Re: argument from authority considered pointless

At 12:06 AM 1/23/03 +0000, Miles Sabin wrote:

>Tim Bray wrote,
> > This may be a fair statement of where the angst is coming from.
> > Let's assume, as a thought experiment, that it is.  So, do we fix the
> > definition of URIs or do we fix the specification of RDF?
> >
> > Since the vast majority of software out there actually does implement
> > URI semantics per RFC2396 (Roy is right; look at a browser or robot
> > source code any time), the cost of changing that definition is higher
> > than the cost of changing RDF, whose deployment is at a much earlier
> > point in the historical curve.
>
>Is it tho'?
>
>RDF doesn't need any changes to URI _syntax_. Nor does it need any
>changes in network protocols, clients or server. All it needs is for
>some bogus metaphysics to be ditched.

I agree with your first sentence, but I must refute that URIs somehow don't 
meet the requirements of RDF.

(Not aimed specifically at you, Miles, but your comment here is a trigger 
for my response.)

1. While it's easy to get into lengthy metaphysical arguments about what 
URIs may denote, the formal semantics of RDF simply doesn't care.  It 
recognizes that, in general, there are many (infinite, even) 
interpretations that can satisfy an RDF graph.  Usually, only one of those 
interpretations is intended, but that doesn't matter to the formal aspects 
of RDF, which provide some ground rules for saying what can be validly 
inferred from a given RDF graph.  So no problem on that score.

2. I've noticed that there may be some problems in building applications 
that use RDF, but I'm not aware that any have hinged on the exact nature of 
a resource;  applications (including rules used with the likes of cwm) 
embed enough knowledge about the resources concerned to do their job.  Of 
course, when different applications need to exchange information, there is 
an obvious need to ensure the URIs are being understood consistently.  The 
RDF application community that I'm aware of is taking a test-case driven 
approach to the evolution of URI vocabularies for shared purposes (e.g. 
calendaring), so that establishing such common understanding and the means 
of testing it is built into the process.  The formal aspects of RDF give us 
some tools to lock down some of that common understanding, and I fully 
expect to see increasing use of such means.

So maybe you're right about ditching the metaphysics, as long as you don't 
mean replacing it with some other controversial view.

#g


-------------------
Graham Klyne
<GK@NineByNine.org>

Received on Thursday, 23 January 2003 06:42:04 UTC