Pat, (01)
Thank you for all the good work. (02)
> But the latest version of Protege apparently places some kind of
restraint on the format of :PAL-CONSTRAINTs, and
> the axioms did not load properly. I will have to change the class
for these axioms to something else that doesn't
> mangle the text. This will be simple enough, but in the present
Protege files on the MICRA site, the axioms
> don't appear in the current import. (03)
Will you kindly take the work past this step (i.e. get the axioms into
Protege, even if they cannot be processed), in the next day or two. That
way, others could pick up from there. Thanks in advance. (04)
> I will try to find some time to resume work on the translation, ... (05)
Will sincerely look forward to that. Keep us posted. (06)
Best of luck to your new job! (07)
Regards. -ppy
-- (08)
Patrick Cassidy wrote Tue, 16 Nov 2004 00:07:06 -0500: (09)
> Concerning translations to other formats: (010)
> The translation from KIF to Protege uses certain devices to work
> around the lack of axiom support in Protege and the lack of support
> for higher-arity relations. For the axioms, in past versions I had
> simply recorded KIF axioms as instances of a subclass of
> :PAL-CONSTRAINT (called :SKIF-AXIOM). (011)
> But the latest version of Protege apparently places some kind of
> restraint on the format of :PAL-CONSTRAINTs, and the axioms did not
> load properly. I will have to change the class for these axioms to
> something else that doesn't mangle the text. This will be simple
> enough, but in the present Protege files on the MICRA site, the axioms
> don't appear in the current import. (012)
> As for the higher-arity relations, instances of the relations are
> visualized as instances of the metaclasses :Proposition or
> :Assertion; and the relation definitions are present as slots with
> added facets which specify the restrictions on the argument types.
> However, unlike with ordinary (binary) slots, there is no way at
> present to specify restrictions on arguments other than the last
> ("value") argument, when those restrictions differ on subclasses of
> the highest-level class on which the relation is defined. This would
> require additional programming, probably overriding existing Protege
> functions, and is not likely to be implemented in the foreseeable future. (013)
> In a discussion of the KIF-Protege translation with Holger
> Knublauch, he suggested that it would be helpful to have an OWL base
> ontology which includes built-in definitions of the Protege classes
> used to work around these problems. It sounds like a way to get
> translation from SKIF to OWL, and also to be able to use the Protege
> visualization and editing tool. I regrettably have much less time to
> do any work on that now, and am not sure when if ever I will be able
> to add any enhancements to the existing import plug-in. As I have
> mentioned before, there is still no export functions, so Protege
> cannot be used with the current import plug-in as an accurate editing
> environment for SKIF ontologies. (014)
> I will try to find some time to resume work on the translation, but
> can't predict when any more progress will be made. For the time
> being, I can import SKIF ontologies into Protege, but only for the
> purpose of visualization.
>
> Pat (015)
> Peter P. Yim wrote Sun, 14 Nov 2004 22:29:27 -0800:
>
>> Thanks, Pat.
>> I assume those who are making translations from Protege (to other
>> formats) could start now ... please post any problems encountered so
>> that we can collaboratively resolve them. It's only 12 more working
>> days to the EIDX panel presentation. Thanks in advance.
>>
>> -ppy
>> --
>>
> (016)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Unsubscribe/Config:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (017)
|