- From: Pat Hayes <phayes@ai.uwf.edu>
- Date: Fri, 5 Apr 2002 12:42:12 -0600
- To: "Seth Russell" <seth@robustai.net>
- Cc: <www-rdf-comments@w3.org>
>Pat Hayes, > >I would like an authoritative interpretation of what the MT is telling us >about blank nodes. In particular if an rdf process reads a bnode from >source A, and copies that node to a source B, but adds a uriref to the node >(making it not blank), has the process changed the meaning of the node ? Not clear what 'changing the meaning' means. Has the process ADDED meaning? Yes, in the precise sense that its 'conclusion' (output) is consistent with, but not entailed by, its input. So it has not changed the meaning in the sense that it has denied what was input to it, but it has indeed altered the meaning in some precise sense. It is like the difference between being told that someone has a ball and concluding that Joe has a ball. >For example: > >Source A reads: >_:a rdf:type :ball >_:a :hasColor :blue. >_:a :hasShape :round. >_:a :ownedBy :Seth. > >Source B reads: ><uuid:1615> rdf:type :ball. ><uuid:1615> :hasColor :blue. ><uuid:1615> :hasShape :round. ><uuid:1615> :ownedBy :Seth. > >Is this tantamount to the process A reading "Seth owns some (maybe only >one) round blue balls." and copying to B "Seth owns a round blue ball."? No, the best way to render that output would be "Seth owns a round blue ball called '<uuid:1615>' ". Now, as long as everyone knows that '<uuid:1615>"'isn't *really* a name, then this indeed hardly amounts to saying more than was input to it. But how does one know, in general, which urirefs are supposed to be real names and which are only these, er, 'blank' names that only give existential import? Once the name is created and set loose upon the world, as it were, it loses all trace of its origin, and is just another uriref. And urirefs are names, on the web, right? Globally unique and supposed to last for ever. So you just created, for all time, and for planet-wide use, a permanent Name for this thing that nobody has a name for. If anyone else ever uses that name for anything else, global confusion could follow, so that uuid has now been used up, for ever. It can never be used by anyone else anywhere on the Web, ever again. I would say to the process that did that, are you SURE you want to publish that name? I wouldn't do it, myself. 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 Friday, 5 April 2002 13:42:12 UTC