W3C home > Mailing lists > Public > public-powderwg@w3.org > November 2008

Re: why "describedby" instead of "meta"?

From: Jonathan Rees <jar@creativecommons.org>
Date: Mon, 24 Nov 2008 11:47:47 -0500
Cc: Public POWDER <public-powderwg@w3.org>
Message-Id: <711BB51A-B6B3-4581-9E20-23BE09DFF3FE@creativecommons.org>
To: Phil Archer <phil@philarcher.org>

Thanks. Actually I don't think you should use "meta" since RDFa (which  
has the only normative definition of "meta" I could find) defines the  
target to be metadata, and metadata is data about data, not data about  
things generally. My house's deed could have metadata, but my house  
couldn't since my house isn't data. I'm fine with "describedby".

Jonathan

On Nov 24, 2008, at 10:55 AM, Phil Archer wrote:

> Jonathan,
>
> I hope you don't mind me copying my reply to the public list but  
> Julian Reschke just asked me about the same thing so I'd like to  
> record the answer.
>
> In our earlier drafts we were going to use rel="powder" with the  
> existing XML and RDF MIME types for POWDER and POWDER-S  
> respectively. However, Mark Nottingham [1] advised that a better way  
> would be to define a more generic relationship type but define  
> specific MIME types - hence one of my jobs this week is to register  
> text/powder+xml and application/powder-s+xml.
>
> Could we use meta? Yes, we could, and we came close to doing so.  
> There are no strong feelings in the group but there was a very  
> slight edge in favour of describedby as meta _appears_ to be  
> associated specifically with RDF. That is, when building application  
> to handle rel="meta" /might/ assume that the target of such a link  
> is RDF. This is wrong - it could be anything - but that might be an  
> assumption in the real world. Equally, it would be wrong to say that  
> describedby will always point to POWDER, that isn't our intention,  
> but, I guess, it is a nod to practicality over theory.
>
> We do actually refer to rel="meta" in our documentation where we  
> discuss a resource linking to a POWDER processor from which RDF  
> triples may be returned to describe that resource [2]. If there were  
> strong feeling that we should use rel="meta" then I don't think we'd  
> put up much resistance (but the request to register describedby has  
> been made to  IANA!)
>
> Meanwhile - a reply to your comments on the formal doc has been  
> prepared and should be posted in the very near future.
>
> Cheers
>
> Phil.
>
>
> [1] http://lists.w3.org/Archives/Public/public-powderwg/2008Aug/0000.html 
>  onwards
> [2] See just below example 3-2 http://www.w3.org/TR/2008/WD-powder-dr-20081114/#eg3-2
>
>
> Jonathan Rees wrote:
>> I just found a definition of "meta", in
>> http://www.w3.org/1999/xhtml/vocab/ which has a date of 2008-10-21.
>> Repeated in http://www.w3.org/TR/rdfa-syntax/.
>> Not sure how it differs from "describedby". What do you think?
>>
>> These documents also give yet another registry of link types... not
>> that I'm collecting them, but I wasn't really aware of it before.
>>
>> Jonathan
>>
>>
>
>
> -- 
>
> Phil Archer
> w. http://philarcher.org/
>
Received on Monday, 24 November 2008 16:48:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:42:13 GMT