DataCite relationships [was: VOResource 1.1 RFC]

[Markus Demleitner]

Hi,

On Thu, Jun 08, 2017 at 01:58:46PM +0000, Sarah Weissman wrote:
> Replying to my text
>> 
>> Not really; the main issue was how IVOIDs would go into DataCite
>> documents.  The current resolution is to use DataCite's URI type
>> rather than try to get a special relatedIdentifierType; I'm not 100%
>
> DataCite 4.0 (http://schema.datacite.org/meta/kernel-4.0/)
> doesn???t have a generic URI type for relatedIdentifierType. The
> allowed values are: ARK arXiv bibcode DOI EAN13 EISSN Handle IGSN
> ISBN ISSN ISTC LISSN LSID PMID PURL UPC URL URN. I don???t
> understand why they do it this way, because, as you say, you can
> generally determine what to do with a URI by looking at its scheme.

Oh, yikes.  Yes, that's even true for DataCite 3.1; and actually, the
vor-to-datacite XSLT uses the the URL name.

I probably have suppressed this memory because I'm very partial to
RFC3986's language:

   Future specifications and related documentation should
   use the general term "URI" rather than the more restrictive terms
   "URL" and "URN" [RFC3305].

This is from 2005.  Sigh.

Anyway, to justify the XSLT's [well, my] choice of "URL": if you check
DataCite 4.0, in the explanation of the URL relatedIdentifierType
(Table 8), they write

  Uniform Resource Locator, also 
  known as web address, is a 
  specific character string that 
  constitutes a reference to a 
  resource. The syntax is: 
  scheme://domain:port/path?query_string#fragment_id

So, they don't restrict to HTTP URIs at all (though the language with
"web address" could be construed to suggest that, but "also known as"
does not pass as binding specification in my book).  And if you look
at RFC3986 (which is of course not explicitly referenced by DataCite
4.0, but I guess just by saying URL, URI, and URN there's an implicit
reference), they define URL as

   The term "Uniform Resource Locator" (URL) refers to the subset of
   URIs that, in addition to identifying a resource, provide a means
   of locating the resource by describing its primary access
   mechanism (e.g., its network "location").

Note the "e.g.".  And for IVOIDs we do provide a means for locating the
resources.

Do I hear anyone call out "sophistry"?

Whether or not someone called out, I'd frankly like it a lot if we
got IVOID into the list of defined relatedIdentifierType, not only to
spare us the sophistry.

Does anyone on this list know what we'd need to do to make that
happen?

       -- Markus