[
Lists Home |
Date Index |
Thread Index
]
Simon St.Laurent wrote:
> Norm Walsh writes:
>>If identifying links in a vocabulary-neutral way is a requirement,
>>let's just fix XLink so that it can do that. Just about the very last
>>thing I want is to have to consult some quasi-structured text document
>>to find the links in my document.
>>
>>Surely that way lies madness...
>
> I dunno. I've found CSS syntax pretty agreeable stuff, not just
> "quasi-structured". It may be a different markup from pointy brackets,
> but it's hardly hieroglyphics.
>
> I vastly prefer its decorative approach to the transformative madness
> XSLT imposes for a wide variety of otherwise simple tasks...
CSS has a pleasant syntax for humans, but it's not always pleasant to
parse. I don't think anyone will be pleased to have to add a
full-fledged CSS parser to their code just to know which attributes
describe links. It's ok if you already have one, otherwise it's painful.
I haven't looked very closely at Opera's syntax (so I might get the
example below wrong, it doesn't matter much), but we probably could come
up with something simple in an XML syntax.
<linking>
<rule match='a'>
<set-link-source using='@href'/>
<use-link-source from='current'/>
</rule>
<rule match='img'>
<replace with='@src'/>
</rule>
...
</linking>
--
Robin Berjon <robin.berjon@expway.fr>
Research Engineer, Expway
|
