[Go-essp-tech] error in "Model" name(s)

Fri Mar 4 03:13:10 MST 2011

Hi

I agree with Stephen's comments.  We use CMOR2 in a completely standard way and it maintains case sensitivity of the model name - we have done nothing to force upper and lower case either in the filename or the header in our processing.  But in any case, as Karl indicated, the system needs to retain case

Mark
___________________________________________________        
Dr Mark Elkington - Met Office Hadley Centre                            
email:  mark.elkington at metoffice.gov.uk <mailto:mark.elkington at metoffice.gov.uk>                                    
tel: 01392  884835                                                                   
location: A2-93  

________________________________

	From: Mark Morgan [mailto:momipsl at ipsl.jussieu.fr] 
	Sent: 04 March 2011 11:02
	To: stephen.pascoe at stfc.ac.uk
	Cc: Stephane.Senesi at meteo.fr; drach1 at llnl.gov; Elkington, Mark; go-essp-tech at ucar.edu
	Subject: Re: [Go-essp-tech] error in "Model" name(s)

	Hi 

	What this does impact upon is customized search engines.  At the IPSL in France we have developed a portal based upon metadata aggregated from theTHREDDS catalogs exposed by the ESG Data-Nodes in the IPSL federation (currently 4 data nodes).   When ingesting this metadata we build case insensitive search indexes so that we resolve potential discrepancies between CMOR2 / ESG-Publisher / DRS controlled vocab.

	Mark

	On 4 Mar 2011, at 10:31, <stephen.pascoe at stfc.ac.uk> wrote:

		Stephane and everyone,

		> CMOR2 enforces model names to be lowercase, so there should be no issue here, except if one admits that avoiding the use of CMOR2 is a
		> way to get more visibility by using an uppercase model name

		We have data from MOHC that has been through CMOR2 with mixed-case model names in the metadata and file names.  Since MOHC did the processing I can't vouch precisely for the method used (maybe Mark wishes to comment) but as far as I know CMOR2 supports mixed-case model names.

		Case sensitivity in DRS components has been a tricky issue in the design of the whole system and we've never really converged on a consistent approach.  The DRS document has always been very clear that model names are mixed case, as a result drslib supports mixed-case model names.  The publisher, by contrast, has usually been configured to use lower-case model names, however Bob and I have agreed that publishers can configure mixed-case models into the publisher configuration.

		Other than the obvious bug of prefixing "model_" to the name, I think the case sensitivity should be retained from what ever is published.  BADC datasets are consistently named in mixed case in the THREDDS catalogs, filenames and NetCDF metadata.

		Cheers,
		Stephen.

		---
		Stephen Pascoe  +44 (0)1235 445980
		Centre of Environmental Data Archival
		STFC Rutherford Appleton Laboratory, Harwell Oxford, Didcot OX11 0QX, UK

		From: go-essp-tech-bounces at ucar.edu [mailto:go-essp-tech-bounces at ucar.edu] On Behalf Of Stéphane Senesi
		Sent: 04 March 2011 08:01
		To: go-essp-tech at ucar.edu
		Subject: Re: [Go-essp-tech] error in "Model" name(s)
		Karl Taylor wrote, On 03/03/2011 18:10:
		Thanks, Nate.

		One other item, which may already be ticketed, but wanted to make sure:

		Except on the home gateway, the model names under search category "Model" appear all in lower case with a prefix of "model_".  For example, on PCMDI's gateway, the model published at BADC is listed as "model_hadgem2-es", but it should appear just as it appears at the BADC gateway: "HadGEM2-ES"

		Hi, all

		CMOR2 enforces model names to be lowercase, so there should be no issue here, except if one admits that avoiding the use of CMOR2 is a way to get more visibility by using an uppercase model name

		(Sorry if the same is said at https://vets.development.ucar.edu/jira/browse/GTWY-2058; I do not have access to this site)

		Regards
		S

		thanks,
		Karl

		On 3/2/11 7:01 PM, Nathan Wilhelmi wrote:
		Hi Karl,

		Thanks for the feedback, a couple of inline comments.

		On 03/02/2011 06:47 PM, Karl Taylor wrote:
		Dear all,

		I hope that the following two items can be implemented soon:

		1.  Users should be able to ask ESG to display only datasets they have permission to download:

		Without this, a "CMIP5 commercial" user will have to subselect (point and click on!) only the models they are allowed to download, and they will have to consult an external table to figure out which tables those are.  If they don't do this, when they go to download the files they will be asked if they want to join the "CMIP5 research" group, and I don't want them to do this if they are going to use the data for commercial purposes.    We're asking a lot of users to expect them to jump through lots of hoops to honor the "terms of use".
		I have added a JIRA ticket for this one: https://vets.development.ucar.edu/jira/browse/GTWY-2182

		Similarly lots of CMIP5 data is already published (and visible via ESG), but hasn't been released for download.  Users who want to download CMIP5 data will have to eliminate one by one all the models that are not yet public before downloading, and it is not obvious which models those are.  This is a pain.

		2.    I noticed that the institute information is lost when a gateway harvests datasets from another gateway (at least it doesn't appear in the "Search Categories").  If you look at the institutes listed at BADC, you find MOHC, but the same datasets listed at PCMDI or NASA JPL don't include the institute information under "Search Categories". 
		We already have this one logged here: https://vets.development.ucar.edu/jira/browse/GTWY-2058

		Thanks!

		-Nate

		Best regards,
		Karl

		_______________________________________________
		GO-ESSP-TECH mailing list
		GO-ESSP-TECH at ucar.edu
		http://mailman.ucar.edu/mailman/listinfo/go-essp-tech

		_______________________________________________
		GO-ESSP-TECH mailing list
		GO-ESSP-TECH at ucar.edu
		http://mailman.ucar.edu/mailman/listinfo/go-essp-tech

		-- 
		Stéphane Sénési
		Ingénieur - équipe Assemblage du Système Terre
		Centre National de Recherches Météorologiques
		Groupe de Météorologie à Grande Echelle et Climat

		CNRM/GMGEC/ASTER
		42 Av Coriolis
		F-31057 Toulouse Cedex 1

		+33.5.61.07.99.31 (Fax :....9610)

		-- 
		Scanned by iCritical.

		_______________________________________________
		GO-ESSP-TECH mailing list
		GO-ESSP-TECH at ucar.edu
		http://mailman.ucar.edu/mailman/listinfo/go-essp-tech

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/go-essp-tech/attachments/20110304/5f11ff5d/attachment-0001.html 