Re: My task from last week: Semantic free identifiers

On Mon, 20 Jun 2011 12:08:43 -0700, Andrea Splendiani  
<andrea.splendiani@bbsrc.ac.uk> wrote:


> - in a continuum between web and semantic web, perhaps IDs are not only
> intended to be 'understood' by machines.
>
> Again, I understand the reason for them. But is it worth the reduced
> intuitiveness ? Or the added complexity to retain a bit of it ?


I couldn't disagree more :-)

I tend to err on the side of doing "the right thing", and ensuring that  
the tooling is there to support "the right thing"... By "right thing" I  
mean that I'm sure Hungarian semantic-webbers would have quite something  
to say about a decision to make the URI "partOf" rather than "A_0001" +  
multi-lingual labels.  It's a bit selfish of us English-speakers to create  
global infrastructures just for ourselves... na?

(though I guess, for them, "partOf" *is* opaque... so...??  Perhaps that  
argument is somewhat spurious??)

Regardless, just as browser bookmarks were created so that we humans  
wouldn't have to remember/type/read URIs, there is no good reason that we  
humans should ever have to read RDF-XML... and if you are expert enough to  
*have* to read it, then you should probably be sophisticated enough to  
deal with opaque identifiers (preferably using appropriate tools ;-) ).   
If we're having trouble constructing SPARQL queries using opaque  
identifiers, lets not solve the problem by building a  
"philosophically/technically-incorrect" global architecture just for the  
sake of convenience, lets fix it at the level of the SPARQL query writer.

$0.02  <-- mark:partOfMine



Mark

Received on Monday, 20 June 2011 19:28:49 UTC