Talk:Acid2
From Wikipedia, the free encyclopedia
[edit] Coverage of Test?
The intent of this test seemed to be to goad Microsoft into improving its layout rendering in fundamental ways. While passing the test is no small feat, shouldn't the article contain a note about the general coverage of the test, that is to say, the test actually doesn't test basics, but is a specific 'stumper' test, whose value is minimal as long as Microsoft ignores the challenge?
[edit] Safari was the first to pass the Acid2 test completely
I'm changing the line saying Safari didn't pass Acid2 first because it didn't hide the scrollbars, as it does on my Safari, and I don't know how long it's been like that... Just look at the date here, http://mac1.no/node/1659 way before Opera 9.
[edit] Konq and iCab do not pass Acid2!
The only browsers that pass Acid2 is Safari and Opera (weekly build). Konq and iCab only claim they pass the test but they don't. They fail to apply one of the styles required by the test:
html { ... overflow: hidden; /* hides scrollbars on viewport, see 11.1.1:3 */ ... }
Konq and iCab don't hide the scrollbar in the viewport. Therefore, they are very close but they don't pass it yet. —The preceding unsigned comment was added by 83.31.113.57 (talk • contribs).
- I tend to not agree, CSS2.1 section 11.1.1 says of overflow:hidden "This value indicates that the content is clipped and that no scrolling user interface should be provided to view the content outside the clipping region." (emphasis mine)
- SHOULD NOT and MUST NOT have a very different meaning in a spec like this (see RFC 2119 referenced by the CSS spec). In fact, iCab and Konqueror MAY both pass the test, as long as the scrollbars are there on purpose :) —The preceding unsigned comment was added by 81.188.84.69 (talk • contribs).
- From RFC 2119: "SHOULD (...) there may exist valid reasons in particular circumstances to ignore a particular item, but the full implications must be understood and carefully weighed before choosing a different course." - one of the implications is that it won't pass Acid2 ;)
- I've blogged here why iCab (and Konqueror) pass the Acid2 test despite showing scrollbars on the viewport. Although both browsers can hide the scrollbars now.
- From RFC 2119: "SHOULD (...) there may exist valid reasons in particular circumstances to ignore a particular item, but the full implications must be understood and carefully weighed before choosing a different course." - one of the implications is that it won't pass Acid2 ;)
- Please point out where the scrollbars are showing in the image at this page or in this image. It appears that Konq does pass the test. — Mperry 18:19, 12 March 2006 (UTC)
- This was fixed in Konqueror 3.5.2, as the article now mentions. (Konq showed the bars in 3.5.1) If you want to see for yourself, you can get 3.5.2 on a live CD, see [1]
[edit] Specifics on the test
Does anyone know what this test specifically tests? Is this really too technical to be included here? My first reaction to the this article is "why on earth is it so hard for these browsers to render allegedley standard HTML?" . --Njerseyguy 21:20, 12 March 2006 (UTC)
- I believe it essentially tests proper rendering of HTML, CSS, PNG and Data URLs, but also how browsers handle invalid code. Full details can be found on their guide page. --ADeveria 21:22, 13 March 2006 (UTC)
[edit] Pictures
What about adding *some* pictures to illustrate this article? I was thinking about one or two screenshots of browsers _incorrectly_ displaying the test...
FiP Как вы думаете? 08:48, 14 April 2006 (UTC)
- I was thinking of this too, seems like a good idea. Mostly just wasn't sure where to begin or what to include. But probably IE6 and/or IE7's renderings would make most sense to have. Maybe also other browser's abilities at exactly the time the test was released? --ADeveria 12:09, 14 April 2006 (UTC)
-
- Perhaps add a screenshot of a browser correctly rendering the acid 2 test, such as Opera 9(which I happen to be using right now) or Safari or simular. But Opera is probably the best well know out of the ones that have passed and is the only windows browser that passes. Beta or otherwise. It sucks but it just happens that the majority of people use windows.
- Robert Maupin 23:41, 19 June 2006 (UTC)
-
-
- What's the point in a screenshot of a browser correctly passing the test? The image looks just like the reference rendering, and we already have an image of the reference rendering. --X-Man 03:43, 22 July 2006 (UTC)
-
[edit] Gecko's Compatability
Gecko's rendering of Acid2 is able to be substantiated, all builds are able to be reproduced by anyone with the proper buildtools and cvs, no build (except for ones with local patches), are closed from public consumption. The "not public build" is a misnomer. --Callek 03:12, 28 April 2006 (UTC)
- An interesting point...though I think that's something worth mentioning in the "notes" section of its entry. Personally I still wouldn't consider it "public" until any average joe can simply download it, install it and see it for himself. Perhaps you can think of a better term for it, indicating that the build requires some work to run? --ADeveria 12:41, 28 April 2006 (UTC)
-
- I think it should be changed to something like "source code is available but requires compiling". The word assembly is strictly wrong since firefox is compiled, not assembled. 218.103.137.41 14:50, 23 June 2006 (UTC)
[edit] Opera
I just downloaded Opera 9 and it fails the Acid2 test. The smiley is the right shape but yellow rows 2-4 are red! Does anyone know why? Also when I zoom in and out the smiley breaks up into little bits even when it goes back to 100% :( The whole browser seems to be a lot stabler and slower. On a brighter note, Wikipedia pages are rendered a lot faster, but this never really mattered since the server response time is much slower. 218.103.137.41 14:50, 23 June 2006 (UTC)
- Do you have the View>Fit to Width" option on, perhaps? There may also be other settings that can cause it to render incorrectly, but if all is correct Opera 9 will pass it. Zooming in and out can indeed mess it up, but that's irrelevant to the actual test. --ADeveria 15:16, 23 June 2006 (UTC)
- See the Acid2#Passing conditions section I've just added. It should explain things. --X-Man 03:40, 22 July 2006 (UTC)
Haha, is the Opera 9 Beta easter egg real? Sounds like someone got a bit cocky. :D 82.109.166.178 03:10, 17 December 2006 (UTC)
[edit] Opera Software's involvement
From the opening sentence:
Acid2 is a test case designed by the Web Standards Project, with assistance from employees of Opera Software
The citation gioven was this CNET article written by Hakon Wium Lie, chief technology officer of Opera Software. Unfortunately that link makes no references to Opera's involvement, it simply states:
"To ensure that IE 7 does not become another failed promise, the Web community will issue a challenge to Microsoft."
I added the bold to make my point clearer. He is not saying Opera Software is doing this, he is saying the web community as a whole are creating Acid2. This includes people from Opera Software, Mozilla, Apple, etc. The statement "with assistance from employees from Opera Software" implies that they were the only ones, which is incorrect. --Lethargy 20:14, 27 June 2006 (UTC)
Opera's involvement is mentioned on the acid2 test page.
Acid2 is a year old now, and recent beta builds of Opera 9 now support the test. Håkon Wium Lie, CTO of Opera Software and one of the contributing authors to the Acid2 test, says: “Some people thought Acid2 would be easy for Opera since we were involved in shaping the test. Not so. Acid2 rightfully exposed many bugs in Opera and we have squashed them, one by one.”
I believe the article should reflect this. --phocks 05:03, 6 July 2006 (UTC)
- IIRC one of the developers at mozilla was also a "contributing author", and probably people from several other companies. I'm not sure if it makes sense to list each individual who contributed, as we may not have the space and it might not contribute much to the article. If we only list a few of the people, it may be POV because it sounds as if they were the only people (or companies) who supported it. --Lethargy 00:07, 3 August 2006 (UTC)
[edit] Firefox download link
In a previous version of this article, there was a download link to the Firefox build. I don't see a purpose for removing it, and I also don't know why was the link dubbed "questionable". And, even if the link is really "bad", why was the coloring and the "unofficially published build" text reverted? Thanks! --NetRolller 3D 16:16, 4 July 2006 (UTC)
[edit] Firefox 3.0 image
The image for the firefox 3 build seems wrong. The picture in the article appears to be the screen shot from the unused branch 1.6. Firefox 3 does properly render the acid 2 test. This site shows the 1.6 and 3.0 images: http://www.howtocreate.co.uk/acid/
- Not entirely correct. The main Firefox 3 branch does render the Acid 2 test as shown in the article, but another branch (the reflow branch, as noted in your link, which will be merged with the main FF3 branch) renders the test perfectly. In fact, I'm running the latest build from that branch right now. :) --Guess Who 00:20, 22 October 2006 (UTC)
[edit] Firefox 3.0 passing
Minefield passes the test the most current windows available build anyway Atomic1fire 06:08, 3 March 2007 (UTC)
[edit] Outdated Test
The test is VERY different now. It can be found at http://www.webstandards.org/action/acid2/
[edit] Acid1
What ever happened to the acid1 test? what if browsers want to try that test out now? and doesn't that test also need an article page? 165.230.46.144 18:16, 26 October 2006 (UTC)
Visit: http://www.w3.org/Style/CSS/Test/CSS1/current/test5526c.htm. 217.44.169.222 18:36, 6 December 2006 (UTC)
[edit] Browser final releases
With the new final releases of IE7 and Fx 2.0, those pictures should be in the article instead of the beta/test images. 165.230.46.144 18:20, 26 October 2006 (UTC)
[edit] Firefox 3 passing?
Im using Firefox 3 alpha 1, Gran Paradiso with Gecko 1.9 alpha 1, but its not rendering properly. The mouth is off to the and theres a white line a the top of the head. I dont think firefox 3 passes acid2 yet.Superway25 04:16, 22 December 2006 (UTC)
- Alpha 1 (Gran Paradiso) doesn't pass. Only nightly builds (Minefield) do. The next public version will most probably pass it. - Sikon 10:51, 22 December 2006 (UTC)
I thought Gran Paradiso was supposed to be better than Minefield, it came out later. Is minefield better?Superway25 19:36, 22 December 2006 (UTC)
- Gran Paradiso is the codename for Firefox 3.0 alpha 1, which was essentially a one-time snapshot of the trunk. Minefield is the codename for the trunk builds, which are essentially the bleeding edge of ongoing development. The reflow branch, which implements proper Acid2 support, was merged with the trunk after the release of the public alpha, therefore the trunk builds pass the test and the public alpha doesn't. - Sikon 07:10, 23 December 2006 (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] FF dev build acid2
havent got time to read everything, but i noticed that it says that "and development builds of Mozilla Firefox." under perfect acid2. To my knowledge the latest gran paradiso Firefox 3.0 Alpha 2 has a dislocated nose (a bit to the left) on the smiley. Does Dev builds mean cvn releases? If im missing the point, then send a mail to comradekingu@gmail.com
-kingu
