- From: Government Linked Data Working Group Issue Tracker <sysbot+tracker@w3.org>
- Date: Fri, 12 Apr 2013 17:04:25 +0000
- To: public-gld-wg@w3.org
ISSUE-62 (DCATscope): Clarify scope of datasets and distributions [DCAT] http://www.w3.org/2011/gld/track/issues/62 Raised by: Richard Cyganiak On product: DCAT This issue combines several items of Last Call feedback. Readers of the spec come away with some uncertainty about its scope and applicability. For example, can it be used to describe APIs? Is everything a dataset? Is it really only intended for downloads? Why the bias towards downloads? At the F2F3, there seemed to be rough consensus on the following: * dcat:Dataset is to be interpreted broadly, it's certainly not just for tabular files * Same for dcat:Distribution * APIs are a kind of distribution, but DCAT does not include properties for describing them in detail because that would be hard, and existing catalogs don't really go into this * We have properties for describing one kind of distributions in more details -- downloads. That's because they are common and easy to describe. This doesn't mean that DCAT isn't applicable to other kinds. * We considered re-introducing a single subclass of dcat:Distribution, for downloads. * It would be reasonable for extensions to define additional subclasses and properties to specify other kinds of distributions, like APIs, in more detail. Now we need to work out what changes are required to the spec in order to clarify the document so that readers get this interpretation.
Received on Friday, 12 April 2013 17:04:26 UTC