Re: LANG + Metadata + unknown attributes

Paul Prescod (papresco@calum.csclub.uwaterloo.ca)
Sat, 08 Mar 1997 22:24:41 -0500


Message-ID: <33222D79.4F90@csclub.uwaterloo.ca>
Date: Sat, 08 Mar 1997 22:24:41 -0500
From: Paul Prescod <papresco@calum.csclub.uwaterloo.ca>
To: www-html <www-html@w3.org>, www-international <www-international@w3.org>
Subject: Re: LANG + Metadata + unknown attributes

Misha Wolf wrote:
> A similar problem arises with the use of Metadata.  At the DC-4 Metadata
> Workshop in Canberra (March 3-5), we agonised over a difficult choice:
> 
>    1.  Use a clean syntax for "qualified" (explained below) Metadata, even
>        though it would rely on a use of attributes not defined in HTML 2.0/3.2.
> 
>    2.  Use a dirty (difficult to parse) syntax, conformant with HTML 2.0/3.2.

The latter approach to meta-data seems doomed to failure in the long
run. Metadata can have structure. Metadata can nest. Metadata can even
recurse. Metadata structure should be validatable. Lobbying W3C and the
browser vendors to support each metadata idea you come up with is  going
to be a painful process. 

Even your "qualified" metadata seems "dirty" to me, in the sense of
difficult to parse. Wouldn't this be better:

<META-BLOCK LANG="Swahili" SCHEME="XYZ" SYSTEM="DC">
    <SUBJECT>...</SUBJECT>
    <DATE DAY=03 MONTH=02 YEAR=1997>
</META-BLOCK>
<META-BLOCK LANG="French" SCHEME="QRS" SYSTEM="TEI">
    <FILEDESC>...</FILEDESC>
....
</META-BLOCK>

I propose, instead of lobbying for a *particular* extension to metadata
today, and another particular extension next year, and another
particular extension the year after that, you should just lobby for the
right to <LINK> to MetaData which can then be an XML document broken
into META-BLOCKs where each META-BLOCK has arbitrarily complex XML
structure.

 Paul Prescod