Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-discussion] DISPLAY_SLOT problem

William Bug William.Bug at DrexelMed.edu
Sat Mar 17 00:56:47 PDT 2007


I hope it can be useful, Tania.

Thanks to all who helped me trouble-shoot the DISPLAY_SLOT problem I  
was having.

One quick correction re: language

I believe I stated the localized language spec (e.g.,  "en-US") was a  
DC Metadata formalism.

In fact, it really derives from RFC 3066 - "Tags for the  
Identification of Languages" (2001)
	http://www.ietf.org/rfc/rfc3066.txt

which the DC metadata language element:
	http://dublincore.org/documents/dces/#language

and the XML Schema language SimpleType
	http://www.w3.org/TR/2004/REC-xmlschema-2-20041028/ 
datatypes.html#language

both conform to.

You might want to ultimately consider redefining the language types  
in the Protege metadata OWL file (http://protege.stanford.edu/plugins/ 
owl/protege) so they conform with this standard.  I would think it  
should be possible to declare the range for this property to be  
xsd:language - i.e.,
	<rdfs:range rdf:resource="http://www.w3.org/2001/XMLSchema#language"/>

Even though that type isn't supported by default in Protege-OWL, it  
certainly is a valid XML Schema type, so the parsers shouldn't choke  
on it.  It's also a supported by the rdfs:range and rdf:datatype  
properties:
	http://www.w3.org/TR/2004/REC-owl-ref-20040210/#DataRange

Then it's just a matter of making certain the Protege code designed  
to parse protege:defaultLanguage is adjusted to process this as  
xsd:language type, as opposed to just xsd:string type.

I suppose this should wait until there is a significant call for this  
language field to support internationalization.

Cheers,
Bill

On Mar 16, 2007, at 5:39 PM, Tania Tudorache wrote:

> Hi Bill,
>
> Thanks a lot for the feedback. I will review the wiki page about  
> hiding the identifiers with labels and update it.
>
> Cheers,
> Tania
>
>
>
> William Bug wrote:
>> Thanks for the feedback, Tania.
>>
>> Please see comments inline below.
>>
>> Cheers,
>> Bill
>>
>> On Mar 14, 2007, at 7:48 PM, Tania Tudorache wrote:
>>
>>> Hi William,
>>>
>>> William Bug wrote:
>>>> Just to clarify the issue for the curious on this list who may  
>>>> have read it, I do believe I understand what the outcome of  
>>>> setting any AnnotationProperty to the DISPLAY_SLOT should be  
>>>> across several Protege-OWL features. My problem comes down to  
>>>> the behavior being different when I set the "LANG" attribute for  
>>>> that AnnotationProperty as opposed to when the LANG attribute is  
>>>> left unset.  When unset, the behavior of several Protege-OWL  
>>>> features to setting an AnnotationProperty other than :NAME or  
>>>> rdfs:label to the DISPLAY_SLOT is as it should be according to  
>>>> my understanding.  When the LANG attribute is set for a given  
>>>> class - to any value, as far as I can tell from my ad hoc  
>>>> testing - then the value of that AnnotationProperty for that  
>>>> class is not as advertised.
>>>
>>> It would help if you could give me an example from the ontologies  
>>> that you have sent in your previous post. If there is a bug in  
>>> the display text of classes when the language is set on the  
>>> browser slot, we should fix it.
>>>
>>>
>>>>
>>>> Regarding the protege:defaultLanguage ontology meta-data  
>>>> AnnotationProperty - thanks for pointing out the inconsistency  
>>>> across our ontology files (the BIRNLex ontology is modularized  
>>>> into a collection of OWL files imported according to a specified  
>>>> hierarchy).  This property turned out to be causing a few  
>>>> problems with various aspects of the integration of the  
>>>> modules.  I tried to fix those problems by including this  
>>>> property in each of the OWL files and setting it to the same  
>>>> value (either 'en' or the more proper according to DC specs 'us- 
>>>> EN').  Neither worked.  If I removed this property from ALL the  
>>>> OWL files, then the import-related problems did go away.
>>>>
>>>
>>> I forgot to mention that you should set the protege:default  
>>> language only on the top level ontology, so on the ontology that  
>>> imports the others. Otherwise, the display text may not work  
>>> correctly. I think this is also documented on the wiki.
>>>
>>> And another thing that I have not tested before is setting the  
>>> language to "us-EN" rather than just "en". I am not sure that  
>>> this works in Protege. You should try first to set the default  
>>> language to "en" and see if it works that way.
>>
>> Yes - this was at the root of the problem.  I figured it out about  
>> a week ago with help from Trish Whetzel on the OBI project.  Once  
>> I set the protege:defaultLanguage for the root ontology to "en",  
>> then the DISPLAY_SLOT assignment worked just fine.
>>
>> The only thing I'd add is the Wiki writeup (http:// 
>> protege.cim3.net/cgi-bin/wiki.pl? 
>> HidingIdentifiersWithLabelsInOWLPlugin - which I'd seen before but  
>> had forgotten about), has one issue that could use to be cleared  
>> up for future users directed to it.  The behavior as described on  
>> that page is not quite correct.  Item 4.3.2 is not the way things  
>> work.  4.3.1 and 4.3.3 are working as advertised, but not 4.3.2.   
>> It's possible if all the classes in all the ontologies imported  
>> via BIRNLex-Main had the LANG attribute for their rdfs:label  
>> property set to 'en', then things would work as described in  
>> 4.3.2.  But this isn't true for some of the external ontologies we  
>> import (e.g., OBI), so it's not really possible to test that right  
>> now.  If having all the LANG attributes of the selected  
>> DISPLAY_SLOT AnntotationProperty set to a single language value IS  
>> a requirement, this should probably be stated on the page.
>>
>> Thanks again for passing on these valuable pointers.
>>
>> By the way, the current root BIRNLex OWL file is available at:
>> http://www.nbirn.net/birnlex/1.2/BIRNLex-Main.owl
>>
>> If you create a dummy Protege-OWL project, import that BIRNLex- 
>> Main.owl file using that URL, then set the DISPLAY_SLOT to  
>> rdfs:label as described on that Wiki page (protege:defaultLanguage  
>> is now set to 'en' in BIRNLex-Main.owl, so you don't have to do  
>> that), then you can see all the classes in BIRNLex-Main and the  
>> several OWL files it imports assigned the rdfs:label in the Class  
>> Browser.
>>
>>>
>>>
>>> Tania
>>>
>>>> Thanks for redirecting me to the proper list.
>>>>
>>>> Cheers,
>>>> Bill
>>>>
>>>> On Mar 9, 2007, at 11:26 AM, Tania Tudorache wrote:
>>>>
>>>>> William,
>>>>>
>>>>> The browser slot is working fine in the OWL ontologies that you  
>>>>> have given as example. What version of Protege did you use? We  
>>>>> recommend that you use the latest release 3.2.1.
>>>>>
>>>>> If you set the browser slot to another slot than the name or  
>>>>> rdfs:label, for example, core:pref-label, then the same rules  
>>>>> for computing the browser text apply as the ones for rdfs:label  
>>>>> and name. I just tested that and it works in Protege 3.2.1.
>>>>>
>>>>> One thing that I've observed in your ontologies, was the the  
>>>>> protege:defaultLanguage was set to "en-US" instead of just  
>>>>> "en". Maybe that was the problem.
>>>>>
>>>>> Please ask in future OWL related questions on the OWL mailing  
>>>>> list.
>>>>>
>>>>> Thanks,
>>>>> Tania
>>>>>
>>>>>
>>>>>
>>>>> William Bug wrote:
>>>>>> Sorry to be posting this a second time, but I'm in the midst  
>>>>>> of a meeting, where it would be really good to be able to  
>>>>>> display the "preferred labels" for all the classes in our OWL  
>>>>>> files.
>>>>>>
>>>>>> Am I getting something wrong here?  Can other folks pick a  
>>>>>> class annotation property as the DISPLAY_SLOT and get it to  
>>>>>> display in the class browser as the class name regardless of  
>>>>>> whether the LANG attribute is set to 'en' for that property or  
>>>>>> not?
>>>>>>
>>>>>> Many thanks ahead of time for any assistance others can  
>>>>>> provide in solving this problem.
>>>>>>
>>>>>> Cheers,
>>>>>> Bill
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Mar 8, 2007, at 1:22 AM, William Bug wrote:
>>>>>>
>>>>>>> Dear Protege Support folks,
>>>>>>>
>>>>>>> We have been having difficulty using the rdfs:label field as  
>>>>>>> a DISPLAY_SLOT, when the language attribute is set to "en".   
>>>>>>> For classes where language is left blank, that field value is  
>>>>>>> used as the display name.
>>>>>>>
>>>>>>> Is this the expected behavior, when setting the DISPLAY_SLOT  
>>>>>>> to a field other than the class :NAME?  I don't know whether  
>>>>>>> this matters, but this for an imported ontology, but I see  
>>>>>>> the same behavior for classes that exist within the OWL file  
>>>>>>> itself.
>>>>>>>
>>>>>>> If you would likely to examine this directly both of these  
>>>>>>> ontologies are available online:
>>>>>>> main ontology file:  http://www.nbirn.net/birnlex/1.2/BIRNLex- 
>>>>>>> Investigation.owl
>>>>>>> imported file: http://obi.sourceforge.net/ontology/OBI.owl
>>>>>>>
>>>>>>> I've attached two screen shots below demonstrating what I  
>>>>>>> mean.  These are illustrative examples, but this behavior is  
>>>>>>> observed across the entire ontology.
>>>>>>>
>>>>>>> Many thanks for your help with this issue.
>>>>>>>
>>>>>>> Cheers,
>>>>>>> Bill
>>>>>>>
>>>>>>> <Picture 15.png>
>>>>>>> <Picture 16.png>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Bill Bug
>>>>>>> Senior Research Analyst/Ontological Engineer
>>>>>>>
>>>>>>> Laboratory for Bioimaging  & Anatomical Informatics
>>>>>>> www.neuroterrain.org
>>>>>>> Department of Neurobiology & Anatomy
>>>>>>> Drexel University College of Medicine
>>>>>>> 2900 Queen Lane
>>>>>>> Philadelphia, PA    19129
>>>>>>> 215 991 8430 (ph)
>>>>>>> 610 457 0443 (mobile)
>>>>>>> 215 843 9367 (fax)
>>>>>>>
>>>>>>>
>>>>>>> Please Note: I now have a new email -  
>>>>>>> William.Bug at DrexelMed.edu <mailto:William.Bug at DrexelMed.edu>  
>>>>>>> <mailto:William.Bug at DrexelMed.edu>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> <Picture 15.png>
>>>>>>> <Picture 16.png>
>>>>>>> _______________________________________________
>>>>>>> protege-discussion mailing list
>>>>>>> protege-discussion at lists.stanford.edu <mailto:protege- 
>>>>>>> discussion at lists.stanford.edu> <mailto:protege- 
>>>>>>> discussion at lists.stanford.edu>
>>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>>>
>>>>>>> Instructions for unsubscribing: http://protege.stanford.edu/ 
>>>>>>> doc/faq.html#01a.03
>>>>>>
>>>>>>
>>>>>>
>>>>>> Bill Bug
>>>>>> Senior Research Analyst/Ontological Engineer
>>>>>>
>>>>>> Laboratory for Bioimaging  & Anatomical Informatics
>>>>>> www.neuroterrain.org
>>>>>> Department of Neurobiology & Anatomy
>>>>>> Drexel University College of Medicine
>>>>>> 2900 Queen Lane
>>>>>> Philadelphia, PA    19129
>>>>>> 215 991 8430 (ph)
>>>>>> 610 457 0443 (mobile)
>>>>>> 215 843 9367 (fax)
>>>>>>
>>>>>>
>>>>>> Please Note: I now have a new email -  
>>>>>> William.Bug at DrexelMed.edu <mailto:William.Bug at DrexelMed.edu>  
>>>>>> <mailto:William.Bug at DrexelMed.edu>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> ----------------------------------------------------------------- 
>>>>>> -------
>>>>>>
>>>>>> _______________________________________________
>>>>>> protege-discussion mailing list
>>>>>> protege-discussion at lists.stanford.edu <mailto:protege- 
>>>>>> discussion at lists.stanford.edu>
>>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>>
>>>>>> Instructions for unsubscribing: http://protege.stanford.edu/ 
>>>>>> doc/faq.html#01a.03
>>>>>
>>>>> _______________________________________________
>>>>> protege-discussion mailing list
>>>>> protege-discussion at lists.stanford.edu <mailto:protege- 
>>>>> discussion at lists.stanford.edu>
>>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>>
>>>>> Instructions for unsubscribing: http://protege.stanford.edu/doc/ 
>>>>> faq.html#01a.03
>>>>
>>>>
>>>>
>>>>
>>>> Bill Bug
>>>> Senior Research Analyst/Ontological Engineer
>>>>
>>>> Laboratory for Bioimaging  & Anatomical Informatics
>>>> www.neuroterrain.org
>>>> Department of Neurobiology & Anatomy
>>>> Drexel University College of Medicine
>>>> 2900 Queen Lane
>>>> Philadelphia, PA    19129
>>>> 215 991 8430 (ph)
>>>> 610 457 0443 (mobile)
>>>> 215 843 9367 (fax)
>>>>
>>>>
>>>> Please Note: I now have a new email - William.Bug at DrexelMed.edu  
>>>> <mailto:William.Bug at DrexelMed.edu>
>>>>
>>>>
>>>>
>>>>
>>>
>>
>> Bill Bug
>> Senior Research Analyst/Ontological Engineer
>>
>> Laboratory for Bioimaging  & Anatomical Informatics
>> www.neuroterrain.org
>> Department of Neurobiology & Anatomy
>> Drexel University College of Medicine
>> 2900 Queen Lane
>> Philadelphia, PA    19129
>> 215 991 8430 (ph)
>> 610 457 0443 (mobile)
>> 215 843 9367 (fax)
>>
>>
>> Please Note: I now have a new email - William.Bug at DrexelMed.edu  
>> <mailto:William.Bug at DrexelMed.edu>
>>
>>
>>
>>
>

Bill Bug
Senior Research Analyst/Ontological Engineer

Laboratory for Bioimaging  & Anatomical Informatics
www.neuroterrain.org
Department of Neurobiology & Anatomy
Drexel University College of Medicine
2900 Queen Lane
Philadelphia, PA    19129
215 991 8430 (ph)
610 457 0443 (mobile)
215 843 9367 (fax)


Please Note: I now have a new email - William.Bug at DrexelMed.edu




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/protege-discussion/attachments/20070317/751dead9/attachment.html>


More information about the protege-discussion mailing list