This is the mail archive of the docbook@lists.oasis-open.org mailing list for the DocBook project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

RE: Markup for personal names


One of the more common schemes for exchange of information about persons is
vCard:

  http://www.imc.org/pdi/

There are a few proposals for representing vCard in XML:

  http://www.vcard-xml.org/
  http://www.w3.org/TR/vcard-rdf

but none with a 'standard' status, AFAIK. 


If someone wanted to investigate this avenue (directories), please also keep
an eye open for LDAP schema. The basic idea is that all objects are
identified by registered OIDs, so that applications can exchange e.g.
address information in an unambigous way.

  http://www.kingsmountain.com/ldapRoadmap.shtml
  http://www.dsml.org/
  http://www.alvestrand.no/objectid/ 


Another interesting source of schemes for a variety of concepts and entities
is STEP, ISO 10303 - Industrial automation systems and integration,
http://www.nist.gov/sc4/. There's an elaborate model for the concept of
organizations, persons, and the relations between them. For an example of a
binding to XML, see

  http://www.stepml.org/pid_class/index.html#poadd


Yet another scheme for information about persons that might be interesting
wrt. publishing is XrML, http://www.xrml.org/. This is "a universal method
for specifying rights and issuing conditions associated with the use and
protection of content", developed by Xerox spin-off ContentGuard. There's a
rather detailed modelling of persons, other legal entities, and how they
relate to content. XrML might appear a bit overkill for many purposes, but
if you need to aggregate content from a lot of different sorces and track
useage and rights, you'll need it.


And then of course, there's the granddaddy of bibliographic metadata, MARC.
There are some efforts to represent MARC in XML, see e.g. 

  http://xmlmarc.stanford.edu/.


I'd better stop now ...

But just in case someone is going to invent yet another set of metadata for
representing persons and how they relate to published work, the best overall
description of intellectual property (and commerce) that I know of is
<indecs>, 

  http://www.indecs.org/pdf/framework.pdf. 


Kind regards
Peter Ring


-----Original Message-----
From: Jeff Iezzi [mailto:jeff.iezzi@semanticedge.com]
Sent: Monday, June 25, 2001 10:28 AM
To: docbook@lists.oasis-open.org
Subject: RE: DOCBOOK: Markup for personal names


> On 22 Jun 2001, at 17:54, Peter Ring wrote:
> 
> While Dublin Core might often be appropriate for *delivery* of common
> metadata, it is not really intended for basic markup in your repository.
  
Gee, I don't know. Is it possible to assure the right intention for 
everything? Some persons use a baseball bat as a form of 
protection; I prefer to take a baseball bat to the batting cages. 
Some people write documents with DocBook; others use it to 
exchange documents. :>) 
 
There are some DTDs which quite effectively use DC elements to provide
metadata (http://www.ontoknowledge.org/oil/dtd/).  
My point was merely whether it is necessary to have a rich 
semantic (as currently in DocBook) or whether a simple semantic 
is sufficient. As well described in the following article you 
mentioned, there are ways to extend DC to make it more rich.

> To learn more about the evolution and scope of the Dublin Core, see
> http://www.dlib.org/dlib/january01/lagoze/01lagoze.html.
 
While I do not suggest this approach, I believe most processing 
applications do not require a highly-defined level of semantic for 
publishing metadata or for doing intelligent searches with such 
content. In the case of marking up first names and surnames, I add 
a transformation step to DocBook documents to dumb the name-
related information down into a DC-compliant format, which is more 
simple to search than a highly-refined semantic. There's only one 
"Jeff Iezzi" a billion "Jeff"s and few less "Iezzi"s within a certain 
context. Replace my name with "Michael Smith" and the results 
are similar once you express a relation to the "Michael Smith" of 
xml-doc. :>)
 
To conclude, if you consider DocBook an interchange DTD, it might 
make sense to publish the metadata encoded in a DocBook 
document in a format that can be made available to metadata 
handling systems.
 
Just trying to contribute ...


[:>|

Jeff Iezzi
jeff.iezzi@semanticedge.com

-- EOF --


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]