Hi Michel,
The work HCLS is doing is open to anyone in the HCLS community. The
best way to influence the process, for the better, is to take some
time and participate in it. We've had several phone calls, publicly
announced. We will have more. Have you joined any of them? There's
also a number of wiki pages, and the invitation to add more material
so that there is more to work with.
Regarding the goals of the report. It is a recommendation on URI
usage. I don't know exactly what it will contain (since that's
something that's being worked on) but I won't rule out
recommendations on URI schemes. Whether or not that happens would
depend on the participants thinking it would be a good idea or not. I
am, personally, not very interested in only giving a neutral history
URI schemes, or only saying what the problem is, even though I expect
that this sort of analysis will be part of the report. The larger
HCLS community is interested in making some progress on this, and
there are frequent calls for specific recommendations, and there a
set of requirements driven by a mixture of existing applications,
past experience with what has worked and not worked, and general
principles of web architecture.
We are ignoring no discussions, but can't be held responsible if we
advertise an open process and interested parties don't show up.
Regards,
Alan
http://esw.w3.org/topic/HCLSIG_BioRDF_Subgroup/Tasks/URI_Best_Practices
On Aug 6, 2007, at 12:48 PM, Michel_Dumontier wrote:
Alan,
The HCLS should outline existing mechanisms by which one can uniquely
identify a resource, independent of its resolution. Bioinformatics has
been forever plagued by innumerate resource identifiers, for which,
finally, semantic web technologies (XML/RDF/OWL) provide the semantics
to support data integration, in the presence of unique identifiers
and,
optionally, schemas/ontologies. A clear description of this problem
and
the requirements for its solutions would be useful outcomes of HCLS
activity.
While there are arguments FOR and AGAINST a great number of URI
schemes,
there is ABSOLUTELY no requirement for a particular resolution
protocol
to be recommended by HCLS. Indeed, by advocating a certain technology,
HCLS "recommendations" will surely alienate a large number of people
that it aims to represent, and ultimately its report will fall by the
wayside. Ignore these discussions at your peril. The fact that the
HCLS
community uses HTTP URIs or LSID URNs or INFO URIs, etc,
illustrates the
breadth of community requirements. An analysis into these motivations
and summarizing existing mechanisms would surely be a useful
contribution for current and future adopters.
Cheers,
-=Michel=-
