Re: [Aperture-devel] vCard Meeting 7/18 - 2 ET?

Hi Bruce,

It was Bruce D'Arcus who said at the right time 24.07.2007 20:10 the 
following words:
>
> On Jul 24, 2007, at 11:04 AM, Leo Sauermann wrote:
>
>>  It was Bruce D'Arcus who said at the right time 23.07.2007 18:17 the 
>> following words:On 7/18/07, Leo Sauermann <leo@gnowsis.com> wrote:
>>>
>>>
>>>>  Aperture will move to the Nepomuk NIE [2] ontology soon, where we 
>>>> have
>>>> further formalized vCard [3], vCalendar, and many other things. We 
>>>> have made
>>>> some suggestions[4] on the calendar-dev mailinglist for needed 
>>>> bugfixes of
>>>> [1], but the list is hibernated, and then we decided to roll our own.
>>>>
>>> Leo -- what does this mean for the ongoing work here on an updated 
>>> vcard-in-rdf?
>>>
>>
>> I cannot say what implications this has for you, because I don't know 
>> your organization nor standardization body nor the work you are doing 
>> in detail.
>>  perhaps you could give me more detail?
>
> I am referring to the work that Norm Walsh, Harry, and Brian Suda 
> first did on an updated vcard-in-rdf ...
>
> <http://www.w3.org/2006/vcard/ns>
>
> ... and the work that Harry and Garrett are now doing to finish that.
ah, ok. good, you should include the URI in all e-mails. :-)

>
> For context about why *I* care about all this, I spent the past 18 
> months developing a new RDF-based metadata system for the OpenDoucment 
> format at OASIS, along with people like Elias Torres.
>
> That work is now done and approved.
>
> Now the obvious question: what now??
>
> I really want to recommend good, clear, stable vocabularies for 
> developers to use. These developers may not have any apriori 
> background or interest in RDF.
we have the same interest and after months of discussion with desktop 
search engine developers, have tried to reach consensus, but never did :-/#
Antoni Mylka discussed with the different developers for months.
It is hard work to explain RDF to developers. Also, the KDE people want 
only to use datatype properties (only literals) whereas Nepomuk needs 
object properties.
In Nepomuk, the project that funds our work, we have to deliver the 
first prototypes in 5 days, and then, evaluate them in a scientific 
experiment (usability studies, desktop search studies, etc). We move 
aperture.sf.net to NIE now and hope that we reach consensus sometimes in 
the future :-/

I would *love* to discuss and merge our two approaches into one, but 
simply said: we cannot do that now.

>
> So I'd like to be able to suggest a representation for agents be used 
> generally in ODF. I'd also like to use pieces of it to represent 
> agents, addresses, and events in the new bibliographic ontology work 
> some of us are involved in, which I also want to use in ODF:
>
>     <http://bibliontology.com>
We haven't focussed on bib in nepomuk yet,
if you convince the bibsonomy.org people to use your ontology, that 
would help to get us in. (they are project member)
:-)
>>  (note: I cannot follow the discussion on semantic-web@w3.org, too 
>> much traffic for me :-/
>>
>>  We created NIE as a standard for the Semantic Desktop, within the 
>> Nepomuk.semanticdesktop.org project, and we plan to implement it with 
>> a reference implementation (aperture.sf.net). Standardization is 
>> first by best practice and community, later we aim for official 
>> standardization bodies (ISO, W3C, OASIS, still an open question).
>
> Good.
>
>>  What may help us both is when you look at NIE [2] and especially the 
>> ontology [5] and if you find bugs, send them to us, if not, just use 
>> NIE as it is.
>>  By this way, we can also both work on Aperture to get good data->rdf 
>> converters.
>>
>>  If you think that NIE is not what you had in mind, you can use 
>> anything else. Ontologies are an open market :-)
>
> But it's not a good market when you 20 different boxes of cereal on 
> the shelf, with little that distinguishes them :-)
There are two extremes here: Dictatorship or Evolution. Either only 
"W3C" approved ontologies are to be used (or OASIS, or ISO, OR 
UN-CEFACT, ) or we wait until evolution shows us what fits the problem 
best. Both is not so fine.
(In Soviet Union they had only one box of cereal)

My suggestion: we ask the W3C RDF Working groups (best 
practices/deployment, SWEO, where I am a member) to start a new working 
group "ontology standardization" which hosts a list of ontologies and a 
mailinglist.
Once anyone has the idea "I want to make a new ontology", this 
mailinglist is the place. ... we should move this to a new thread. So, 
we announce that "I will make an ontology for X" before it happens :-)

best
Leo
>
> Bruce
>


-- 
____________________________________________________
DI Leo Sauermann       http://www.dfki.de/~sauermann 

Deutsches Forschungszentrum fuer 
Kuenstliche Intelligenz DFKI GmbH
Trippstadter Strasse 122
P.O. Box 2080           Fon:   +49 631 20575-116
D-67663 Kaiserslautern  Fax:   +49 631 20575-102
Germany                 Mail:  leo.sauermann@dfki.de

Geschaeftsfuehrung:
Prof.Dr.Dr.h.c.mult. Wolfgang Wahlster (Vorsitzender)
Dr. Walter Olthoff
Vorsitzender des Aufsichtsrats:
Prof. Dr. h.c. Hans A. Aukes
Amtsgericht Kaiserslautern, HRB 2313
____________________________________________________

Received on Wednesday, 25 July 2007 16:17:38 UTC