W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > February 2002

Subtler simplified datatypes.

From: Pat Hayes <phayes@ai.uwf.edu>
Date: Wed, 20 Feb 2002 16:29:17 -0600
Message-Id: <p05101436b899d0665dba@[65.212.118.219]>
To: w3c-rdfcore-wg@w3.org
Ok, guys. Another version now at
  http://www.coginst.uwf.edu/users/phayes/simpledatatype1.html
This changes the previous one as follows.

1. S-B idiom now works with rdfs:drange; in fact, the S-A and S-B 
idioms are synonymous and can be interchanged with each other.

In some ways, this makes it all even simpler. We can tell people: 
Look, you can just use literals as objects and rdfs:dtype your 
properties, and as long as you don't make any obvious booboos like 
assigning clashing types to the same property, you will be OK. (What 
you always wanted, right?)

If you want to get fussy about it, then 'stretch' each triple into 
two connected by a bnode, using rdfs:dlex to keep your lexical forms 
and your values separate; and BTW, this same trick also lets you 
attach the datatype name directly to the literal if you want to get 
very 'local'. Oh, and be a *bit* careful because that drange stuff 
still works, so now you *can* shoot yourself in the foot.

2. Cost is that the untyped S-B idiom now does NOT say that the 
object is the string. It can be used that way, but that's not 
strictly a valid assumption, so if you base any conclusions on it, 
you may come unglued if datatype information is added later.

  Still, I bet the people who want to use it with simple 
string-matching (eg Dublin Core) aren't going to be making (m)any 
inferences that would get upset by later datatyping, anyway, so they 
aren't going to care. Still, we do have a slight dirty nonmonotonic 
secret here now.

3. Although I didn't go into this in the above, you could still say
rdfs:range rdfs:subPropertyOf rdfs:drange .
and then use simple rdfs:range to attach datatypes, if (a) you want 
to and (b) you are willing to take the risks. But without dtype, the 
risks are far less than they were, since the local scopes are now 
very local indeed.

I'll send answers to my quiz from both versions later today.

Pat
-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes
Received on Wednesday, 20 February 2002 17:29:19 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:45:18 EDT