W3C home > Mailing lists > Public > public-rdfa@w3.org > April 2010

Rename @prefix

From: Doug Schepers <schepers@w3.org>
Date: Fri, 23 Apr 2010 14:44:35 -0400
Message-ID: <4BD1EA93.80702@w3.org>
To: public-rdfa@w3.org
Hi, folks-

I raised this once before in the public-rdf-in-xhtml-tf mailing list 
[1], but I haven't seen any follow-up, and I wondered if there had been 
any discussion.

Quite apart from any issues about whether the 'prefix' attribute is the 
right mechanism, it is a confusing and unintuitive attribute name.  It 
describes a syntactic convention, not the underlying concept which is 
its true function.

Consider this hypothetical alternate way of establishing a namespace 
scope (in ad-hoc pseudo-code):

with: "http://example.org/schema/"
   property: type; content: "definition";
   property: heading; content: "abstract";
   property: body; content: "generalized by reducing the information 
content of a concept or an observable phenomenon";

In this syntax, I've established that the 'type', 'heading', and 'body' 
properties are to be evaluated in the context of the example.org schema 
resource.  This is essentially what CURIEs, namespaces, etc. do, using 
the syntax 'prefix:subject|object|predicate'.

My point is that 'prefix' is a term that only represents how a 
particular syntactic convention represents the data, and is too far 
abstracted from the *meaning* of the mechanism: to point to a resource 
in which the terms in question are evaluated, and which distinguishes 
them from the same term (as a string) used in another context. This 
makes it harder to explain or to learn, needlessly.

My suggestions for a more accurate attribute name are:

* 'ontology'
* 'context'
* 'scope'
* 'reference'
* 'model'

... or anything similar.


-Doug Schepers
W3C Team Contact, SVG and WebApps WGs
Received on Friday, 23 April 2010 18:44:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:15:06 UTC