Multi-conference report: VO and SW

[Doug Tody]

In our case we are purposely using URIs to identify objects but not
locate them - if we want to locate a particular instance we use a URL.
A given URI may resolve into multiple URLs pointing to multiple instances.

For example, we use an "ivo:" URI as a location independent dataset
identifier.  A query can find actual instances of the data, in which case
we have resolved the reference to the level where URLs are appropriate.

 	- Doug



On Mon, 5 Dec 2005, Tony Linde wrote:
>> I would appreciate an elaboration of why it ivo: is bad. It
>
> I think it isn't Norman saying this but the W3C. See his section
> http://nxg.me.uk/note/2005/vo-and-sw/#dcc. They're simply saying that a
> proliferation of naming schemes is bad practice, presumably because only
> those in the know can interpret those naming schemes - they aren't globally
> interpretable.
>
> Personally, it doesn't worry me because a) our URI scheme generally follows
> the guidelines for naming; b) as is pointed out, plenty of others are doing
> the same; c) we also provide DC standard details so the metadata is globally
> available even if the data isn't; d) no one but an astronomer could possibly
> interpret our resources anyway and they'll all be 'in the know'.