Re: SW Graphical Notation

I did not know TBC had added that facility, it seems quite handy for doing
minor changes.  I find that UMLish notation fairly good, and I am one of
the people who use it from time to time on slides for visualizing
ontologies.

> Graphical notations IMHO cannot scale to editing large ontologies and -
as you state - cannot efficiently capture all semantic constructs of OWL
(or SPIN or whatever).

It depends on how large 'large' is. We have good success building
ontologies with on the order of 1500 classes and 700 properties. I have
never felt the urge to author in another manner.  It might even work if I
was building an ontology that was one or more orders of magnitude larger,
but I have not thought about it.  What does not scale so well is searching
and navigating in a purely visual environment, but we don't do that, we use
a hybrid approach.   We export the Visio diagram to OWL and load it into
Protege or Topbraid Composer.


On Sun, Jun 30, 2013 at 12:29 AM, Holger Knublauch <holger@knublauch.com>wrote:

>  On 6/30/2013 4:51, Michael F Uschold wrote:
>
>  I agree that it is probably too early to have a standard, we should
> first have some good candidates that have been used for a while and think
> about standards afterwards.
>
>  Note that there is a big difference between a graphical tool for
> authoring vs. purely visualizing.  Tools for the latter normally use a lot
> of automation to get good layouts and there is usually no attempt to
> completely show all the axioms which are needed for authoring.  There are a
> lot of tools for visualizing, but they all have major shortcomings, and
> none are very adequate overall.
>
>  I am not aware of any commercial tools that are widely available and in
> use for authoring OWL.   I used a tool from Cerebra some years ago, but
> they were acquired by Web Methods and that tool is not available.  It may
> be getting use in house.  Gary Ng was one of the developers.  Sandpiper
> Software has been preparing one for a while called the Visual Ontology
> Modeller, but it does not yet seem to be available for purchase on their
> web site.
>
>  My colleague at Semantic Arts, Simon Robe, gave a short talk at SemTech
> in S.F. a few weeks ago about *Graphically Authoring OWL Ontologies in
> Microsoft Visio.*   He developed a tool we use in house for authoring OWL
> enterprise ontologies for large companies.  If printed out using a plotter
> and put on a wall, they would be about ten feet high and anywhere from ten
> to twenty feed wide.
> *
> *
> I have been using this tool for  the last few years and find it very
> convenient.  While there are pros and cons, I have come to much prefer the
> graphical authoring approach compared to using tools like Protégé or
> Topbraid Composer.  So I author using our Visio plugin and use Protégé
> and/or Topbraid Composer mainly for viewing and querying in SPARQL.   It is
> not for sale at this time, but it freely available to our clients.
>
>
> Hi Michael,
>
> are you aware that TopBraid Composer's Diagram tab is also editable, see
>
>
> http://composing-the-semantic-web.blogspot.com.au/2012/06/graphical-ontology-editing-with.html
>
> (it used to be a pure viewer until mid 2012). The notation we use there is
> close to UML and should therefore be familiar to most users. I definitely
> wouldn't claim it's perfect - we never invested enough time on this kind
> graphical editing, because it is limited to certain scenarios only:
> Graphical notations IMHO cannot scale to editing large ontologies and - as
> you state - cannot efficiently capture all semantic constructs of OWL (or
> SPIN or whatever). However, this kind of visualization is useful to get
> quick overviews of ontologies that have been given to you, and are good to
> create outlines that are later refined with form-based editing. TBC users
> can switch back and forth between the two notations, because the diagrams
> are generated on the fly from the current context.
>
> Holger
>
>
>
>
>   The original goal was to have a format for showing our clients their
> enterprise ontologies in a way that offered much greater than epsilon
> probability that they could understand what was going on without a lot of
> formal training.  We have had some success with this with the more
> technically oriented folk.  We have developed a variety of other techniques
> for explaining and socializing enterprise ontologies that we use.  Some of
> this is described in an invited talk (In the Trenches with Enterprise
> Ontology <http://www.youtube.com/watch?v=lbR8UzlF9B8>) I gave in Galway
> in October 2012 at the European Knowledge Acquisition Workshop.
>
>  Michael
>
>  ABSTRACT: Graphically Authoring Ontologies in OWL
> This tool supports drag-and-drop stencil-based OWL ontology development
> and is intended to complement existing IDE-style ontology editors such as
> Protégé. Our goal was for the diagramming notation to be sufficiently
> compact and intuitive to allow the review of ontologies with subject matter
> experts after only fifteen minutes of training in the conventions. The tool
> has been successfully used to author and review several large and complex
> enterprise ontologies. It has also proved valuable as a simple and
> intuitive training tool for new ontologists. The demo will show the
> creation, serialization and validation of an OWL ontology from within the
> familiar Microsoft Visio environment using the e6tOWL toolset. -
>
>
>
>
>
> On Fri, Jun 28, 2013 at 10:36 AM, Daniel Schwabe <dschwabe@inf.puc-rio.br>wrote:
>
>> Dear all,
>> I agree with Peter's statement. Regarding graphical notations, I would
>> like to call everybody's attention to the work of Daniel Moody [1], an
>> excellent work setting the foundations of why, what and how graphical
>> notations can be helpful. It should certainly be taken into account if such
>> a notation should be proposed, but I also agree that at the current stage
>> it is still more a research problem than a standardization problem.
>>
>>
>>  [1] Daniel L. Moody: The “Physics” of Notations: Toward a Scientific
>> Basis for Constructing Visual Notations in Software Engineering. IEEE
>> Trans. Software Eng. 35(6): 756-779 (2009). Pdf at
>> http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=05353439.
>>
>>
>>  On Jun 28, 2013, at 01:39  - 28/06/13, "Peter F. Patel-Schneider" <
>> pfpschneider@gmail.com> wrote:
>>
>>  I'm having trouble understanding the purpose of having a
>> common/accepted visual representation of an ontology.  Would this
>> representation help in standardizing ontologies?  Would it help in
>> transmitting ontologies?  Would tools be required to consume it?
>>
>>  This is not to say that there is not a pressing need for more ontology
>> visualization tools.  On the contrary, every time I look at ontologies of
>> any size, I become depressed at how bad ontology visualization tools are.
>>  (Of course, what I want is to see just what I need to see, arranged in
>> just the way that makes it easiest for me to understand aspects of the
>> ontology that I understand.)  This seems to point out a need for research,
>> not standardization, however.
>>
>>  Peter F. Patel-Schneider
>>
>>  On Jun 27, 2013, at 8:39 PM, Renato Iannella <ri@semanticidentity.com>
>> wrote:
>>
>>
>>  RDF/OWL have well-defined technical encodings (xml, turtle, etc) but
>> there is no such common/accepted representation for a graphical notation.
>> That is, a visual representation of an ontology that captures
>> (graphically) all the semantics of RDF/OWL.
>>
>>  I have collected a few examples of various graphical notations here:
>> http://www.w3.org/wiki/SemWebGraphicalNotation
>>
>>  Is there any interest from members of the SWIG to look at this in more
>> detail, and potentially propose such graphical notation for RDF/OWL?
>> (This could be via this IG or a new Community Group.)
>>
>>   Cheers...
>> Renato Iannella
>> Semantic Identity
>> http://semanticidentity.com
>> Mobile: +61 4 1313 2206 <%2B61%204%201313%202206>
>>
>>
>>
>>   []s
>> D
>>
>>
>
>


-- 

Michael Uschold
   Senior Ontology Consultant, Semantic Arts
   http://www.semanticarts.com
   LinkedIn: http://tr.im/limfu
   Skype, Twitter: UscholdM

Received on Monday, 1 July 2013 01:16:40 UTC