[uf-discuss] Re: [hcite] indentifier

[Ryan Cannon]

On Feb 22, 2007, at 9:03 AM, Brian Suda wrote:
> KEYS for things like BibTeX are not globally unique, but unique
> only to the bibtext file. This can be generated by any transforming
> application and are not needed in the mark-up. But KEYS are not the
> same as IDENTIFIERS. Identifiers should be human-readable because
> these are used to uniquely identify for find the object in a physical
> world. Things like key-value-pairs are machines use only and shouldn't
> be part of hCite.

We're agreeing, just coming at it from different directions. Perhaps we
should change the description of identifier in the straw man to:

"An identifier used to used to uniquely identify for find the resource.
The identifier SHOULD be a URN, but may also be the identifier assigned
by a specific resource, such as pubmed ID number or Library of Congress
call number. Identifiers that are not URN's should use the TYPE
subproperty to disambiguate."

I'm not sure we want to get into the morass of defining values for TYPE,
but if an identifier is for a specific resource, we should allow a way
to tell which resource it's from.

Examples:

     <a class="identifier url title" href="http://www.free- 
culture.cc/">Free Culture</a>

     ISBN: <abbr class="identifier" title="urn:isbn: 
1594200068">1594200068</abbr>

     <span class="identifier">
       <span class="type">Library of Congress</span>:
       <span class="value">KF 2979 .L47 2004</span>
     </span>

-- 
Ryan

http://RyanCannon.com