W3C home > Mailing lists > Public > public-vocabs@w3.org > October 2012

Re: Meaning of property "url"

From: Dawson, Laura <Laura.Dawson@bowker.com>
Date: Tue, 23 Oct 2012 11:10:47 -0400
To: Adam Wood <adam.michael.wood@gmail.com>
CC: Michael Hopwood <michael@editeur.org>, Ed Summers <ehs@pobox.com>, Cord Wiljes <cwiljes@cit-ec.uni-bielefeld.de>, "public-vocabs@w3.org" <public-vocabs@w3.org>
Message-ID: <E2AE2E19-8C43-4DD5-BAB0-58B88384D6E2@bowker.com>
Text-based identifiers make me very nervous....

On Oct 23, 2012, at 11:08 AM, Adam Wood <adam.michael.wood@gmail.com<mailto:adam.michael.wood@gmail.com>> wrote:

A record in a database can refer to a person, and does so with a "key."

Think of a URI as a database key for the worlds most ridiculously
disorganized database (the internet).

If a hundred different articles all mention Bach, how does a computer
program know they all refer to the same Bach (Johann? Sebastian?
By agreeing that some URI (for the wikipedia page, for example) is the
key for a single guy, and then referring to that URI in your markup,
we can all know we're talking about the same thing.

So, no- there is no big philosophical issue with having URIs for people.

On Tue, Oct 23, 2012 at 10:00 AM, Michael Hopwood <michael@editeur.org<mailto:michael@editeur.org>> wrote:
"Any information that can be named... e.g... a person..." There may be some philosophical issues there; is a person "information"?

Sure, an antelope in a zoo may be a document (http://pages.gseis.ucla.edu/faculty/maack/BrietPrePress.htm) but isn't this taking it a little far? ;)

-----Original Message-----
From: ed.summers@gmail.com<mailto:ed.summers@gmail.com> [mailto:ed.summers@gmail.com<http://gmail.com>] On Behalf Of Ed Summers
Sent: 23 October 2012 15:43
To: Cord Wiljes
Cc: public-vocabs@w3.org<mailto:public-vocabs@w3.org>
Subject: Re: Meaning of property "url"

On Mon, Sep 10, 2012 at 7:27 AM, Cord Wiljes <cwiljes@cit-ec.uni-bielefeld.de<mailto:cwiljes@cit-ec.uni-bielefeld.de>> wrote:
Hi Michael,

(schema.org<http://schema.org>'s "url" property) means something very specific:
"this is the Web location of" _____ (where ______ is some network
addressable digital file).

That is what I thought, too. But the I wonder why:

"url" is a property of class "Thing" (instead of just class
"CreativeWork") there is no property "homepage" for class "Person" or
"website" for class "Organization"

The notion of a URI identifying a "file" is terribly antiquated. So many URIs identify resources that have representations (html) assembled on the fly as the result of queries to databases and whatnot. I prefer to think that url being a property of Thing was an intentional move, because the author chose to sidestep the
httpRange-14 issue, and let URLs identify any type of resource, as is the case in Roy Fielding's description of resource:

The key abstraction of information in REST is a resource. Any information that can be named can be a resource: a document or image, a temporal service (e.g. "today's weather in Los Angeles"), a collection of other resources, a non-virtual object (e.g. a person), and so on. [1] """


[1] http://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm#sec_5_2_1_1

Laura Dawson
Product Manager, Identifiers
Received on Tuesday, 23 October 2012 15:11:12 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:29:25 UTC