Re: Draft Relevant Technologies (and vocabularies section)

On 4/19/11 6:10 PM, Karen Coyle wrote:
> Quoting Antoine Isaac <aisaac@few.vu.nl>:
>
>
>>
>> Trying to understand, is something like Poolparty [1] at the right level of tooling you'd be expecting (for SKOS in that case)?
>
>
> The only example of a tool that is at the right level of service that I know of is the Open Metadata Registry. A Poolparty implementation might also be an example (I looked at it briefly but couldn't get a good understanding so quickly). The registry is a tool with a full user interface that requires very little (if any) knowledge of the underlying LD code. Users fill in a form for names, labels, descriptions, broader terms, and that's all. The service then provides all of the translation of this to RDF/OWL/SKOS.
>
> If I were to give requirements for a tool, it would be:
> - requires no technical skills beyond those of a general computer user (meaning either online forms or a program that an average Windows user can install)
> - the interface is expressed in the language of the user
> - provides help and direction where decisions must be made
>
> Note, however, the the Registry and Poolparty are designed for DEVELOPMENT of metadata properties and vocabularies, and what we really need are tools that help us create and use the instance data. That's the tool that I am asked about every time I speak about LD. The people who would create the properties in Poolparty are not the same people who would create the instance data (in most cases) and tools for the latter must be even closer to the mental models used by those data creators.



Right. But then you're really focusing on one specific category of tool, which is useful but should not be our only point. We should be careful about it: as Lukas and Jeff pointed out in their respective contributions, 'a ready-made "marc-2-rdf" converter' and other developer-oriented tools would be interesting, as well, of course.

Antoine

Received on Tuesday, 19 April 2011 17:06:03 UTC