I was hoping that this conversation would end on Bryan's statement or
thereabouts
Actually, I think we really want:
>
> http://cfconventions.org/something
>
> (We own cfconventions.org, so we can point it where we want!)
>
> And I would like in 2011 to have settled on a permanent answer to that
> question!
>
to which I would like to add Amen! (but I did not want any more e-mails)
It is, after all, a pithy answer to the original question.
But it was not to be. So now I try to fix the confusion I created earlier,
and add my point-of-view to what followed.
First of all, we can establish a set of URIs for the standard_name
vocabulary without establishing a service, as nice as a service might be.
Two separate questions.
Secondly, the two sets of SKOS documents are significantly different, and
both are missing the essential connection to CF that was in the original MMI
construction (my earlier misunderstood point), namely
cfsns:standard_name
a property that connects the concept to the string that appears in a dataset
documented according to the CF convention.
skos:prefLabel is not good enough, because it does not mean that in theory
(it is a display label). On a practical note, skos:prefLabel is the
natural place for language translation, or even getting rid of the _, in
which case it will be multivalued with each label tagged by a different
language. I think. And BODC serves other vocabularies that have
skos:prefLabel that are not cf:standard_names. And we do not want to create
the situation where skos:prefLabel has this additional meaning when served
from this particular url on this particular server.
This is not saying SKOS is a bad thing, in fact BODC has SKOS-relations
between standard_names that do not even exist elsewhere, but neither of
these representations is complete enough to connote standard_names. So
like all representations, they are partial. We should be mapping between
them, not picking a victor. Someone needs to serve the mapping. And CF
needs to serve the mapping between URIs and attribute strings. *That* is
the standard.
And then there is something I really do not understand. Both BODC and MMI
have versioning in the terms, very good for a vocabulary maintenance point
of view. But CF does not have versioning on the standard_name attributes in
a file. So all the standard_name s that have the same string representation
in CF attributes have to be the same in the fundamental sense that given a
dataset with a standard_name tag, it has to map to all the
URIs-with-version that have that standard_name tag. So isn't carefully
versioning the terms just creating a mess in URI space? If you are going
to version whole sets, aren't you obligated to map between versions so that
we can figure out which terms have not really changed? I understand that if
one is picking one-from-a-set and you change the set, you could change the
pick, but we cannot really afford that, since old datasets are not getting
relabeled and we cannot find the version, anyway. If we do version
standard_name values, than shouldn't the URI's start from when that is
required?
Benno
On Wed, Dec 15, 2010 at 4:46 AM, Bryan Lawrence
<bryan.lawrence at stfc.ac.uk>wrote:
> Hi Folks
>
> I have to declare a preference to pointing to the BODC copy as the
> primary copy since we've built our CF vocab management tooling to ingest
> new vocab members directly into their vocabuarly managment system - and
> if we have any problems, debugging them can be done on the same
> timezone.
>
> Cheers
> Bryan
>
>
> > Ditto! Implementing what is suggested would be around 20 lines of
> > code. We've done the same for SeaDataNet to get their vocabs within
> > their namespace such as
> > http://www.seadatanet.org/urnurl/SDN:P021::TEMP
> >
> > Cheers, Roy.
> >
> > ________________________________________
> > From: cf-metadata-bounces at cgd.ucar.edu
> > [cf-metadata-bounces at cgd.ucar.edu] On Behalf Of John Graybeal
> > [jbgraybeal at mindspring.com] Sent: 14 December 2010 17:56
> > To: Steve Hankin
> > Cc: cf-metadata at cgd.ucar.edu
> > Subject: Re: [CF-metadata] Web reference to a standard name?
> >
> > In theory, MMI can support that approach (redirecting
> > cfconventions.org, ideally transparently). (That is, I worked it
> > out in my head once. Haven't actually tried it in practice. I
> > *think* it's "just a simple matter of programming". ;->) I think
> > that's the best combination to support both appropriate branding and
> > the kind of support a "full time dedicated" repository can give.
> > Roy's server perhaps the same. (But not both at once!)
> >
> > Just to be crystal clear, the places where you have '16' could also
> > have 'current' (if I understand correctly what Roy was saying about
> > their server), and the mmisw one could also be served with a
> > particular version ID (analogous to the NERC example).
> >
> > If there is someone on the CF (or other) team that wants to play with
> > us on the 'simple matter of programming' part, that would be a
> > collaboration we'd be happy to encourage.
> >
> > John
> >
> > On Dec 14, 2010, at 08:16, Steve Hankin wrote:
> > > I second that thought.
> > >
> > > On 12/14/2010 3:12 AM, Bryan Lawrence wrote:
> > >> Hi Dom
> > >>
> > >> Actually, I think we really want:
> > >>
> > >> http://cfconventions.org/something
> > >>
> > >> (We own cfconventions.org, so we can point it where we want!)
> > >>
> > >> And I would like in 2011 to have settled on a permanent answer to
> > >> that question!
> > >>
> > >> Cheers
> > >> Bryan
> > >>
> > >>> Thanks all,
> > >>>
> > >>> Interesting to see the new MMI ontology work.
> > >>> I take from this discussion then that there is not currently a CF
> > >>> 'blessed' method to reference a term?
> > >>>
> > >>> E.g. for air_temperature I can choose from any of these:
> > >>>
> > >>> http://mmisw.org/ont/cf/parameter/air_temperature
> > >>> http://vocab.ndg.nerc.ac.uk/term/P071/16/CFSN0023
> > >>> http://cf-pcmdi.llnl.gov/documents/cf-standard-names/standard-nam
> > >>> e-ta ble/16/cf-standard-name-table.html#air_temperature plus:
> > >>> http://cf-pcmdi.llnl.gov/documents/cf-standard-names/standard-na
> > >>> me-ta ble/16/cf-standard-name-table.xml
> > >>>
> > >>>
> > >>> Cheers,
> > >>> Dom
> > >>>
> > >>> On 14/12/10 08:18, Lowry, Roy K. wrote:
> > >>>> Dear All,
> > >>>>
> > >>>> Working to find a way of harmonising the two sources of
> > >>>> web-served Standard Names is on by ?To-Do? list for next year
> > >>>> (part of an issue list fed into the NETMAR project by John G.)
> > >>>>
> > >>>> Cheers, Roy.
> > >>>>
> > >>>> *From:* cf-metadata-bounces at cgd.ucar.edu
> > >>>> [mailto:cf-metadata-bounces at cgd.ucar.edu] *On Behalf Of *John
> > >>>> Graybeal *Sent:* 13 December 2010 22:28
> > >>>> *To:* Benno Blumenthal
> > >>>> *Cc:* cf-metadata at cgd.ucar.edu
> > >>>> *Subject:* Re: [CF-metadata] Web reference to a standard name?
> > >>>>
> > >>>> A minor clarification as to how MMI is doing things:
> > >>>>
> > >>>> Benno, all,
> > >>>>
> > >>>> On Dec 13, 2010, at 07:25, Benno Blumenthal wrote:
> > >>>> We have tried to get standard URIs for CF concepts, but the
> > >>>> effort bogged down.
> > >>>>
> > >>>> As for standard names, MMI established a set of URIs at
> > >>>>
> > >>>> http://marinemetadata.org/cf
> > >>>>
> > >>>> but has not been keeping up to date.
> > >>>>
> > >>>> Thanks for the mention. This is our older URL, sorry; we are now
> > >>>> maintaining CF standard names in our ontology repository, using
> > >>>> resolvable URLs. [1]
> > >>>>
> > >>>> There was a period when MMI was not keeping up to date with CF
> > >>>> changes, but in the recent past we have been continuously
> > >>>> updating the SKOS ontology for the parameter names at the site
> > >>>> Carlos mentioned [1] (We will forward the misleading /cf URL
> > >>>> above, or fix it to make clear in some other way the current
> > >>>> status; thanks for calling that out.)
> > >>>>
> > >>>> As with Roy's handling, we provide URLs both for the current
> > >>>> version at the reference URL, and for individual releases of the
> > >>>> vocabulary.
> > >>>>
> > >>>> John
> > >>>>
> > >>>> [1] http://mmisw.org/ont/cf/parameter
> > >>>>
> > >>>>
> > >>>>
> > >>>> I have written a XSL transform at
> > >>>>
> > >>>> http://iridl.ldeo.columbia.edu/ontologies/xslt/.cfsn2rdf.xsl
> > >>>>
> > >>>> that can be applied to
> > >>>>
> > >>>> http://cf-pcmdi.llnl.gov/documents/cf-standard-names/standard-na
> > >>>> me- table/16/cf-standard-name-table.xml
> > >>>>
> > >>>> to generate an up-to-date version, e.g.
> > >>>>
> > >>>> xsltproc
> > >>>> http://iridl.ldeo.columbia.edu/ontologies/xslt/.cfsn2rdf.xsl
> > >>>> http://cf-pcmdi.llnl.gov/documents/cf-standard-names/standard-na
> > >>>> me -table/16/cf-standard-name-table.xml
> > >>>>
> > >>>> On the other hand, Roy has created opaque URIs for standard
> > >>>> names at
> > >>>>
> > >>>> http://vocab.ndg.nerc.ac.uk/list/P07/9 contains all the Standard
> > >>>> Names that have ever been published, including names that have
> > >>>> been aliased
> > >>>>
> > >>>> http://vocab.ndg.nerc.ac.uk/list/P071/9 contains the Standard
> > >>>> Names that are currently valid (i.e. those that have not been
> > >>>> deprecated through aliasing)
> > >>>>
> > >>>> http://vocab.ndg.nerc.ac.uk/list/P072/9 contains the Standard
> > >>>> Names that have been deprecated through aliasing
> > >>>>
> > >>>> The mapping between deprecates and their replacement aliases my
> > >>>> be obtained through the following API call
> > >>>>
> > >>>> http://vocab.ndg.nerc.ac.uk/axis2/services/vocab/getMap?subjectL
> > >>>> ist
> > >>>> =http://vocab.ndg.nerc.ac.uk/list/P072/current&predicate=255&ob
> > >>>> ject
> > >>>> List=http://vocab.ndg.nerc.ac.uk/list/P071/current&inference=fa
> > >>>> lse
> > >>>> <http://vocab.ndg.nerc.ac.uk/axis2/services/vocab/getMap?subjec
> > >>>> tLi
> > >>>> st=http://vocab.ndg.nerc.ac.uk/list/P072/current&predicate=255&
> > >>>> obje
> > >>>> ctList=http://vocab.ndg.nerc.ac.uk/list/P071/current&inference=
> > >>>> fals e>
> > >>>>
> > >>>>
> > >>>>
> > >>>>
> > >>>> he changes them with each version, so it is both more definitive
> > >>>> and harder to use. It is also pure SKOS, so one needs to add
> > >>>> structure to actually connect it with a CF-described dataset. I
> > >>>> plan to write out a map between the two, so that I can use his
> > >>>> relationships between CF standard_names, but I do not have it
> > >>>> yet.
> > >>>>
> > >>>> Additional information and ontologies describing the CF
> > >>>> structure both as attributes and conceptually are available at
> > >>>> http://iridl.ldeo.columbia.edu/ontologies/, including the
> > >>>> connections between the standard_name ontologies and the CF
> > >>>> structure.
> > >>>>
> > >>>>
> > >>>>
> > >>>> Benno
> > >>>>
> > >>>>
> > >>>>
> > >>>> On Mon, Dec 13, 2010 at 9:22 AM, Dominic Lowe
> > >>>> <dominic.lowe at stfc.ac.uk
> > >>>>
> > >>>> <mailto:dominic.lowe at stfc.ac.uk>> wrote:
> > >>>> > Hello all,
> > >>>> >
> > >>>> > What's the current CF recommended scheme for referencing a
> > >>>> > standard
> > >>>>
> > >>>> name on
> > >>>>
> > >>>> > the web? (either by URL or URN).
> > >>>> >
> > >>>> > I know about the vocab server and the xml/html online
> > >>>> > versions and could choose to use URLs to point to items in
> > >>>> > any of these but is there a definitive pattern that should
> > >>>> > be used? (or at least a best practice that should be
> > >>>> > encouraged?).
> > >>>> >
> > >>>> > Much appreciated,
> > >>>> >
> > >>>> > Dominic
> > >>>> > --
> > >>>> > Scanned by iCritical.
> > >>>> > _______________________________________________
> > >>>> > CF-metadata mailing list
> > >>>> > CF-metadata at cgd.ucar.edu<mailto:CF-metadata at cgd.ucar.edu>
> > >>>> > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > >>>>
> > >>>> --
> > >>>> Dr. M. Benno Blumenthal benno at iri.columbia.edu
> > >>>> <mailto:benno at iri.columbia.edu>
> > >>>> International Research Institute for climate and society
> > >>>> The Earth Institute at Columbia University
> > >>>> Lamont Campus, Palisades NY 10964-8000 (845) 680-4450
> > >>>>
> > >>>> _______________________________________________
> > >>>> CF-metadata mailing list
> > >>>> CF-metadata at cgd.ucar.edu<mailto:CF-metadata at cgd.ucar.edu>
> > >>>> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > >>>>
> > >>>> John Graybeal<mailto:jgraybeal at ucsd.edu>
> > >>>>
> > >>>> phone: 858-534-2162
> > >>>>
> > >>>> System Development Manager
> > >>>>
> > >>>> Ocean Observatories Initiative Cyberinfrastructure Project:
> > >>>> http://ci.oceanobservatories.org
> > >>>>
> > >>>> Marine Metadata Interoperability Project:
> > >>>> http://marinemetadata.org
> > >>>>
> > >>>>
> > >>>> --
> > >>>> This message (and any attachments) is for the recipient only.
> > >>>> NERC is subject to the Freedom of Information Act 2000 and the
> > >>>> contents of this email and any reply you make may be disclosed
> > >>>> by NERC unless it is exempt from release under the Act. Any
> > >>>> material supplied to NERC may be stored in an electronic
> > >>>> records management system.
> > >>>>
> > >>>>
> > >>>>
> > >>>> _______________________________________________
> > >>>> CF-metadata mailing list
> > >>>> CF-metadata at cgd.ucar.edu
> > >>>> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > >>
> > >> --
> > >> Bryan Lawrence
> > >> Director of Environmental Archival and Associated Research
> > >> (NCAS/British Atmospheric Data Centre and NCEO/NERC NEODC)
> > >> STFC, Rutherford Appleton Laboratory
> > >> Phone +44 1235 445012; Fax ... 5848;
> > >> Web: home.badc.rl.ac.uk/lawrence
> > >> _______________________________________________
> > >> CF-metadata mailing list
> > >> CF-metadata at cgd.ucar.edu
> > >> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > >
> > > _______________________________________________
> > > CF-metadata mailing list
> > > CF-metadata at cgd.ucar.edu
> > > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> >
> > John Graybeal <mailto:jgraybeal at ucsd.edu>
> > phone: 858-534-2162
> > System Development Manager
> > Ocean Observatories Initiative Cyberinfrastructure Project:
> > http://ci.oceanobservatories.org Marine Metadata Interoperability
> > Project: http://marinemetadata.org
> >
> > _______________________________________________
> > CF-metadata mailing list
> > CF-metadata at cgd.ucar.edu
> > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> > _______________________________________________
> > CF-metadata mailing list
> > CF-metadata at cgd.ucar.edu
> > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
> --
> Bryan Lawrence
> Director of Environmental Archival and Associated Research
> (NCAS/British Atmospheric Data Centre and NCEO/NERC NEODC)
> STFC, Rutherford Appleton Laboratory
> Phone +44 1235 445012; Fax ... 5848;
> Web: home.badc.rl.ac.uk/lawrence
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
--
Dr. M. Benno Blumenthal benno at iri.columbia.edu
International Research Institute for climate and society
The Earth Institute at Columbia University
Lamont Campus, Palisades NY 10964-8000 (845) 680-4450
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cgd.ucar.edu/pipermail/cf-metadata/attachments/20101215/106ec8cd/attachment-0001.html>
Received on Wed Dec 15 2010 - 15:22:48 GMT
