[whatwg] comments

dolphinling dolphinling at myrealbox.com
Sat Jan 1 08:54:44 PST 2005


Anne van Kesteren wrote:
> dolphinling wrote:
> 
>> Also, in the repetition model, I _REALLY_ don't like the repeat 
>> attribute applying to arbitrary non-form-related elements. It seems 
>> incredibly hackish to me--why should trs, ps, lis, etc. get a new 
>> attribute just because forms have been updated? I'd much prefer that
>> the new attribute only apply to fieldsets, and those can be wrapped
>> around the stuff you want repeated (which is actually correct
>> semantically, too).
> 
> 
> That complicates it actually. For a lot of forms it is far more
> semantically correct to use multiple table rows. If you need to use
> FIELDSET with 'display:table-row' and some other CSS tricks which will
> not even work because you can not style FIELDSET properly this feature
> can better be dropped.
> 
> However, the feature is very useful and having an attribute that applies
> to multiple elements is not that bad and can actually be quite useful to
> achieve semantically correct results.

Actually, I meant something like

<fieldset repeat='template' repeat-start='blah'>
   <tr>...</tr>
   <tr>...</tr>
</fieldset>

...which I see now is invalid. I wish it weren't, though (even without 
templates I think fieldset should be able to go around table rows, or 
pretty much wherever else it wants), and I don't think there are many 
(if any) cases where it would be more semantically correct to not have 
fieldsets. The only ones I can think of are where there's only one input 
that's being repeated, but that's just semantically redundant, not wrong.

-- 
dolphinling
<http://livejournal.com/users/dolphinling>
<http://dolphinling.net> coming soon…


More information about the whatwg-whatwg.org mailing list