[whatwg] reversed lists

[Ian Hickson]

On Tue, 26 Feb 2008, Tab Atkins Jr. wrote:
> 
> As noted by Jonas Sicking, reverse-ordered lists will require a browser 
> to read the entire list in before numbering, or else update on the fly 
> (not acceptable).  However, as Ian notes, this isn't a problem with 
> variable-width tables.  We accept that certain classes of tables can't 
> be displayed until the entire thing has been read and computed, and we 
> will just have to accept that with reverse-ordered lists as well.
> 
> The exception would be if we adopted the rule, suggested by Simon 
> Pieters, that the start= attribute apply to the first *lexical* element 
> in the list, rather than the first *ordinal* element.  This would allow 
> browsers to render reversed lists immediately when it is present.  I 
> like this compromise.

This is basically what the spec does now, I think.


> > ::stuff about step= attribute::
> 
> I can't think of any use cases for a step= attribute currently, at least 
> none that wouldn't be served best by *arbitrary* number generation.  
> Frex, numbering a list with the successive squares or primes.  While 
> fancy, these are just cute tricks, and not actually generally useful as 
> far as I can tell.  The same would be true for the step= attribute.

Agreed; step="" doesn't seem necessary yet.


On Wed, 23 Apr 2008, Christoph Päper wrote:
> Ian Hickson schrieb:
> > On Fri, 25 Jan 2008, Christoph P�per wrote:
> > > 
> > > I think it has been shown, that the meta attribute |reverse| would 
> > > not work in HTML, it would have to be a "command" attribute, i.e. it 
> > > doesn't describe the ordering of the following list items, but their 
> > > indended display. This would make it presentational and thereby not 
> > > suitable for HTML. It would belong into CSS, but that isn't very 
> > > good at reordering boxes.
> > 
> > I don't really follow. What's wrong with how the spec works now?
> 
> Without rereading or much rethinking the thread, the current spec is 
> right in that |reversed| describes the actual order of |li|s -- which 
> is, what markup should do --, but this doesn't degrade well and it's not 
> incremental, because you need to know the number of |ol|'s children 
> (which you could hardcode with |start|) in advance to number the first 
> item. Therefore someone proposed a command-like |reverse| (no 
> participle) attribute that would keep the numbers, but reorder the |li|s 
> with them, which is backwards-compatible, but works just as bad for 
> incremental rendering (though in a different way) and is not very 
> markupish and -- if at all -- should be done on the styling level.
> 
>   Logical markup order              Presentational markup order
> 
>   <ol><!--spec, compat-->           <ol><!--messy-->
>     <li>First        1. First         <li>Third        1. Third
>     <li>Second       2. Second        <li>Second       2. Second
>     <li>Third        3. Third         <li>First        3. First
>   </ol>                             </ol>
> 
>   <ol reversed><!--messy-->         <ol reversed><!--spec-->
>     <li>First        3. First         <li>Third        3. Third
>     <li>Second       2. Second        <li>Second       2. Second
>     <li>Third        1. Third         <li>First        1. First
>   </ol>                             </ol>
> 
>   <ol reverse><!--a proposal-->     <ol reverse><!--messy-->
>     <li>First        3. Third         <li>Third        3. First
>     <li>Second       2. Second        <li>Second       2. Second
>     <li>Third        1. First         <li>First        1. Third
>   </ol>                             </ol>

Reversing the rendered order seems like it would cause more trouble than 
reversing the numbers, so I think we should stick with what we have now.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'