[kepler-dev] RE: [seek-kr-sms] computation ontologies

[Matt Jones]

Hi,

Although customization is a good feature, I thought that Rich was 
reiterating the point that we had arrived at in earlier discussions that 
the best default view is a domain-specific view.  That's why we had the 
idea originally of having an ontology view that was 'switchable', and 
our original design had a preference dialog tab that allowed the user to 
choose their preferred view based on domain (e.g., the 'Ecology' view, 
the "Geoscience" view, the 'Electrical engineering' view, etc.).

For details of our previous discussion, see pages 8 and 9 of the meeting 
notes:

http://cvs.ecoinformatics.org/cvs/cvsweb.cgi/~checkout~/kepler/docs/dev/meeting-notes/kepler-sba-mtg-20040602.pdf?rev=1.1&content-type=application/pdf

This is sort fo a middle ground where we provide several classifications 
and the user chooses among them without the ability to further customize 
(at least at first).  Presumably such a mechanism could be extended to 
allow the user to provide their own classifications, possibly by 
modifying an existing one (I could envision a bookmark manager sort of 
interface built right into the tree on the left ('Add new subcategory', 
'Delete subcategory', with DnD operations to put actors in categories 
[which actually would just annotate the actor with additional category 
metadata]).

Matt

Chad Berkley wrote:
> Hey Rich,
> 
> As I have been trying to classify the library, I've mostly come to the 
> conclusion that your ideas on multiple classifications are correct.  My 
> classification seems just as arbitrary as the one that is already there. 
>  Mark and I are going to try to come up with an ecology-based 
> categorization next week.  Hopefully that will at least make more sense 
> to ecologists.
> 
> I'm kind of thinking right now that after we get a decent hierarchy 
> working for the original actors, the best thing we could work on after 
> that is allowing the user to customize that classification to his/her 
> own needs.  It would be interresting to allow a user to upload his own 
> classification to the grid to allow others to use it, although I'm not 
> sure how many people would actually be willing to put in the time to 
> customize their tree.
> 
> Hopefully Mark and I can come up with at least a preliminary ontology 
> for the library next week so we can have something working by october 15.
> 
> Thanks for your 2 cents.
> 
> chad
> 
> Rich Williams wrote:
> 
>> Hey Chad -
>>
>> I haven't worked on anything like an ontology of computation.  I've done
>> some work a number of ontologies that might be of interest:  1) an 
>> ontology
>> of statistical concepts, so that could be of use for describing actors 
>> that
>> implement statistical methods  2) an ontology of biodiversity indices 
>> and 3)
>> an ontology of ecological models.  But I suspect that these are more
>> high-level than the kind of granularity you will need to classify the 
>> actor
>> library as I know it.
>>
>> Of course in some sense the actor library is already classified, as the
>> existing directory structure can be seen as a hierarchical classification
>> scheme.  Perhaps the problem is that ecologists and ecoinformatics people
>> don't like the classification scheme of the engineers and computer
>> scientists.  One issue I've thought of in this is that there will 
>> probably
>> need to be multiple classification schemes, and that the 
>> classification of
>> most interest to an end-user will depend on the domain expertise of that
>> user and on the problem they are addressing.  For example, GARP is an
>> ecological method, of interest when looking at spatial distributions of
>> species on a landscape.  It is also a statistical method and could be 
>> of use
>> in domains other than ecology and problems other than spatial 
>> distributions.
>>
>> Rich
>>
>>
>>> -----Original Message-----
>>> From: seek-kr-sms-admin at ecoinformatics.org
>>> [mailto:seek-kr-sms-admin at ecoinformatics.org]On Behalf Of Chad Berkley
>>> Sent: Wednesday, September 15, 2004 9:59 AM
>>> To: rich at sfsu.edu; Shawn Bowers; Matt Jones;
>>> kepler-dev at ecoinformatics.org; seek-kr-sms at ecoinformatics.org
>>> Subject: [seek-kr-sms] computation ontologies
>>>
>>>
>>> Hi Rich,
>>>
>>> Shawn, Matt and I were having a conversation on IRC this morning about
>>> ontologies in Kepler.  For the last couple days, I've been trying to
>>> classify the actors in the actor library.  Shawn said you may have some
>>> sort of computation ontology that might be helpful in doing this.  Do
>>> you have anything of that sort?  Do you have any recommendations on how
>>> I (we) should go about doing this?  The goal is to have (at least) a
>>> preliminary categorization of the actors for the 10/15 kepler release.
>>> Shawn has gotten Jena working in kepler so we now have an ontology based
>>> search engine, but no real ontology to search with.  If you have any
>>> ideas on this, we'd be stoked to hear them.
>>>
>>> thanks,
>>> chad
>>> _______________________________________________
>>> seek-kr-sms mailing list
>>> seek-kr-sms at ecoinformatics.org
>>> http://www.ecoinformatics.org/mailman/listinfo/seek-kr-sms
>>
>>
>>
>> _______________________________________________
>> kepler-dev mailing list
>> kepler-dev at ecoinformatics.org
>> http://www.ecoinformatics.org/mailman/listinfo/kepler-dev

-- 
-------------------------------------------------------------------
Matt Jones                                     jones at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Fax: 425-920-2439    Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)
University of California Santa Barbara
Interested in ecological informatics? http://www.ecoinformatics.org
-------------------------------------------------------------------