Re: Questions about metadata vocabulary and model for tabular data and metadata on the Web

Dear Esther,

as you can see, I've copied this mail to our mailing list, because others in the group may have better answers than I do!

My answers are based on what we call the editors' draft of the specifications, ie, versions that are significantly ahead of the published documents. (We plan to publish those as final designs in a few weeks from now, b.t.w.) With that caveat:

- The issue on the list of accepted columns I can see two possibilities right now (a simple and a complex one):
 - a string datatype can be 'restricted' through the usage of the the 'format' parameter[1] which is, essentially, a regular expression
 - it is possible to add an '@id' parameter[2] (in the new version) that refers to an externally defined datatype. That can be defined in OWL, in XSD or whatever else. The downside is that standard validator software are not required to follow up and handle those.

- The issue of 'same as', 'see also', etc: that can be solved by the usage of virtual columns. The editor's draft document contains some examples[3] that can be easily converted to your use case I believe.

But, again, my colleagues may come up with better answers!

Thanks for your interest

Ivan Herman

[1] http://w3c.github.io/csvw/syntax/#formats-for-other-types
[2] http://w3c.github.io/csvw/metadata/index.html#h-derived-datatypes
[3] http://w3c.github.io/csvw/metadata/index.html#h-use-of-virtual-columns




> On 02 Jul 2015, at 11:32 , Ivan Herman <ivan@w3.org> wrote:
> 
> Guys,
> 
> forwarding this mail to the mailing list because (a) there are interesting questions and (b) because it is good to have information about other use cases!
> 
> Ivan
> 
>> Begin forwarded message:
>> 
>> From: Esther DZALE YEUMO <edzale@versailles.inra.fr>
>> Subject: Questions about metadata vocabulary and model for tabular data and metadata on the Web
>> Date: 2 Jul 2015 11:25:08 CEST
>> To: ivan@w3.org
>> 
>> Greatings,
>> I am writting on behalf of the Wheat Data Interoperability working group. We are considering following the recommendations provided by the CSV on the Web working group to provide guidelines on how to publish wheat data in tabulated files, specificaly germplasm passport information.
>> We started for instance a metadata document available at the following: https://github.com/edzale/wheat/blob/master/mcpd.json
>> I would appreciate if you could help us answer the following questions:
>> Is there any way to specify a list of accepted values for a column? For instance, for the column "ORIGCTY" in our schema, how could we indicate that the values for this coumn should come from an existing controlled list (either available at a specific URL or embeded within the schema)? Also, for the same column, how could we establish a link out (same as, see also, etc.) to the geonames ontolotogy?
>> Thank you very much.
>> Regards,
>> Esther Dzalé Yeumo
>> --
>> --
>> <Mail Attachment.gif>
>> 
>> Esther DZALE YEUMO KABORE
>> Chef de projet / Project manager
>> Esther.Dzale-Yeumo@versailles.inra.fr
>> Direction déléguée à l’IST / Scientific Information department
>> Tèl. : +33 1 (0)1 30 83 34 44
>> Fax : +33 1 (0)1 30 83 34 40
>> RD10
>> 78 026 Versailles cedex
>> France
>> www.inra.fr
> 
> 
> ----
> Ivan Herman, W3C
> Digital Publishing Activity Lead
> Home: http://www.w3.org/People/Ivan/
> mobile: +31-641044153
> ORCID ID: http://orcid.org/0000-0003-0782-2704
> 
> 
> 
> 


----
Ivan Herman, W3C
Digital Publishing Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: http://orcid.org/0000-0003-0782-2704