[datacatalogs] Request for Comments on Draft Data Catalog Standard (Schema and Protocol)

Thu Jun 14 14:05:31 UTC 2012

On 12 June 2012 17:02, Ed Summers <ehs at pobox.com> wrote:
> On Tue, Jun 12, 2012 at 11:22 AM, Rufus Pollock <rufus.pollock at okfn.org> wrote:
>> One question of course is whether the URL is the API url or the url of
>> the human readable version (sure with content-negotiation this isn't
>> so relevant but will everyone support that ...)
>
> Yes, I agree. One approach to entertain would be support for a JSON
> equivalent of Atom's link element in the changes.json:
>
>    <entry>
>    ...
>    <link rel="alternate" type="text/html"
> href="http://example.org/dataset/1234.html">
>    <link rel="altnernate" type="application/json"
> href="http://example.org/dataset/1234.json">
>    <link rel="alternate" type="application/rdf+xml"
> href="http://example.org/dataset/1234.rdf">
>    </entry>
>
> This allows clients to pick the mediatype they understand and act
> accordingly. If the server supported negotiation at one URL it could
> supply a JSON version of this instead:
>
>    <entry>
>    ...
>    <link rel="alternate" type="text/html"
> href="http://example.org/dataset/1234">
>    <link rel="alternate" type="application/json"
> href="http://example.org/dataset/1234">
>    <link rel="alternate" type="application/rdf+xml"
> href="http://example.org/dataset/1234">
>    </entry>
>
> I imagine conneg wouldn't get implemented that much, but you never
> know. For what this might look like in JSON the page on RESTful JSON
> [1] and in particular this post by Subbu Allamaraju might provide some
> guidance [2]. I could sketch up an example if you are interested.

That would be great. I think we are getting to the stage of actually
drafting a patch and then reviewing.

>>> 2) Are the full representations of the dataset made available in the
>>> changes.json, or will clients need to fetch the dataset to get the
>>> full information?
>>
>> They will need to fetch. changes.json is lightweight.
>
> All the more reason for supplying a URL for the dataset so that
> clients can fetch it easily without needing to construct URLs.
>
>>> 4) Are both JSON and RDF representations of the dataset required?
>>
>> No only the JSON representation is *required* atm with option to
>> provide other formats (n3 / rdf/xml etc)
>
> I think the link constructs discussed above, or something like them,
> would allow servers to describe what representations are available.
> Maybe?

I'd need to think harder or see an example.

> Thanks for the feedback Rufus, and for talking openly about this as it
> is being developed--much appreciated.

:-)

Rufus