RE: XBL 2 and xml:id

On Wed, 5 Jul 2006, Doug Schepers wrote:
> Ian Hickson wrote:
> |  
> | It requires authors to understand two specs instead of one.
> <snip />
> | It requires authors to understand why they can use xml:id on 
> | XBL but not  on HTML, or, if implementations support 
> | both in HTML, why in HTML they can have two IDs but they  
> | can only have one in XBL.
> 
> The 2 points above (non-exhaustively) exemplify the argument that we can 
> somehow sheild the author from learning about namespaces. Whether or not 
> namespaces are the ideal mechanism for disambiguating nodes, they are a 
> reality of XML today, and I can't see them going away soon.

If we want XML to be used "on the Web" (as opposed to on the server-side 
or in specific vertical markets), they are going to have to either go away 
or be dramatically simplified.


> XBL2 is about binding multiple languages together.  I would be very 
> surprised if the authors who are going to be using XBL2 will struggle 
> with the concept of namespaces.

Mozilla has 7 years of experience with XBL, and even in that community, 
which is not exactly the "amateur Web developer" community, developers 
have had problems with XML namespaces in an XBL context. I don't see why 
this would be any less of a problem with XBL2, especially considering that 
the net effect of making XBL more widely available would be to lower the 
average technical level of the authors using XBL.


> If they don't already know about them, and how they are used, they are 
> certainly going to have to learn them in order to effectively author 
> XBL2 documents.  The idea that you can use XBL without dealing with 
> multiple specifications and languages is an oxymoron.

While that is currently true (though not necessarily true going forward, 
if we can simplify the namespace syntax in XML), it is not a reason for 
making it even more complicated (with "xml:id").


> Let's not underestimate authors in what they know or what they are able 
> to learn.

I'm not trying to estimate anything; I'm only looking at the available 
evidence (XBL experience in the Mozilla community, the experience reported 
by members of the XForms community, the evidence visible in the majority 
of files intended to be SVG files, the many documents labelled as "XHTML" 
on the Web, etc) and trying to build new technologies with that evidence 
as a guide.


> XBL2 is rather advanced stuff, and isn't going to be used by people 
> posting poodle pictures, but rather by those making advanced sites that 
> need maximum potential in their language.

I wouldn't be surprised if XBL2, once (and if) widely deployed, was used 
more for reordering content HTML from CSS than for any "advanced" use. And 
that doesn't actually require any namespaces in the XBL file other than 
XBL, and doesn't require any prefixes at all. (Prefixes are what authors 
have the most trouble with, based on the aforementioned evidence.)

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Wednesday, 5 July 2006 08:30:20 UTC