Wordnet URIs?
Ryan Shaw
Where do the Wordnet URIs come from? E.g. http://openguid.net/e69ec40b-da25-102b-9a03-2db401e887ec
says it is identical to http://wordnet.princeton.edu/synset/104148054,
but when I dereference the latter I don't get anything. Is it an
official Wordnet URI? Why not use a Wordnet Linked Data URI[1]
instead, so that it can be dereferenced?
Ryan
[1]http://www.w3.org/TR/wordnet-rdf/
ja...@openguid.net
A snippet from the roadmap says "WordNet associations are preserved by
means of a fictitious URI at princeton, from which the original sysnet
id can be gleaned".
I would very much like "Identical To"s to be dereferenced links.
However, I did not discover stable URIs for WordNet 3.0. The W3C
document does list version 2.0 URIs, which may be good enough, but
might not be complete. There are some notes at the bottom of that
spec (http://www.w3.org/TR/wordnet-rdf/#issues) indicating possible
ways of handling versions, none of which seem to be implemented.
Those notes suggest possibly complicated mappings between versions, so
I was hesitant to declare that 3.0 URIs map exactly to 2.0 URIs. Does
anyone familiar WordNet versioning know whether there are safe
mappings from 3.0 to the W3C URIs?
For now, I just wanted to save the synset id's for future linking, and
since it was fictional, I did not make them href's.
What do you think?
> Where do the Wordnet URIs come from? E.g.http://openguid.net/e69ec40b-da25-102b-9a03-2db401e887ec
> says it is identical tohttp://wordnet.princeton.edu/synset/104148054,
Ryan Shaw
> A snippet from the roadmap says "WordNet associations are preserved by
> means of a fictitious URI at princeton, from which the original sysnet
> id can be gleaned".
>
> I would very much like "Identical To"s to be dereferenced links.
> However, I did not discover stable URIs for WordNet 3.0. The W3C
> document does list version 2.0 URIs, which may be good enough, but
> might not be complete. There are some notes at the bottom of that
> spec (http://www.w3.org/TR/wordnet-rdf/#issues) indicating possible
> ways of handling versions, none of which seem to be implemented.
> Those notes suggest possibly complicated mappings between versions, so
> I was hesitant to declare that 3.0 URIs map exactly to 2.0 URIs. Does
> anyone familiar WordNet versioning know whether there are safe
> mappings from 3.0 to the W3C URIs?
>
> For now, I just wanted to save the synset id's for future linking, and
> since it was fictional, I did not make them href's.
>
> What do you think?
I think your solution makes sense for now; though if you're going to
mint URIs, maybe they should be in a namespace you control to prevent
confusion? Then it would be clear that these are URIs you are creating
and not ones Wordnet is providing.
Ideally the people doing the W3C Wordnet-as-Linked-Data stuff would
update to WN 3.0 and then you could use those.
Incidentally, I was surprised to see that Wordnet includes proper
names of people, something I wouldn't have discovered without your
great tool.
Cheers,
Ryan