Re: CSS2.1: Markers are needed!

On 8/5/02 3:29 PM, "Peter Sheerin" <> wrote:

> I'm troubled with the removal of markers from the CSS2.1 spec, and take issue
> with the stated reason of allowing user agents to become compliant. The
> removal of this and other features does not, in fact, accomplish this, because
> browsers are not following the CSS2 spec in other more significant ways. All
> it serves to do is remove features that are, or would be, useful to Web
> authors and designers at large.

This is a standard(s) misconception.

Features that are not implemented are of zero usefulness to Web authors and
designers at large.

As far as features that "would be", look for them in CSS3.

> With this reasoning, shouldn't position:fixed
> and overflow:auto be removed from CSS2.1, since IE/Win doesn't support the
> former and Opera doesn't support the latter?

position:fixed is supported by two (actually more) user agents today - and
quite interoperably as well.

similarly with overflow:auto, although overflow could use some good test
cases to work out interoperability issues.

> [12] Markers. The one that most troubles me is the elimination of markers.
> Since list-style provides no way to size or style bullets separately from the
> list content, and doesn't provide any means of specifying the use of
> characters as bullets, the removal of markers takes away significant
> functionality.
> Also, it would be incorrect to say that no browser supports this, since
> Netscape 6.x and higher (including Netscape 7.0 and Mozilla), (though they
> display the bullet in addition to the marker when used with OL or UL lists--an
> easy bug to fix, and it's not present when using markers on elements that
> don't generate markers on their own). Try this test page as an example:

Please see the CSS3 Module: Lists.

Markers have been obsoleted and superceded by the mechanisms described in
that draft.

> [12.2] Counters. I'm less concerned about counters being removed, since I'm
> not aware of any working implementations. Also, I'm not sure that the
> description in CSS2 is written clearly enough for either implementors or
> authors. For instance, I would argue that one needs to be able to reference
> the counter value in other elements, so that one could number paragraphs, for
> instance, and cross-reference them by referencing the ID value of an element,
> and repeating its counter value (e.g. Paragraph {counter(ContractExemption)}
> shall be interpreted...). It's not clear to me if such cross-referencing is
> possible in CSS2, and the lack of that feature combined with the lack of
> implementations makes it acceptable (to me, at least) to defer until CSS3,
> when this and other issues can be resolved.

There has been discussion regarding Counters, and it looks like they will be
added back to the next draft.  I believe Opera implements them, and there is
an expectation of at least one more compliant interoperable implementation
during the CR period.



Received on Monday, 5 August 2002 19:07:19 UTC