W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > May 2001

Issue http://www.w3.org/2000/03/rdf-tracking/#rdfms-abouteachprefix

From: Stephen Petschulat/CanWest/IBM <spetschu@ca.ibm.com>
Date: Tue, 29 May 2001 22:51:50 -0700
To: w3c-rdfcore-wg@w3.org
Message-ID: <OF199F85A0.D958E8FB-ON88256A5C.0012A9AB@mkm.can.ibm.com>
I missed some of the recent discussion on this issue in the ns-prefix
thread due to holidays, but as Martyn & others pointed out, it seems clear
that by and large people want this feature gone from the core:

http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0112.html
http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0123.html

For those not familiar with the issue, here is a quick summary:

aboutEachPrefix was introduced to deal with the case where a web site
developer may want to assert something about a (potentially large) number
of resources on their site such as "all the documents under the directory
http://www.mysite.com/secure/ require a password" where the URI is a
substring prefix. M&S uses the example of copyright:

<rdf:Description aboutEachPrefix="http://foo.org/doc">
    <s:Copyright> 1998, The Foo Organization</s:Copyright>
</rdf:Description>

This would pattern match http://foo.org/doc/doc1.pdf,
http://foo.org/doc/doc2.pdf, etc.

There are a number of obvious problems:
  - why just a prefix substring? what about port numbers messing up the
prefix match? why not full pattern matching? [1] [2]
  - aboutEachPrefix is described in the syntax, but not in the model [3]
  - it is difficult to implement [4] [5]
  - it creates problems for the formal model [6]

In addition, TimBL weighs in with "yes, abouteachprefix was a mistake in
retrospect." [7]

Is this an open & shut case or is there a group with a vested interest in
it (web developers?) that may want it to stay in? I don't know all of the
history of it, but perhaps some of the participants of the original WG can
shed some light on this. If no one speaks up in its defense, I assume
deprecation is the right route.

cheers,

- steve

[1] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Oct/0015.html
[2] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Sep/0003.html
[3] http://lists.w3.org/Archives/Public/www-rdf-interest/2001Apr/0349.html
[4] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Sep/0134.html
[5] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Feb/0109.html
[6]
http://lists.w3.org/Archives/Public/www-rdf-comments/2000JulSep/0034.html
[7] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Feb/0172.html

Stephen Petschulat
Common Architecture and Strategy Team
Center for IBM e-business Innovation::Vancouver
4611 Canada Way, Burnaby, BC, Canada
Phone: 1-604-297-3231    Fax: 1-604-297-3040
Received on Wednesday, 30 May 2001 01:56:54 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:35:53 EDT