PubDIDs (and DIDs in general, maybe)

[Norman Gray]

Arnold, hello.

On 2014 Jan 15, at 15:29, Arnold Rots <arots at cfa.harvard.edu> wrote:

> The format is: ivo://ADS/<Publisher>#<InternalToPublisher>

> The <Publisher> tags are issued by the ADS.
> The data publishers provide the ADS with URLs to resolve the URIs
> and the fragment allows the publishers to organize their holdings in
> their own fashion.

The problem with this is that the RFC which defines URIs requires this #<internaltopublisher> part to be removed before the IVORN is dereferenced.  Thus the principal uses of such IVORNs violate the URI spec, and so probably shouldn't be encouraged.

> See, for instance:
> http://cxc.cfa.harvard.edu/cda/datasetid.html
> http://aas.org/data-sets/formatting-data-sets-aas-submissions
> http://aas.org/data-sets/participating-data-centers

The use of the '#' in the text at <http://aas.org/data-sets/formatting-data-sets-aas-submissions> doesn't violate this.  But as soon as a ivo:// is put in front of this, and it's called a URI, that fragment should be stripped before any network action.

ivo://ADS/<Publisher>%23<InternalToPublisher> would be OK, though....

Markus pointed to <http://www.ivoa.net/documents/Notes/URIFragments/index.html>, where there's more extensive discussion of this.

See you,

Norman


-- 
Norman Gray  :  http://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK