RE: MBUI limitation

Dear all,

It seems to me that UML metamodels and OWL ontologies are located at the
same level of abstraction. In the case of MB-UIDE, the initial UML metamodel
has been transformed into a XML Schema. Several different transformations
are possible for this purpose. OWL is more expressive than a XML Schema
since classes, instances, and properties could be expressed. In addition,
OWL benefits from several mechanisms such as symmetry and inverse
properties, which XML does not.

Second, a UML metamodel could be transformed into OWL. See for instance:
http://topquadrantblog.blogspot.com/2011/02/converting-uml-models-to-owl-par
t-1.html . XML could be also transformed into OWL (see for instance
http://hal.archives-ouvertes.fr/hal-00624055/), but the resulting OWL
ontology is less expressive than the initial UML metamodel.

For me, there are several advantages of having an OWL ontology (as we did in
http://www.w3.org/wiki/images/5/5d/UsiXML_submission_to_W3C.pdf):
- it is a W3C standard that has been largely used in certain domains like
FOAF and Dublin Core
- it is at the same level of abstraction of UML class diagram (which is a
OMG standard)
- it comes already with its own representation, like OWL2 XML syntax (no
problem with different XML syntaxes)
- it supports triple-stores that are supported by various development
environments

My 2 cents,
Jean



Début du message réexpédié :

> Renvoyé-De : public-mbui@w3.org
> De : Fabio Paternò <fabio.paterno@isti.cnr.it> Objet : R: MBUI 
> limitation Date : 20 août 2012 09:37:33 UTC+02:00 À : 'Gerrit Meixner' 
> <Gerrit.Meixner@dfki.de>, 'Dave Raggett' <dsr@w3.org>
> Cc: public-mbui@w3.org
> Répondre à : fabio.paterno@isti.cnr.it
> 
> I doubt that practitioners are interested in a RDF, OWL specification 
> of the task models. I don't know how many interactive application 
> developers use ontologies in their work.
> 
> Of course it is possible to do it but I am not sure about the  
> concrete added value in a similar exercise. XML descriptions in 
> general are easier to understand and manipulate than the others even 
> if they have some limitations in terms of expressiveness.
> 
> 
> Fabio
> 
> 
> 
> -----Messaggio originale-----
> Da: Gerrit Meixner [mailto:Gerrit.Meixner@dfki.de]
> Inviato: lunedì 20 agosto 2012 06:29
> A: 'Dave Raggett'
> Cc: public-mbui@w3.org
> Oggetto: AW: MBUI limitation
> 
> Hi Dave,
> 
> thanks for forwarding the comment by Yan.
> 
> Here I guess we can see the different views people have on a 
> specification again. XML-based languages in MBUID have a long 
> tradition and are well established in this community.
> Also in practice XML-based languages for e.g., infotainment systems 
> are specified in such a way (e.g., OEM-XML, AbstractHMI).
> We had many talks about the different possibilities for using a 
> specific notation. People have arguments for paper-based informal 
> specifications and people have arguments for (more) formal specifications.
> It depends on the user (group) you are asking. Often the least common 
> denominator is a XML-based language. But in the end it doesn't matter 
> which way you go (XML or RDF/OWL) because you will always need 
> powerful tools to support people.
> 
> What do the others think about the comment of Yan?
> 
> Best regards
> Gerrit
> 
> ========================
> Dr.-Ing. Gerrit Meixner
> Head of the Human-Machine-Interaction group
> 
> German Research Center for Artificial Intelligence (DFKI) Innovative 
> Factory Systems (IFS) Trippstadter Strasse 122
> 67663 Kaiserslautern, Germany
> 
> Tel./Fax/Mobile/E-Mail/Web
> +49 (0) 631 / 205 75 3415
> +49 (0) 631 / 205 75 3402
> +49 (0) 157 / 725 95 865
> Gerrit.Meixner@dfki.de
> http://www.dfki.de
> ========================
> Legal statement: 
> Deutsches Forschungszentrum für Künstliche Intelligenz GmbH 
> Trippstadter Strasse 122
> 67663 Kaiserslautern
> 
> Geschäftsführung: 
> 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
> ========================
> -----Ursprüngliche Nachricht-----
> Von: Dave Raggett [mailto:dsr@w3.org]
> Gesendet: Donnerstag, 9. August 2012 21:40
> An: Yan Bodain
> Cc: public-mbui@w3.org
> Betreff: Re: MBUI limitation
> 
> Thank you for your feedback, I am forwarding it to the MBUI list for 
> comments.
> 
> On 04/08/12 22:28, Yan Bodain wrote:
>> Hi
>> I am a software developper with a strong background in Cognitive 
>> Ergonomics (PhD degre from Ecole Polytechnique of Montreal) and also 
>> in semantic technologies.
>> 
>> I am writing this email because I was extremly surprised by the first 
>> draft of MBUI.
>> 
>> I think you have mist a great opportuny to make your model usefull by 
>> using a XML Schema instead of using a combinaison of semantic 
>> descriptions (RDF, OWL).
>> 
>> For example, in many places, you have listed the values that are 
>> legal for some XML attribut. But if you had choosen to link these 
>> values to an ontology class instead, it would have made your model 
>> more usefull (because it would let the system navigate inside a 
>> hiearchy of task and find a task directly related to the local 
>> context or domain)
>> 
>> To resume, the first draft of MBUI is great a excercice for someone 
>> working as ergonomist, cognitive engineering or artificial 
>> intelligence but it as poor value for someone who build software, 
>> portal solution or knowledge system for entreprise.
>> 
>> Coming from the W3C, I was hopping that the MBUI documentation would 
>> not only help ergonomist but also practionner.
>> 
>> If you decide to develop your model further, I will be happy to help 
>> you to migrate your XML model to a semantic one.
>> 
>> Or, at least, find someone who is familiar with semantic technologies 
>> (why not ask Tim Berners-Lee?) in order to validate the context of 
>> application of your MBUI draft.
>> 
>> Best regards
>> Yan Bodain, PhD.
>> 
> 
> 
> 
> 
>

Received on Thursday, 23 August 2012 15:37:04 UTC