W3C home > Mailing lists > Public > public-lld@w3.org > July 2011

Re: Review of the "Available Vocabularies and Datasets" section

From: Antoine Isaac <aisaac@few.vu.nl>
Date: Mon, 25 Jul 2011 22:06:05 +0200
Message-ID: <4E2DCCAD.4090107@few.vu.nl>
To: "Haffner, Alexander" <A.Haffner@d-nb.de>
CC: public-lld@w3.org
Dear Alex,


> first of all, I really like this section - good work guys!
> I only have three little comments which maybe could improve the
> understanding.


Thanks very much for your kind words! And for the feedback. This was in fact overlapping with some comments we've received later, so we postponed our answer until we could fully work out the answers to the others...


  
> After the announcement of the RDA implementation decision in the US I
> would appreciate to highlight the RDA element sets as an example for
> metadata element sets.


RDA elements are already mentioned in the specific section on metadata element sets in the side deliverable we've written, via the work happening at the Metadata Registry:
http://www.w3.org/2005/Incubator/lld/wiki/Vocabulary_and_Dataset#Functional_Requirements_for_Bibliographic_Records_.28FRBR.29_and_related_ontologies

We however understand the wish for more visibility, especially in our community. We've also added a sentence
"Resource Description and Access (RDA) defines elements for cataloging, based on the FRBR model."
next to the FRBR example:
http://www.w3.org/2005/Incubator/lld/wiki/Vocabulary_and_Dataset#Introduction:_Scope_and_Definitions

(the report section will be updated with the new content of the side deliverable on vocabularies and datasets, as soon as the changes made there receive positive feedback)

  
>   I am not really good with listing GeoNames as value vocabulary.
> GeoNames is an pretty complex dataset - just see the offer of premium
> data access. In the report we say value vocabularies focus: "on the
> management of discrete value/label literals for use in metadata
> records". This conflicts with my understanding of the GeoName approach
> and maybe confuse readers.


We agree that GeoNames is a complex dataset. We however stand behind the fact that it has huge potential for being used as a "digital gazetteer", a kind of resource we list among other value vocabulary types. We've also added "features" as a clarification for what we consider to be a "value vocabulary" in Geonames. (again, in the side deliverable, but we'll port it to the report section if people agree with our changes).


On the issue of conflicting definitions, we indeed face a difficult situation. The readers familiar with the linked data approach will be aware of the situation presented in the last sentence that focuses on the use of URIs for resources:
" Value vocabularies often have http URIs assigned to the value, which would appear in a metadata record instead of or in addition to the literal value."
To these, using literals as object of triples may seem worthless, while URIs could be made available and used in description.
On the other hand, the readers accustomed to the traditional approach the community has had for the past decades, using codes or natural language labels in metadata records, may have a hard time seizing what is at stake in using URIs.

We believe we can't really avoid the conflict, with both practices bound to co-exist for some time. We hope however that the current wording of the introduction at
http://www.w3.org/2005/Incubator/lld/wiki/Vocabulary_and_Dataset#Introduction:_Scope_and_Definitions
would make the report section easier to read.


>
> Regarding metadata element sets for libraries I am missing one or two
> sentences which recommend a coordinated use of element sets at least in
> the community - without naming particular element sets. Out of my point
> of view it's difficult to establish union catalogs like VIAF if every
> institution forges own new element sets. VIAF is this successful because
> all institutions provide for the match&  merge processes similar
> datasets (currently in MARC 21). I reckon in middle- and long-term
> library committees should give recommendations for the re-use of
> established linked data element sets for particular entity descriptions.
> I'm in fear that we otherwise remove again from interoperability and
> internationalization approaches. You touch this subject already in the
> "Linking" section, maybe we should more emphasize this issue in here.


We are keen on adding one sentence that re-uses much of your wording:
http://www.w3.org/2005/Incubator/lld/wiki/index.php?title=Draft_Vocabularies_Datasets_Section&diff=5345&oldid=5220

We hope you will agree with these answers and suggested changes,

Best,

Marcia, Antoine, Jeff, William




> Hi,
>
> first of all, I really like this section - good work guys!
> I only have three little comments which maybe could improve the
> understanding.
>
> After the announcement of the RDA implementation decision in the US I
> would appreciate to highlight the RDA element sets as an example for
> metadata element sets.
>
>   I am not really good with listing GeoNames as value vocabulary.
> GeoNames is an pretty complex dataset - just see the offer of premium
> data access. In the report we say value vocabularies focus: "on the
> management of discrete value/label literals for use in metadata
> records". This conflicts with my understanding of the GeoName approach
> and maybe confuse readers.
>
> Regarding metadata element sets for libraries I am missing one or two
> sentences which recommend a coordinated use of element sets at least in
> the community - without naming particular element sets. Out of my point
> of view it's difficult to establish union catalogs like VIAF if every
> institution forges own new element sets. VIAF is this successful because
> all institutions provide for the match&  merge processes similar
> datasets (currently in MARC 21). I reckon in middle- and long-term
> library committees should give recommendations for the re-use of
> established linked data element sets for particular entity descriptions.
> I'm in fear that we otherwise remove again from interoperability and
> internationalization approaches. You touch this subject already in the
> "Linking" section, maybe we should more emphasize this issue in here.
>
> Best, alex
>
Received on Monday, 25 July 2011 20:04:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 25 July 2011 20:04:56 GMT