Re: [CSS21] bidi, text-align, and list markers

We (same group as my post yesterday morning) achieved consensus on the
following at the "Additional Requirements for Bidi in HTML" f2f yesterday
afternoon:

(Section 3.10) CSS3 will include a new property, list-style-direction, with
the values left, right, start, and match-me. [Aharon: the last is a
placeholder name until we find something better]. The CSS initial value is
start, which means according to the list item’s direction. The match-me
value is like start, but is inherited as a computed value of either left or
right. To get markers to appear all on one side in most cases, the default
style sheet will specify ":not(li) > ol, :not(li) > ul {
list-style-direction:match-me;}". This will still align the list items by
their own direction, and the user will have to explicitly use li
{text-align:match-parent}. We can not make this the default without breaking
the inheritance of text-align.

When one actually wants the markers to appear on different sides, one needs
to set up margins or padding appropriately.

[End of quote]

The rationale here is that opposite-direction list items as they are
currently displayed are broken. The marker gets cut off in most cases, and
in different ways in different browsers. The reason it gets cut off is that
it, like the same-direction marker, is supposed to show up "in the gutter",
i.e. in the margin or padding. The list sets up the margin on the start
side, so the same-direction list items' markers have where to show up, but
the list does not set up a margin on the end side, and the
opposite-direction markers get cut off. This can not be fixed by the list
automatically leaving a margin there because this would waste real estate in
the 99% case where there are no opposite-direction list items.

So, given that opposite-direction list items are broken, and we thus don't
need to worry about backward compatibility, and that all markers on one side
is what's usually wanted anyway, we allowed ourselves to try to make it the
default.

Fantasai and Tab, I do not remember, why did this have to be done in the
default style sheet? I.e., why couldn't the CSS initial value for
list-style-direction be match-me to begin with?

Aharon

On Tue, Jun 8, 2010 at 8:56 AM, CE Whitehead <cewcathar@hotmail.com> wrote:

>
> I am sending the test files I promised though this does not seem to be what
> this group is focused on today -- I have viewed them in IE; IE automatically
> puts in different bullet for different list levels which helps solve the
> problem of clear display of levels;
> ie actually does inherit the side (left or right) the bullet will be
> displayed on from the higher list level unless you explicitly override that
> by setting the directionality of the list item.
> So I feel I have control enough I guess as things are -- and this does not
> seem to be on today's agenda anyway.
> My mistake.
>
> Best,
> C. E. Whitehead
> cewcathar@hotmail.com
>
>
> ------------------------------
>
> From: cewcathar@hotmail.com
> To: public-i18n-bidi@w3.org
> CC: aharon@google.com; adil@diwan.com; shachar@shemesh.biz;
> textexin@xencraft.com; fantasai.lists@inkedblade.net; addison@lab126.com;
> jackalmage@gmail.com; ambrose.li@gmail.com; alan@css-class.com
> Date: Mon, 7 Jun 2010 22:19:33 -0400
>
> Subject: RE: [CSS21] bidi, text-align, and list markers
>
>
>
> Hi.
> Tex, O.k. if these are embedded block elements -- I could not see that from
> the other example; from your example yes I did see that.
> These are good examples of a problem -- perhaps a good reason to have the
> bullet on the same side for all list items -- but I will have to try some
> examples first, and will get back on this tomorrow first thing.
> Thanks.
> Best,
> C. E. Whitehead
> cewcathar@hotmail.com
>
> From: Tex Texin <textexin@xencraft.com>
> Date: Sat, 5 Jun 2010 20:03:41 -0700
>
> > I thought the footnote example was reasonable.
> > . . .
> > However, if bullets are only follow one direction, it can cause
> confusion.
> > (Might be unavoidable.)
> > For example, the footnote example confused me a bit because I wasn’t sure
> if
> > the ending periods were bullets or not.
>
> > If the chosen direction is RTL, then it might be confusing to have LTR
> > items:
> >
> >
> Item
> > 1
> >
> Item
> > 2
> > A long multiline item that wraps perhaps 3
> >                       And so this item and the next 4
> >                                        items are now unclear.
> >
> > am trying to show that the 4 could be a bullet number or could be part of
> > the wrapped phrase “next 4 items”.
>
> > Having multiple directions also makes sublists more complex…
>
> > tex
>
>
>