W3C home > Mailing lists > Public > whatwg@whatwg.org > April 2005

[whatwg] [WA1] The profile Attribute

From: Ian Hickson <ian@hixie.ch>
Date: Sun, 17 Apr 2005 13:22:02 +0000 (UTC)
Message-ID: <Pine.LNX.4.61.0504171317540.20636@dhalsim.dreamhost.com>
On Sun, 17 Apr 2005, James Graham wrote:
> > > > 
> > > > The only way I can see to avoid this is to use only one profile, 
> > > > since then you can't ever get clashes.
> > > 
> > > There are other ways I've seen proposed, such as using namespaces:
> > > 
> > > http://www.protogenius.com/rel-schemas/draft-scheid-rel-schemas-00.htm
> > 
> > Namespaces are not an option. Authors simply don't understand them.
> 
> Respectfully, I think namespaces are the only sensible solution here and in
> other situations where the document is mixing semantics from multiple sources.

The recent microformats trend (using profile="") is one other solution, 
which seems to be at least as sensible.


> What's the evidence that authors don't understand namespaces?

One source for example is Micah Dubinko's statistic that 90% of all the 
queries about XForms that he receives are asking for him to explain 
namespaces. [1]

  [1] http://www.w3.org/2004/04/webapps-cdf-ws/papers/verity.html

This certainly has also been my experience in dealing with Web authors who 
are trying to use languages that rely on namespaces.


> Does it all come from XML namespaces (which are more complex than 
> anything we would need for this type of problem*)? In any case I think 
> this is a situation where, with sensible defaults, we can provide a 
> useful feature that will be well within the grasp of the small subset of 
> authors who actually want to use it.

By "namespaces" here I am refering to XML namespaces and similar solutions 
that require declaring a prefix and then using that prefix elsewhere.


> For example we could define <profile name="foo" 
> href="http://example.com/profiles/#foo" /> and then require a profile 
> attribute for elements with rel values not assosiated with the default 
> profile, which would be given by the value of the profile attribute in 
> <head> or the last <profile> element with no <name> value. That seems 
> much simpler than XML namespaces.

That seems a lot more complicated than the current proposed solution with 
profile="".

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Sunday, 17 April 2005 06:22:02 UTC

This archive was generated by hypermail 2.3.1 : Monday, 13 April 2015 23:08:22 UTC