Paola, (01)
As a counter example, we have customers who want to ensure that certain
ontologies (even "open source" ontologies) that they elect to depend on
are developed and managed in processes similar to those of typical
standards bodies. For instance, at OMG, there is work underway to
develop an SBVR vocabulary to represent dates and times (in the SBVR/UML
modeling standard, and based on numerous standards and related papers,
including Jerry Hobbs' ontology, Pat's catalog, several ISO standards,
etc.), which we plan to map to OWL. We would like to have a repository
at OMG to host and manage all of the relevant artifacts together -- the
documentation, SBVR models, and any additional relevant artifacts,
including the OWL representation, which will be lossy, but useful
nonetheless. We believe that this repository, which we would like to
make generally searchable, will be an important asset to the broader
community (OMG members or otherwise). (02)
At our December OMG meeting in Santa Clara, we heard presentations on
Stanford's BioPortal and Collaborative Protege, as well as on a new
repository effort at GSA, and will be pulling a wiki together over the
coming months to develop requirements for building this out. The
BioPortal approach is appealing for a number of reasons, particularly to
those of us who are familiar with Stanford's work. Other ideas include
use of a back-end MOF repository, which would allow the UML models to be
searchable, also very appealing from an OMG perspective. (03)
The repository will be focused on standards produced by the OMG
community, including potentially vocabularies or ontologies that make
sense for that community to author and publish, with an emphasis on
relating all of the relevant artifacts to one another and managing them
as elements of the standards. It is not intended to provide a more
general catalog of ontologies on the web that might be useful, though --
only those that have been developed through the OMG standards process. (04)
I would argue that as we develop validation/vetting/best practices
towards developing standard vocabularies at OMG, it may be that those
processes are more valuable to the community than the repository
initially, but over time, a well managed repository of standard
vocabularies whose representational forms include UML models, ODM/OWL
ontologies, ODM/RDF vocabularies, ODM/CL ontologies, SBVR vocabularies,
and so forth, will indeed have value to commercial and government
organizations, especially those who are leery of depending on ontologies
that have not been evaluated/vetted as rigorously. They will likely
only be useful for certain contexts/applications as most models tend to
be, with potentially lossy mappings among the artifacts, and thus the
documentation defining the intended context and mapping coverage will
also be important. They may also provide the basis for additional
research, but are likely to be somewhat behind the proverbial research
curve, by nature. Those artifacts that would typically live in the web,
such as RDF vocabularies, OWL ontologies, and XMI documents, will be
made available through traditional web access methods, but they would be
managed in the repository with the other relevant model artifacts.
Mechanisms allowing community input on model utility, enhancement
requests, and so on are also of interest to us -- inspired by the
Collaborative Protege presentation. Unless a particular vocabulary or
ontology is sponsored by a group of devoted developers, who are
committed to their user community, and listen and respond to feedback in
a timely manner, I find it difficult to believe that it will be long
lived, hanging out on the web on its own. (05)
Best regards, (06)
Elisa (07)
paola.dimaio@xxxxxxxxx wrote:
> John, and all
>
> Of course, like many others I am interested, but some questions were
> posed at the last summit, where some ideas related to the repository
> project were first asked, and never answered
> at least not on the public list
>
> I remember, for example, that my first reaction to the ideas presented
> by Mike was
> 'it is so un-LOD' , meaning: ontologies live on the web, the web is
> the natural repository.
>
> (ref. my question at the end of mike dean's presentation on the record)
>
> To create 'another' repository would be redundant, unnecessary, a
> waste of resources and unsustainable. (I am sure the above statement
> is debatable but in short the above are some key points that have
> never been discussed)
>
> To produce a, say, directory of all existing ontologies on the web it
> is essential that each resource owner/administrator produces some uri
> with some metadata, and updates that (as being discussed in parallel
> in other lists), and submits it to the directory index
>
> A welll maintained directory could be easily used as a knowledge base
> for individual developers to produce queries, scripts and apis to do
> what they want with the ontologies, including knitting conceptual and
> semantic lace from
>
> Now, that's probably more realistic 'semantic' view of the repository world
>
> Thus, assuming any of the above points is valid,
> the repository effort is going to need some serious rethinking
>
>
> pdm
>
> On Tue, Jan 13, 2009 at 3:45 AM, John Graybeal <graybeal@xxxxxxxxx> wrote:
>
>> There are existing projects that are geared toward meeting both these
>> goals, are there not? Many that are putting forth semantic wikis (for
>> the purpose of defining semantic concepts in a wiki-like way) and a
>> few projects that are targeted (broadly) at a more formal ontology
>> presentation space for community ontologies.
>>
>> I don't have examples of the first in hand (many are known), but in
>> the second we are describing Knoodl (Revelytix), NeOn's work, and the
>> (early stage discussions) Open Ontology Repository project (by Ontolog
>> group, previously mentioned in the thread). The last is noteworthy
>> because many requirements have been defined in public pages.
>>
>> I ask because I'm not sure why this group is devoting time discussing
>> design of a system, when the interested parties might instead agree on
>> basic goals, pick a system, and start work? Or else I am missing
>> something.
>>
>> John
>>
>>
>> On Jan 12, 2009, at 6:58 AM, Ron Wheeler wrote:
>>
>>
>>> I would suggest a 2 pronged approach. The formal ontology repository
>>> should have a proper governance structure and peer review by whatever
>>> body is setup to do that.
>>>
>>> The wiki should be more like Wikipedia with the emphasis on collecting
>>> ontologies and building up a set of documentation about each one,
>>> comments from users, links to compatible ontologies, links to
>>> alternatives and comments from reviewers regardless of their
>>> "officialness".
>>>
>>> The formal repository governing body should find this a useful
>>> resource
>>> both as a source of candidate ontologies and as a source of potential
>>> SMEs and reviewers. It will also identify topics and ideas that the
>>> official reviewers may want to include in their analysis.
>>>
>>> The less bureaucracy in the wiki, the better. It has worked very well
>>> for Wikipedia.
>>> I doubt if we would have more vandalism than Wikipedia does,
>>> although we
>>> do get some heated discussion here.
>>> If it does become a problem, the easiest way to fix that is by
>>> requiring
>>> people to get permission to have access to writing.
>>> Wikipedia has not had to resort to that and they draw from a much
>>> wider
>>> audience with all kinds of commercial and competitive interests.
>>>
>>> Ron
>>>
>>>
>>> John F. Sowa wrote:
>>>
>>>> Azamat and Ron,
>>>>
>>>> There are two separate issues:
>>>>
>>>> 1. Developing the ground rules and policies for an ontology
>>>> registry.
>>>>
>>>> 2. Setting up a registry and maintaining the contributed ontologies.
>>>>
>>>> These two goals can be pursued in parallel, but #1 should be started
>>>> first. Then an implementation, #2, would give us further experience
>>>> and ideas about how to develop #1 further.
>>>>
>>>>
>> _________________________________________________________________
>> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
>> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
>> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
>> Shared Files: http://ontolog.cim3.net/file/
>> Community Wiki: http://ontolog.cim3.net/wiki/
>> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
>> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>>
>>
>>
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
>
>
> (08)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|