Re: List styles - a wish list item

On Wed, 3 Jun 1998, Grzegorz Staniak wrote:

> Hello to all,
> 
> I noticed this "list-style wish-list" thread and I would like to ask for
> your opinion about an idea brought forward by some people on a Polish
> newsgroup, namely, a property that would enforce alphabetical order on
> unordered lists. The idea is to make things easier for the author, who
> would be able to just prepare a list, and then have it rendered in order
> by entering a CSS rule into a stylesheet. Possibly, the order could be
> reversed, too.  Obvious difficulties with this approach include the need
> to store (or retrieve from the system - are there such resources
> available in operating systems?) tables/arrays specifying the
> alphabetical order for a large number of languages; delays in document
> rendering, especially for long lists etc. I'm not sure CSS is the best
> place for this kind of functionality, since it's more a HTML tree
> transformation than application of style info. Perhaps browsers could
> offer a "sort lists" switch somewhere in their configuration. 
> 
> What is your opinion? Is implementation of such a feature faesible, and
> if it is, is it worth the effort? Myself I think a utility (external or
> built in an editing environment) is a better idea, if only because of
> the unevitable delays in the parsing and rendering of HTML documents,
> but I lack experience necessary to assess the value of such an idea. 

This is really better done with scripting, rather than CSS. Given that the
HTML character set is now Unicode, it would be very difficult to provide
browers with general rules for lexicographical ordering. Having authors
supply their own algorithms means they can not only concentrate on just
that subset of Unicode which they intend to use, but also specify exactly
how they want case, accents, punctuation marks, etc. to affect the
ordering. As someone who is in the middle of translating a fairly simple,
intuitive ordering algorithm over ISO Latin-1 from hand waving into
computer code I know just how important (and difficult) these things are. 

Tim Bagot

Received on Wednesday, 3 June 1998 13:37:29 UTC