior, so the incorrect behavior will still be there unless the new content wants IE to be in standards-compliant mode, and then they will ask us, and then we will show perfectly standard picture." (emphasis mine) —Remember the dot (talk) 02:11, 7 January 2008 (UTC)

[edit] External Links

Seems to be a lot of external links. Can someone more knowledgable than I on the subject clean it up? The freddinator (talk) 21:06, 30 January 2008 (UTC)

[edit] Name of article

I suggest this article should be given a more generic name, or perhaps broken down into several articles, now that Acid3 is out.

  • Acid browser tests (main)
    • Acid1
    • Acid2
    • Acid3

And Acid 4 when it comes out... Some info about formal browser test suites also would be fine.--itpastorn (talk) 19:30, 3 February 2008 (UTC)

I split the Acid3 stuff into its own article, and created Template:Acid tests and Category:Acid tests to keep track of the Acid test articles. I think that this is better than having one monolithic article on every Acid test that has been made and will be made. —Remember the dot (talk) 02:38, 4 February 2008 (UTC)

[edit] Time to add IE8!

Microsoft finally caved. Time to add IE 8 to the list :).  —CobraA1 03:12, 4 March 2008 (UTC)

Excellent! I am very excited. —Remember the dot (talk) 05:52, 4 March 2008 (UTC)

I notice the Beta has been added as non-compliant. This is only partially true: it renders the one at webstandards.org correctly and thus DOES pass the test. They've explained it already at [8]. --80.164.91.26 (talk) 23:57, 5 March 2008 (UTC)

Yeah, I'd agree that IE8 is compliant. If you set up the test exactly the same way as the web standards project has it set up, just on a different web site, then IE8 would pass. The copies of the tests actually change a seemingly insignificant part of the test (they are by and large not true copies).
That said, if the web standards people come out and say that this issue prevents IE8 from passing, I'd defer to them. —Remember the dot (talk) 05:47, 6 March 2008 (UTC)

[edit] Ian H says IE still does not pass the test

IE8 may render the page correctly, provided its taken at the correct URL. This will succeed:

These two will not:

It's an Active X security check that fails, and according to the spec it should then provide fallback content, which it does not! See Ian's comment at http://blogs.msdn.com/ie/archive/2008/03/05/why-isn-t-ie8-passing-acid2.aspx#8059324 I will remove IE 8 from the table, if no one comes up with strong counterargument.--itpastorn (talk) 09:34, 9 March 2008 (UTC)

Given that all we have is a post on a blog by someone who calls himself Ian Hickson but misspells his name, I think we should wait until we have a reliable source before we say that IE8 does not pass Acid2. Remember that verifiability, not truth, is the threshold for inclusion. -- Schapel (talk) 13:46, 9 March 2008 (UTC)
Well. Do we have a source that discuss these circumstances and say that they do not matter and that IE 8 beta 1 indeed does pass the test, then? Without such a source we should not include IE at all.--itpastorn (talk) 16:28, 9 March 2008 (UTC)
I don't see that we have any reliable source that discusses the circumstances in question. I do see reliable sources that state IE8 passes the Acid2 test, and those should be cited in the article. -- Schapel (talk) 18:48, 9 March 2008 (UTC)
(1) Those sources pre-dates the finding of this issue. (2) Ian's comment (and I have confirmed through e-mail that it really is Ian!) is quite clear. The guy who wrote the test does not consider IE8 to pass it! (3) The article I referenced clearly says that IE8 fails Acid 3, if it is taken through any other URL than "www.webstandards.org/...". One of the key points in Acid2 is to check for correct fallback behavior on the object element. IE8 beta 1 does not have such correct behavior, but luckily gets the graphical rendering right sometimes. How can that be considered passing the test?--itpastorn (talk) 19:11, 9 March 2008 (UTC)
I'm not arguing whether or not IE8 passes the test. What reliable sources say is that IE8 does pass the test. The threshold for inclusion in Wikipedia is verifiability, not truth. Because we cannot verify your personal communication with Ian Hickson, we cannot add that information to the article. Even if he publicly states that IE8 does not pass, we will probably still have to say that IE8 passes, but attribute the opinion that IE8 does not pass to Ian Hickson. Remember, we need to keep a NPOV and present all points of view neutrally (for example, Microsoft's claim that IE8 does pass Acid2 even though it renders differently from different URLs), even if we know one of the points of view is actually incorrect. -- Schapel (talk) 19:22, 9 March 2008 (UTC)
So as long as any unclued journalist, writing out of his ignorance, has said IE passes it goes into the article. I checked your reference. It does not discuss this point of concern at all. By what measure is that a better source than the man who wrote the test and who is probably one of the very leading experts in the world about standards? Should we include every opinion no matter how uninformed the author is? BTW, here is another reference: http://www.downloadsquad.com/2008/03/05/why-internet-explorer-8-might-not-pass-the-acid2-test/ And here is some more: http://ajaxian.com/archives/ie-8-on-acid And one more: http://www.builderau.com.au/blogs/betaliving/viewblogpost.htm?p=339270935 http://www.betanews.com/article/IE8_Beta_1_experiences_Acid2_hiccups_while_Acid3_is_rolled_out/1204822946 --itpastorn (talk) 21:56, 9 March 2008 (UTC)
There's no need to argue with me. Say whatever you want to say, as long as you provide a reliable source so we can all verify the information. We'll also need Microsoft's POV that they believe IE8 passes Acid2 so the information is NPOV (yes, even if that POV is incorrect). -- Schapel (talk) 01:16, 10 March 2008 (UTC)
It seems what this really boils down to is which url constitutes the Acid2 test. I do not see Ian H saying anything to the effect of "IE8 fails Acid2." He merely comments that what apparently is not Acid2 is not rendered correctly. I am removing the implication that Ian Hickson does not consider IE8 to pass Acid2. - Josh (talk | contribs) 18:26, 11 March 2008 (UTC)
OK, this is OR, but I happen to know that he really meant it that way. I have talked with him both through e-mail and on IRC. Asked about it he thought his comment on the IE blog was clear enough - and really, it is!--itpastorn (talk) 18:32, 11 March 2008 (UTC)
Although I do not disagree that that's what he meant, and that that's what he communicated in an e-mail to you, the fact remains that we do not seem to have a reliable source where we can verify that he said it. Therefore, we should not put the information in the article. The threshold for inclusion is verifiability, not truth. Sometimes Wikipedia contains incorrect information for this reason. Let's wait until all readers can verify that Ian Hickson says that IE8 beta 1 does not pass Acid2 according to a reliable source. -- Schapel (talk) 18:57, 11 March 2008 (UTC)

So how about this then: http://keryx.se/dev/ian-hickson-about-msie8-beta1-acid2.txt--itpastorn (talk) 20:43, 12 March 2008 (UTC)

No, a text file on some web server is not a reliable source. Articles should rely on reliable, third-party published sources with a reputation for fact-checking and accuracy. -- Schapel (talk) 20:59, 12 March 2008 (UTC)
And if you read what he actually says, he says the behavior is a bug, but he doesn't really say it causes IE8 to not pass Acid2. The failing he refers to is IE8 failing to load the requested resource. He doesn't say straight out This causes IE8 to fail Acid2. Remember, I'm not arguing with you, just pointing out the rules of Wikipedia and what you can and cannot say. You cannot read something into a blog post, verify over email that the person who claimed to say it said it, put up the email on a text file on your web server, and then use all that to justify making a change to a Wikipedia article that will surely be contested. Let's have it from a reliable source that says IE8 fails Acid2. -- Schapel (talk) 21:05, 12 March 2008 (UTC)
If you're not arguing, why do you refuse the edit ? The WP guidelines don't dictate what you "can and cannot say". They are just guidelines. And there are these two guidelines too : Don't game the system. and ignore all rules. --Fenring (talk) 01:03, 13 March 2008 (UTC)
I'm not refusing anything. I'm just saying that according to the guidelines, we shouldn't make the edit. You are welcome to make the edit and see what happens. My guess is that what will happen is someone will immediately challenge the edit, and without any reliable source to verify that IE8 fails Acid2, and with many reliable sources that verify that IE8 passes Acid2, the article will need to say that IE8 passes. -- Schapel (talk) 02:03, 13 March 2008 (UTC)
I agree with Schapel. As state in Wikipedia:Verifiability, The threshold for inclusion in Wikipedia is verifiability, not truth. DO NOT replace verifiable information with unverifiable one. Such actions will be considered offenses against Wikipedia policy.--216.254.162.74 (talk) 05:51, 13 March 2008 (UTC)
I have yet to see one single source that addresses the issue of IE 8 beta 1 not providing fallback to the object element and still thinks that it's OK. I would also seriously challenge your definition of reliable sources. Most journalists writing on IT-magazines are absolutely clueless about this issue. I provided links above that discuss this issue thoroughly and even MSIE's own developers say that if you take the test at any other url than http://www.webstandards.org/files/acid2/test.html it will not render correctly. That's verifiable. What follows is simple logic. IE 8 beta 1 does not pass the test. It is common sense. The test is checking if an UA provides fallback. It should for whatever reason it does not use the outer object. Ergo: IE8 beta 1 not passing has been proved. If Ian H himself is quotable could perhaps be argued, since it was in a comment where he accidentally misspelled his name. I think it's wise to include it, since a lot of people - who have no clue whatsoever that there even is a discussion about this issue - are going to say "but I've read that it passes". Seriously, there is no better judge in the entire world than Ian H on this issue. How about this for an analogy. Scientist X applies the law of gravity. However sometimes his calculations don't add up, because he has used the in such a way that through pure luck the result is sometimes right. He publishes his findings. People write about it. Then Einstein shows up and says he got it wrong. Now, who are we going to believe? (Yes, I know Einstein is dead.) --itpastorn (talk) 08:18, 13 March 2008 (UTC)
No, it is not really the "common sense", but largely original research. The URL you mentioned is the official one, and as stated in the article, it passes the official test. As for mirrors at other locations, the article has also already explained why the result is somehow different, including the reasoning of fallback mechanisms to the object element in a different domain. So such discussions are already included, as they are verifiable, unlike the ones you have claimed.--216.254.162.74 (talk) 17:21, 13 March 2008 (UTC)

Oh, and one source more: http://rutsum.com/internet-explorer-8-beta-fails-the-acid2-and-acid3-test "It clearly fails even the Acid2 test."--itpastorn (talk) 08:40, 13 March 2008 (UTC) And while I'm at it: http://www.w3.org/TR/REC-html40/struct/objects.html#h-13.3.1 Quote: 2. If the user agent is not able to render the object for whatever reason (configured not to, lack of resources, wrong architecture, etc.), it must try to render its contents. (emphasis added)--itpastorn (talk) 08:45, 13 March 2008 (UTC)

I have yet to see one single reliable source that addresses the issue of IE 8 beta 1 not providing fallback to the object element and says IE8 fails the test because of it. If that's what you want to say in the article, that's what you'll need to find. If you want to argue about Wikipedia's guidelines, this is not the place to do it. And no, you can't make up for a lack of sources with [WP:OR|original research]]. As I say, go ahead and make the edit if you think you can justify it. My one piece of advice is simply not to get into an edit war over the issue. -- Schapel (talk) 11:52, 13 March 2008 (UTC)

IMHO it's reasonable to conclude that IE8 fails the Acid2. Thus we should remove mention of IE8 passing, since it isn't verifiable that IE8 handles all exceptions correctly. --Kjoonlee 12:41, 13 March 2008 (UTC)

It may be reasonable to conclude, but according to Wikipedia guidelines, we should not put our own conclusions in articles, as that constitutes original research. But if you want to make the edit, go for it. -- Schapel (talk) 13:17, 13 March 2008 (UTC)
Removing data should not count as including conclusions, though. --Kjoonlee 13:23, 13 March 2008 (UTC)
Alright, let's say you remove all the claims that IE8 passes Acid2 from the article. Someone else will simply put them right back in, complete with citations from reliable sources. If you keep removing the properly cited information, that's vandalism. If it's your wish to get into an edit war and vandalize the page, that's your prerogative, though. -- Schapel (talk) 13:32, 13 March 2008 (UTC)
Nope, if I had time on my hands, I'd mention that the MSIE team claims they pass Acid2. I'd also add that IE8 doesn't pass the Acid2, depending on the URLs. I'd add the blog post from the MSIE team itself as a citation. --Kjoonlee 13:37, 13 March 2008 (UTC)
Kindly assume good faith (WP:AGF) and think once more about WP:NPOV, please. --Kjoonlee 13:39, 13 March 2008 (UTC)
I am assuming good faith. I think you truly believe that IE8 does not pass Acid2, and that you're in the right. For arguments sake, let's say I also believe in my heart that IE8 does not pass Acid2. But removing information that is backed up by reliable sources and replacing it with information that is not does not follow Wikipedia's guidelines. If you get into an edit war with someone and continue to make those changes, it will be seen as vandalism by an admin, so don't be surprised if one blocks your account. I'm not an admin, and I will not participate in these edits, so consider what I say to be friendly advice. I was blocked in 2006 by reverting edits that said Firefox 2 was released, even though I had a reliable source stating that it had not yet been released. I broke the 3-revert rule and was blocked for a day. -- Schapel (talk) 15:07, 13 March 2008 (UTC)
But adding information which is backed by an incorrect source is worse. I didn't get into an edit war and I don't remember editing the Acid2 page. Don't brandish warnings in vain, please. --Kjoonlee 15:13, 13 March 2008 (UTC)
I agree it might be "worse" in some philosophical sense, but Wikipedia policy is that what the sources say takes precedence over what's correct. I'm simply explaining the rules to you. If you wish to ignore the rules, go ahead. -- Schapel (talk) 15:29, 13 March 2008 (UTC)
I suggest you take another look at WP:V, which states that statements should be backed by the person adding them. What's incorrect is in fact unverifiable, thus should not be added. --Kjoonlee 21:58, 13 March 2008 (UTC)

[edit] Ian H verification now provided

http://lists.w3.org/Archives/Public/www-archive/2008Mar/0018.html

As for me personally, I am going to heed Ians advice in the lasts lines. I asked for a "formal ruling", as the question about whether IE 8 beta is in or out is boolean. I believe my case to be settled with this clarification. But it is only a matter of credibility for WP. Not an issue for web developers, since they ought to know anyway that IE 8 b1 has this bug. By the time IE 6 and 7 are dead and gone I am 100 % sure MS will have fixed this bug. In fact I am sure it will be fixed for beta 2. A few years from now we can use the object element and get predictable results. Jay! Anyway, do what you want with this information. This discussion is over for me.--itpastorn (talk) 08:47, 14 March 2008 (UTC)

So now you will spend your time doing something productive and useful to society instead of bickering over meaningless questions. Great! -- Schapel (talk) 11:35, 14 March 2008 (UTC)
Congratulations! It seems that you finally understands what is productive & useful to society and what is meaningless struggles. Good luck for your future!--128.100.109.33 (talk) 21:11, 14 March 2008 (UTC)

I'd like to clarify that Ian H didn't say that the object fallback issue is irrelevant to passing Acid2. He said that it's a waste of time to argue whether or not a browser passes Acid2 because the real issue is interoperability.

I've updated the article to say that IE8 passes at www.webstandards.org but fails at webstandards.org, which I think captures the whole situation pretty well. We can still list IE8 as a complaint application and include it in the timeline, just note that there are still interoperability issues not fully captured by the test. —Remember the dot (talk) 05:13, 21 March 2008 (UTC)

[edit] Microsoft stated that passing Acid2 is a goal for Internet Explorer 8

Correct me if I'm wrong, but I was under the impression that passing Acid2 was not a goal for IE8 but a consequence of following specs better.Lyml (talk) 00:01, 18 March 2008 (UTC)

No, it was definitely their goal to pass Acid2. They had to implement some new "very large" features in order to pass. This video has more information. —Remember the dot (talk) 04:25, 18 March 2008 (UTC)

[edit] IE8 and Standards Compliance

Alot of sections relating to IE8 mention that by default standards mode will not be enabled. This decision was reversed by Microsoft at the beginning of March [9]. Please can somebody confirm that the notes relating to IE8 not being in standards mode by default can be removed? James (talk) 13:50, 18 March 2008 (UTC)

In this article? Where? -- Schapel (talk) 14:04, 18 March 2008 (UTC)

[edit] Browsers not really passing Acid2?

Firefox 3b4 and the latest WebKit nightlies don't pass Acid2; the noses are slightly the wrong size. Is that good enough to pass already? --M1ss1ontomars2k4 (talk) 00:29, 2 April 2008 (UTC)

[edit] Missing screenshots

It would be really great if we could get screenshots of the versions of Safari, Konqueror, iCab, and Prince that had been officially released by the time the test was released. This would help illustrate just how spectacularly all the browsers failed. Unfortunately, being a Windows/Linux user, I have no way to get a screenshot of Safari 2.0 taking the test. Is there a Mac user out there who could get one, preferably at 800x600 resolution? —Remember the dot (talk) 05:26, 3 April 2008 (UTC)

[edit] Firefox 3 Beta 4

Firefox 3 Beta 4 passes the test. Can we add that in somewhere? Altonbr (talk) 14:59, 4 April 2008 (UTC)

The status of how well various builds of Firefox do on Acid2 seems to be well documented in the article. -- Schapel (talk) 15:00, 4 April 2008 (UTC)

[edit] Iris Browser

I added the Iris Browser to the list, although I can't find any source to explicitly state that it passes the test (albeit I didn't find any mention of Acid2 in the reference linked for Prism 0.8 neither), but it's publicly available and it passes the test for my WM6 device. It's also the first publicly availabe mobile browser to pass the test I believe. Ufopedia (talk) 07:13, 4 June 2008 (UTC)

"Remember the dot" has removed it for "notability" reasons, although I'm uncertain of the precise part guideline you may have violated.  —CobraA1 17:10, 12 June 2008 (UTC)