W3C home > Mailing lists > Public > www-style@w3.org > June 2011

Re: [css3-lists] Published as WD!

From: Håkon Wium Lie <howcome@opera.com>
Date: Wed, 22 Jun 2011 17:51:29 +0200
Message-ID: <19970.3969.236456.644479@gargle.gargle.HOWL>
To: "Tab Atkins Jr." <jackalmage@gmail.com>
Cc: www-style list <www-style@w3.org>
Also sprach Tab Atkins Jr.:

 > > So, when does the ambiguity appear?
 > 
 > The counter style written above produces lists like:
 > 
 > abc. First item
 > abc. Second item
 > abc. Third item
 > abc. Fourth item
 > 
 > Your proposal would make it ambiguous whether it should produce the
 > above, or instead produce:
 > 
 > a. First item
 > b. Second item
 > c. Third item
 > a. Fourth item

I suggest resolving that ambiguity by saying that, when only
one string is specified, it should be interpreted as a list of single
glyphs.

 > > Examples are good, but I still think you need to describe what
 > > "doubling" or "tripling" means. It's defined in the pseudo-algorithm,
 > > but that's hard to read.
 > >
 > > How about:
 > >
 > > The ‘symbolic’ counter type cycles repeatedly through its provided
 > > glyphs. At the end of a cycle, the string length is increased by
 > > adding one more glyph of the same kind.
 > >
 > > Example:
 > >
 > > @counter-style foo {
 > >   type: symbolic;
 > >   glyphs: 'A' 'B';
 > > }
 > >
 > > will produce a list like this:
 > >
 > >   A.
 > >   B.
 > >   AA.
 > >   BB.
 > >   AAA.
 > >   BBB.
 > 
 > I couldn't come up with a way of avoiding using "doubling" without
 > either being unclear or just restating the pseudoalgorithm in text.
 > Instead, I've added a simple inline example to the text.  Please let
 > me know if this is okay!

Examples should be marked up as such.

I still think the terms "doubling" and "tripling" are problematic and
think they can be avoided.

 > (I've also finally moved the footnotes counter-style to being an
 > example for symbolic, like I said I'd do earlier.)

You write:

  It can be used for footnote-style markers, and is also sometimes
  used for alphabetic-style lists for a slightly different
  presentation than what the ‘alphabetic’ type presents.

I suggest rewording this text from normative text to examples.

 > > I suggest listing them in an appendix as examples for others to
 > > cut/and paste from. But I don't think they should be predefined.
 > >
 > > I specifically suggest make all the predefined styles non-predefined
 > > in the next version the WD.
 > 
 > I think the current predefined styles are useful, and would prefer
 > keeping them the way they are.  If you feel strongly about this, I
 > suggest bringing it up as an issue for the WG to decide on at a future
 > telcon.

I'm bringing it up here, and I ask you -- as editor -- to make note of the issue.

 > > I suggest the critera to be that a counter style is actively used in
 > > print and in web documents, so common that people hardcode them into
 > > HTML to achieve the effects. Listing 10 different pages in the wild
 > > would be a requirement for starting discussions.
 > 
 > One problem with doing that is that I don't read anything other than
 > English, and thus don't have any way to do the research for this.

I'm not asking you to do it. The draft could ask reviewers to submit examples. 

 > > Was the other alternative presented? I suggest listing the glyphs
 > > needed for these languages from 1-100 so that we more easily can
 > > compare the alternatives.
 > 
 > Given that the alternatives were "define it up to 10k (or 100k) with a
 > simpler algorithm" vs "define it up to 10^16 with a more complex
 > algorithm", and the WG explicitly chose the larger and more
 > complicated of the two, I highly doubt that "define it up to 100"
 > would have gotten a significant number of votes.  ^_^

Goo the WG isn't voting, then :)

I'd still like to see the alternate representation -- what do the
algorithms in section 10.2 and 10.2 generate from 1-100? Can I ask you
to provide that?

There's an unresolved CVS conflict in the version of the document
found here:

  http://dev.w3.org/csswg/css3-lists/

Cheers,

-h&kon
              Håkon Wium Lie                          CTO °þe®ª
howcome@opera.com                  http://people.opera.com/howcome
Received on Wednesday, 22 June 2011 15:52:00 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:41 GMT