Re: making statements on the semantic web

Greg Tyrelle wrote:
> http://purl.uniprot.org/uniprot/P00750 is a permanent URL for a
> uniprot database entry. A database entry is an information resource,
> hence http://purl.uniprot.org/uniprot/P00750 should not return 303,
> but instead the database record. Assuming for a moment you're using a
> browser that  would be the HTML representation.
 >
 > Or are the semantics of http-range14 going to break PURL servers,
 > which are using 303 as a simple redirect ? In other words, 303 in
 > http-range14 suggests that 'see other' means see this for data *about*
 > the non-information resource being identified. Rather than get the
 > information resource *itself* being identified at its current
 > (possibly temporary) URL ?

 From a practical point of view, it is conceivable that some people would 
want to make statements about a web page or another specific representation 
of this "database entry", so we need a different URI for each:

http://purl.uniprot.org/uniprot/P00750 <- "database entry"
http://beta.uniprot.org/uniprot/P00750 <- an HTML representation
http://beta.uniprot.org/uniprot/P00750.rdf <- an RDF representation
http://beta.uniprot.org/uniprot/P00750.fasta <- a FASTA representation

Now we could have a lot of fun arguing over what exactly this "database 
entry" is, whether or not it corresponds to someone's concept of "protein", 
and more generally, what any of this is meant to reflect in the "real" 
world. But I don't see any point in such a discussion -- not unless someone 
can demonstrate that there are practical implications for any of this.

Received on Monday, 27 August 2007 16:59:24 UTC