Re: Best Practices doc section 1.3 Vocabulary Selection

On 08/02/2012 19:49, Bernadette Hyland wrote:
> Hi Phil,
> First, thanks for circulating.  While the short extract is helpful, reading it without the context of the entire document leaves me with a number of questions.

Sorry, yes.

I don't know if this doc you're writing is meant to be informational or 
hard core guidance for EU publishers.

It was for a very specific target market, yes - working groups set up 
under the particular EU activity, certainly not W3C WGs.

   The except you attached in PDF seems more the later, but let's 
discuss further as vocabs is on Thursday's GLD WG agenda.

Happy to if you want me to. I expect to be on the call tomorrow.

> Feedback with the caveat that I don't how you plan to modify this content for the BP section on vocab selection:

I was offering it more as some text that might be useful to pick bits 
out of for the BP doc *if* the relevant editors felt it useful. I won't 
feel in the slightest way upset if the nect button pressed is marked 
'Delete' !

> * p. 30 - you reference "the Working Groups" - are these W3C working groups, SEMIC working groups, gov't working groups?

See above.

> * p. 30 - need definition of "Core Vocabularies" - no doubt that is specified in the first 29 pages, but is that a set of Core Vocabularies proposed by ISA or a given gov't authority?

It was in the ISA context, although I think it is probably valid for us 
too. The ISA Programme definition is: "A Core Vocabulary is a 
simplified, reusable, and extensible data model that captures the 
fundamental characteristics of an entity in a context-neutral fashion." 
It's the context-neutral aspect that I think is most important. Within a 
context of course one can be very specific.

> * p. 30 - would you consider adding to the first bulleted list:
> a) Who is the authority publishing the vocab?
> b) Does the publishing authority have an institutional commitment to keeping it current and available?

Certainly. The list in the BP doc now is better than I came up with!

> * p. 30 - bulleted list of vocabs, consider adding:
> 1) The Vocabulary of Interlinked Datasets (VoID) defines key metadata about RDF datasets. It is intended as a bridge between the publishers and users of RDF data, with applications ranging from data discovery to cataloging and archiving of datasets. Authorities MUST publish a VoID description so others can reuse it.
> 2) The GeoNames Ontology is a geographical database containing over 10 million geographical names.
> 3) The Semantically-Interlinked Online Communities vocabulary (SIOC, pronounced “shock”) is designed for developers to describe information about an online community sites, such as users, posts and forums.

Yes, all those fit the bill.

> ** The word "must" is something I hope we can discuss on tomorrow's telecon vis a vis informative vs. normative guidance.  See IETF RFC 2991,

Yeah I don't think RFC2119 keywords are relevant in that doc - I 
shouldn't have included them.
> Re: Vocab Working Note
> The longer bulleted list that pass your tests and are relevant to eGov, I suggest putting into a Working Note.  I was thinking about our discussion during the F2F about not recommending specific vocabs but rather providing a checklist of things to review and determine whether they are a good fit for a given gov't authority.
> A good way to handle this may be to put in a GLD Working Note.  In a WN, we can say, "We recommend ...<bulleted list>" giving the 'force of law in the absence of other guidance.'  Readers recognize that these vocabs may change over time but we satisfy the recommendations that some of us feel strongly is necessary but keep the people who feel vocabs change and we shouldn't recommend specific ones beyond a very short list (ie., those called out in the charter).
> Thoughts??

It's a tricky one. If we don't mention any vocabs at all then it's 
almost certain that someone will go off and create ex:author somewhere 
rather than use dcterms:creator. But, if we include a list then however 
many times we say "this is not normative or comprehensive, there may be 
more, look at the criteria" - some people will take that list as 
definitive whether it's in a Note or the Rec itself.

Personally I think we're better just to list the criteria and perhaps 
say that DC and maybe one other fit the bill. One compromise might be to 
list some of the things for which we already have vocabularies:
- publishing;
- eCommerce;
- citations and bibliographies;
- licences;
- dataset.

then say "right, now we've given you a steer, good hunting." ??


> On Feb 8, 2012, at 12:24 PM, Phil Archer wrote:
>> Fulfilling my Action-40 [1] in which I was asked to "Reflect on SEMIC advice on vocab selection etc. to see if there is more to contribute to the BP doc" here are some reflections.
>> The attached PDF is a (short) extract from a process document I wrote for the creation of core vocabularies under the EU's ISA programme. The doc itself isn't available publicly, at least partly because it needs updating, but there's no prob sharing this bit which is about vocabulary selection.
>> It sets out some criteria - actually just 3 - for selection and then gives a list of some that meet them. That list as just a modified version of the one in Tom Heath and Chris Bizer's book.
>> HTH
>> Phil.
>> [1]
>> --
>> Phil Archer
>> W3C eGovernment
>> +44 (0)7887 767755
>> @philarcher1
>> <Review Existing Vocabs.pdf>


Phil Archer
W3C eGovernment
+44 (0)7887 767755

Received on Wednesday, 8 February 2012 22:00:35 UTC