Re: New metadata draft

Hi Jeni,

thanks. Some specific comments:

- On Issue 1 (and actually Issue 2): I think we should decide on this soon, and before publication. In my view, it is important to make the JSON LD compliant, ie, to vote for that compliance in both issues. If so, the example should be the JSON-LD variant that is in the current example text.

- That being said, in the JSON-LD version, the duplication of "GID" vs. "_:GID" is sort of ugly. I am not even sure it would work in general, because a primary key can be a combination of several columns (note that the context in the document would not work either for that case). I think we should simply leave it out; I do not see it as a problem if _in the metadata_ the primary key is not translated into explicit links. Applications, converters, etc, can make the links (eg, for RDF) if needed.

- You say: "Given the location of the CSV file, this metadata file can be located by appending .csvm to the URL". But this is only one option for the location of the metadata file; we also talk about the link header. 

(The "known name" approach would actually violate the R-IndependentMetadataPublication requirement; also, I think it is expected to have the same metadata serve several datasets...)

Or is the word "can" in this sentence meant to be a MAY?

- Section 2.1.1: what is the priority for various properties for the same field? I would presume field has the highest priority but what if the same property is set both through a row and a column?

- Section 3.3.1: is it necessary to list all these DC properties? Why don't we normatively refer to the relevant DCMI documents? After all, it is up to DCMI to specify, eg, the range of the various properties...

- Section 3.4.1: shouldn't it be 'Required Properties' as opposed to 'Required Fields' in the title? Same for 3.4.2?

- Section 3.4.1: What happens if a number of CSV documents share metadata? The @id property becomes problematic...

- Section 3.5.2, title property; I do not understand "if the value of title is an object, the string or strings that are the value of the property of that object whose name is the column language"

That is for now...

Cheers

Ivan



On 21 May 2014, at 19:02 , Jeni Tennison <jeni@jenitennison.com> wrote:

> Hi,
> 
> I’ve done some fairly substantial work on the metadata draft [1] to get the structure and content more towards where I think we want it to head, including trying to map the existing data package structures into something that makes (more) sense if we’re viewing the metadata documents as JSON-LD structures with a metadata vocabulary.
> 
> There’s still a lot of work to do (and loads of issues as you’ll see), but I think it’s a little more internally consistent now. Comments appreciated.
> 
> Jeni
> 
> [1] http://w3c.github.io/csvw/metadata/
> --  
> Jeni Tennison
> http://www.jenitennison.com/
> 


----
Ivan Herman, W3C 
Digital Publishing Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
GPG: 0x343F1A3D
WebID: http://www.ivan-herman.net/foaf#me