Glenn Thanks for your comments. The nc suffix was an example, but I found other cases in the documentation. I found a lot of differences between the convention and the list of requirements and recommendations.
My work is to highlight the differences between a netcdf file (bathymetric data) and the CF convention And what is the way to make these files CF compliant... Best regards Alain MAZURIER Responsable Unité Technologique Logicielle ____________________________________________________________________ Technopôle Brest Iroise Site du Vernis – CS 23866 29238 Brest Cedex 3 Tél : + 33 2 98 05 43 21 Tél direct: + 33 2 98 05 71 90 Mobile : 06 08 92 10 11 Fax : + 33 2 98 05 20 34 www.altran.com Before printing, think about ENVIRONMENTAL responsibility -----Message d'origine----- De : Comiskey, Glenn [mailto:[email protected]] Envoyé : mercredi 3 août 2011 17:48 À : [email protected] Cc : Mazurier Alain Objet : [FRANCE] Question about CF Convention Alain, Further to my earlier comment, and to address your question regarding file name suffix, my company makes data available via THREDDS and ERDDAP neither of which I have found has any problem using files that do not have a ".nc" suffix. When editing the metadata for files to be published via THREDDS and ERDDAP the use of other file suffix, such as ".grb" for GRIB files, has not caused any issues or stopped us from state conformance with conventions such as CF. Regards, Glenn -----Original Message----- From: Comiskey, Glenn Sent: 03 August 2011 16:19 To: '[email protected]' Cc: '[email protected]' Subject: RE: CF-metadata Digest, Vol 100, Issue 2 Bonjour Alain, Being new to the world of meteorological/oceangraphic data myself recently, I had the same problem you have when reading the various conventions available. Based upon discussions about this matter, and how others are using and interpretating the conventions, I personally have adopted a position whereby the term "should" is regarded as a mandatory requirement. That is to say I interpret "should" as "must". While I agree that the general definition of "should" would imply something being optional, the general definition also implies a strong sense of obligation. That is to say you must do something unless you have a good reason not to, i.e. I should not kill anyone (I use this as an example of how strongly "should" could be regarded as obligatory, rather than my homicidal tendencies :-) Hope this helps, though I am sure others will also comment. Salut, Glenn ------------------------------ Message: 6 Date: Wed, 3 Aug 2011 16:31:11 +0200 From: "Mazurier Alain" <[email protected]> Subject: [CF-metadata] [FRANCE] Question about CF Convention To: <[email protected]> Message-ID: <ee5e0b0870aeb74bbf8c9c7bddc9c06004646...@xvs-dcfr-23.europe.corp.altran .com> Content-Type: text/plain; charset="iso-8859-1" Hi all, I am new on this mailing list and I hope my question is not so obvious My problem is with: At the adress : http://cf-pcmdi.llnl.gov/conformance/requirements-and-recommendations/1. 5/ I find ? 2.1 Filename Requirements: * Filename must have ".nc" suffix." But If I read the CF documentation http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.5/cf-conventions-mul ti.html I find "2.1. Filename NetCDF files should have the file name extension ".nc". " For me there is a difference between "must" and "should"?. If i have another suffix than nc, can i be CF compliant? Best regards Alain MAZURIER Responsable Unit Technologique Logicielle _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
