W3C home > Mailing lists > Public > www-style@w3.org > March 2008

Re: Fwd: [cssom-view] small update

From: Garrett Smith <dhtmlkitchen@gmail.com>
Date: Thu, 6 Mar 2008 14:00:55 -0800
Message-ID: <c9e12660803061400s64bcf2fay2c5d7492cd08e1c4@mail.gmail.com>
To: "Anne van Kesteren" <annevk@opera.com>
Cc: Www-style <www-style@w3.org>

On Thu, Mar 6, 2008 at 12:55 PM, Anne van Kesteren <annevk@opera.com> wrote:
> On Thu, 06 Mar 2008 21:36:29 +0100, Garrett Smith <dhtmlkitchen@gmail.com>
>  wrote:
>
> >>  I think it's ready for Last Call.
>  >
>  >  Not without more tests its not.
>
>  Formal tests are typically done during Candidate Recommendation stage.
>
>
>
>  >  The way you've spec'd offset* doesn't really reflect what browsers do.
>
>  As far as I can tell it matches three out of four browsers relatively
>  well. Not perfect yet, of course, but that's to be epxected.
>

Even a basic, simple test would fail what CSSOM:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
        "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html lang="en">
 <head>
  <title>offset: margin on BODY</title>
  <script src="support/test.js"></script>
  <link rel="stylesheet" href="support/test.css">
  <style>
   body { margin:10px; position: relative; }
   #t { position: relative; top: 0 }
  </style>
 </head>

 <body>
  <p id="t">TEST</p>
  <hr>
  <script> test('BODY', 10, 10) </script>
 </body>
</html>

This is testing the AVK CSSOM Spec. I do not agree with the expected
result which (coincidentally) is only attained in Opera.

This is not an edge case at all.There are also more complex, yet
benign examples which will fail.

But since I (and the other 3 browsers) do not agree on the expected
result, it seems only fair to ask the question:"What should the
expected result be?"

To answer this question, I think a test is the best approach. I do not
think this is unreasonable. Nor do I find it to be inappropriate.
Putting up little tests on each thread is not exactly the way to
maintain a test suite.

Spec'ing existing features is not the same thing as designing
futuristic HTML 5 features. If you make stuff up, it causes problems.

There's still an unanwsered email in the archives, and plenty of questions.

Garrett

>
>
>  >  You seem to be in denial about this, or are unwilling to investigate
>  >  the matter more. You also seem to want to avoid a lot of questions and
>  >  don't want to test. I'm not sure what the reluctance is. What is the
>  >  problem, Anne?
>
>  I have tried to answer your questions, but it seems we simply disagree on
>  what the specification should be based. I'm not sure why you keep being so
>  hostile in your comments as I mean no harm.
>

You're calling me hostile, huh?

Things like putting: "..." after I asked you if you're waiting for me
to get tired replying should be considered what, "harmless"? Some
questions I've asked have been snipped, unanswered or answered
carelessly, and I've had to repeat the questions, which has led away
from the original point.

It is difficult for me to participate in a discussion with such distractions.

Now, regardless of my own opinion on your behavior, there are several
problems with the spec. Condemning you as harmless would do nothing to
improve the spec, would it?

There's still an unanwsered email back in the archives, if I'm not mistaken.

Garrett

>
>  --
>
>
> Anne van Kesteren
Received on Thursday, 6 March 2008 22:01:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:02 GMT