Skip navigation.

miscoded

the web is a hack

Worthless? Certainly not, sir.

I was looking over the spreadsheet covering Mozilla’s Acid3 failures, and it struck me that very few of the fixes would substantially improve the Web or the browser. They are bugs and they will be fixed (except maybe SMIL… wtf?), but they don’t impact authors or users at all. Looks mostly like an opportunity for grandstanding about “commitment to standards.”


So says Rob Sayre, who seems to be a Mozilla contributor or developer. (Sorry to not know exactly who you are Rob, but when neither this nor this were of any help I gave up.. p:)

A quick look at the Moz bug spreadsheet he refers to shows me for example:
  • Several DOM Range issues - every author of Rich Text Editor components really struggles with Range incompatibilities
  • Several dynamic cloning of form element bugs - nasty, potentially dataloss-causing gotchas
  • event.stopPropagation() and capture - we've been a victim of Gecko event capture issues that made websites test broken code, every detail of capture and propagation really matters

I'd like to remind Rob Sayre that the smallest incompatibilities can be the most harmful. Details and corner cases are not "worthless", they are important.

Joel is from Mars, browsers are from Venusit's so nice to say ... goodbye!

Comments

shadowk 30. March 2008, 20:14

Passing Acid2 was for a long time, not important, passing Acid3 also looks like it's getting the same treatment.

Some say that as long as IE doesn't pass any Acid test it's not important for others to pass.

Then again if there are standards and tests to show how browsers don't follow them. Standars are called standards because everybody once upon a time agreed to followed them, so we can write everything once and have it run.

Of course if you read the blog posts of the maker of Acid3 then you can see that he is kinda targeting Opera and Webkit.

iheartwp 30. March 2008, 22:33

sound a little jealous to me.
whats with your verification codes.
http://mz.iheartwp.com/2008/03/30/opera-jealous/

Andrew Gregory 31. March 2008, 01:52

To be fair, he does say "very few", implying that some are worthwhile.

For example, I'm sure someone could have turned up a more important consistency problem than the "-0 toFixed" issue?

I also feel that speed tests (part of test 26) are a low enough priority to be not worth a spot in an acid test. There's already a wide variety of JS performance tests around - another one wasn't needed.

Still, I think Acid3 is on the whole, something much needed - a wide ranging implementation check helping to ensure all browsers are working from the same page.

Cross-browser consistency is, IMO, Opera's biggest "compatibility" problem.

BleedingHeart 31. March 2008, 05:33

From a quick google search he appears to be Mozilla QA.

If everyone being on the same page is important is it safe to say Acid2 only comes into its own when FF3 and IE8 are released, and Acid3 won't come into its own until finals of all the main browsers are released with passing grades?

iheartwp if FF was ahead in the context of Acid3 and Opera behind then you could brush off an Opera response as jealousy, when the positions are reversed within the same context calling them jealous seems strange, unless you specify what they are jealous of beyond that context.

I would of left a comment on your blog but it says you require my email and I hate that, also you don't have to be registered to leave a comment on my.opera blogs its just Hallvord has chosen it that way.

Kelson 31. March 2008, 16:11

Any Acid-style test only becomes useful once the majority of browsers support the features tested. The whole point is to motivate browser vendors to give web developers more tools, and web developers can't reliably use those tools until they can count on them being available for most visitors.

It's a network effect. Consider telephones. If no one you knew had a telephone (think back 100 years or so), it would be worthless. Why would you buy one? But if everyone you know has a telephone, it's very useful indeed.

How to use Quote function:

  1. Select some text
  2. Click on the Quote link

Write a comment

Comment
(BBcode and HTML is turned off for anonymous user comments.)

Type the two words displayed in the image below:


Smilies