|
Posted by Jukka K. Korpela on 11/02/06 22:01
Scripsit Joel Shepherd:
> "Jukka K. Korpela" <jkorpela@cs.tut.fi> wrote:
>
>> Seriously, you don't need to know about these [rel] attributes,
>> since they are virtually unsupported by browsers and vaguely
>> defined, except for rel="stylesheet" in <link>, and any CSS tutorial
>> should explain that part.
>
> It might be more truthful to say that rel attributes are virtually
> unsupported by _websites_.
To some extent, this is a chicken and egg problem: few people want to use
attributes that have no useful impact on anything; and few browser vendors
are willing to implement support to attributes that are not used by authors.
But basically, the browser vendors haven't even figured out how those
attributes _should_ be supported. And I don't blame them much for _this_.
What should a browser do when it encounters
<a rel="mun kiva linkkini" href="foo">bar</a>
or something similar? For a long time, there was no authoritative
_definition_ of the _meanings_ of rel (and rev) values. I would still say
so, since the current excuse for a specification of HTML semantics (HTML
4.01, that is) is particularly vague when it comes to those values. Most of
the values have little use in <a> elements, and regarding the use of <link>
elements for navigation, it was a great idea in the middle 1990s but it's a
lost cause: everyone uses <a> elements for navigation (except the misguided
souls that use navigation based on non-HTML approaches), so <link> is just
pointless and potentially confusion duplication now.
> Several more sophisticated browsers have
> supported them just fine for quite a while: Opera and Mozilla variants
> are two that I've used personally.
I cannot see what you see as interesting in "supporting" something that is
so vaguely defined. Surely you _can_ make a browser display a rel attribute
value on mouseover or to show all links with rel="cool" as blinking, but
seriously, what is the point?
--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
Navigation:
[Reply to this message]
|