Re: Vocabularies/dataset deliverable

Thanks for sharing the Vocabulary and Dataset draft, Antoine! It's an impressive undertaking with great content!
http://www.w3.org/2005/Incubator/lld/wiki/Vocabulary_and_Dataset

I see from the listserv message you referenced [1] that the audience is a library reader who wants an overview of what vocabularies and datasets already exist/are in common use. I'm thus envisioning two library audiences: as novices who want an overview and experts who want a quick lookup/refresher.

From the perspective of giving an overview, I especially like the "cloud" mockup, where an arrow means that a vocabulary is being used in another. (I'd consider switching the direction of the arrows, but maybe that's just me!)
http://www.w3.org/2005/Incubator/lld/wiki/File:LLD-MetadataElementSetCloudMock.png

For the eventual report, I think a table might be easier to read, with:
-abbreviation/short name
-name
-xmls namespace
-domain or usage example

I'm torn about whether the use cases are relevant to the audience -- I think a *very short* wiki page which provided an index of vocabularies by use case might be more useful, since I'm envisioning the wiki (rather than a static appendix) as the "home" of the use cases.

While it's probably too hard to make subgroupings by domain area, a section for library-specific ontologies could be helpful: first to librarians (who will recognize these more quickly than, say vCard or DOAP), and second to semantic web specialists who may learn more about the library vocabularies/datasets from the added attention this would bring.

I see that the overall grouping is currently based on "metadata element sets" versus "value vocabularies and datasets". I know we had extensive discussion on that distinction but must admit that I'm still less-than-clear on it! Good news is it looks like our definition [7] is planned to be included [8] someplace in the prelude to this section!

Also, this may belong elsewhere in the report, but I'd like us to explain what's different when talking about vocabularies and datasets in the Semantic Web context, compared to earlier treatments of metadata schemes (e.g. [5,6]).

Another topic that the report may need to cover is *finding* vocabularies; one challenge raised by the issues/recommendations subgroup is finding and selecting general-purpose vocabularies. Beyond places and persons (which are well-covered in your draft) there are events (maybe in CIDOC CRM? If so, more details on that could help!), measurements,  colors,  etc. 

-Jodi

> [1] http://lists.w3.org/Archives/Public/public-xg-lld/2011Mar/0006.html

[5] http://www.niso.org/publications/press/UnderstandingMetadata.pdf
[6] Pricilla Caplan's Metadata Fundamentals for all Librarians
[7] http://www.w3.org/2001/sw/wiki/Library_terminology_informally_explained#Vocabularies.2C_Element_sets.2C_Datasets
[8] http://www.w3.org/2005/Incubator/lld/wiki/DraftReport#Vocabularies

On 13 Apr 2011, at 08:59, Antoine Isaac wrote:

> Dear all,
> 
> Here is the current draft Marcia, Jeff, William and me have been assembling so far:
> http://www.w3.org/2005/Incubator/lld/wiki/Vocabulary_and_Dataset
> 
> It is still work in progress, but we believe it gives a good idea of what we're aiming at. It's a quite straightforward implementation  of the plan at [1].
> 
> Of course we'll happily welcome every feedback. But your input would be especially appreciated on the following points:
> 
> - graphs: do you agree this can be useful and worth working on? Have you got alternative options to the ones we're exploring now?
> 
> - description grain: are our (two lines intro + references + namespace pointer) good enough for the purpose? Should we really aim at something more complete and structured, in the line of LOV Bernard has worked on [3,4]?
> 
> - coverage: should we include other items? Especially, do you think we're missing reference vocabularies (values or element sets) that are really important for our domain and not referred to in the use cases? Should that include stuff not published in RDF, or "non-official" versions?
> 
> - categorization: we are thinking of arranging together value vocabularies that are designed to handle agents, places names, and all others (topics). That could be useful for the report section, but if you have any input on whether/how it is feasible, please send it!
> 
> 
> Finally, we of course still intend to come with a proper section for the report, including:
> - pointers to the LLD cloud
> - representative vocs and datasets--a mere selection of items from the separate deliverable.
> - gaps, maybe with discussion on data management and publication, if not redundant with the "problems and limitations" or "recommendation" sections--perhaps we could just fit our stuff there, as we discussed last week for [2]
> - work in progress *at the time of the report*, which can solve the gaps. And what should be do to fix the remaining issues--again, if not redundant with the "recommendation" section.
> 
> But we'll work it a bit later. We feel that the current gathering/description/visualization work we're doing now for the report is good for better grasping the situation, and thus to come with relevant stuff in the report.
> 
> Cheers,
> 
> Antoine
> 
> [1] http://lists.w3.org/Archives/Public/public-xg-lld/2011Mar/0006.html
> [2] http://www.w3.org/2005/Incubator/lld/wiki/Vocabularies#Vocabulary_discussion_in_Pittsburgh
> [3]  http://labs.mondeca.com/dataset/lov/
> [4]http://labs.mondeca.com/dataset/lov/details/vocabulary_dc.html
>