Re: "placeholder link"

On 25 Jun 2007, at 0951, Sander Tekelenburg wrote:

> At 16:09 +0200 UTC, on 2007-06-25, Simon Pieters wrote:
>
>> We can define the semantics for elements.
>
> Right. But let's be honest, when you need to read someone else's  
> HTML, seeing
> <a> doesn't exactly immediately make you realise that it means <span>.

That's beside the point as <a> does not mean the same thing as  
<span>. <a> is intended to mean: "this would be a link, but there is  
nothing to point it at". The fact that it is treated similarly to  
<span> for styling purposes is a coincidence.

The current draft of the spec provides an example of where this would  
be useful. The reference links currently have gibberish in their  
@href because the section they point to does not yet exist. Where a  
link placeholder element available, the UA would not lie to the user  
(implying a non-existent resource exists) and the HTML source would  
still indicate that the references are intended to be links.


> HTML is not for UA consumption only.

True. But I don't see how defining <a> as a "link placeholder" would  
harm a human's comprehension of the document.


> [...]
>
>>>>
>>>> and the stylesheet rules will be simpler.
>>>
>>> How?
>>
>>     menu a { display:block; border:solid; }
>>     menu a:link { background:lime; }
>
> menu * { display:block; border:solid; }
> menu a:link { background:lime; }

It seems to me that 'menu a' would be a clearer statement of the  
author's intent than 'menu *', at least to a human reader of the CSS.


Scott.