[Attempto] executable specifications

[Kenneth Jones]

Norbert,

Thank you for filling in some of the gaps in my knowledge of your work prior
to ACE.  I was aware of your work with Schwitter, but not with Fromherz.  I
have been able to find some of the publications you and he published and
they look like very interesting reading for someone like me who is primarily
interested in using CNLs to write executable specifications.

I too am not sure that ACE has not been applied to specification writing in
industry, that's why I added "as far as I know" to my statement.  Actually,
someone else asserted that "ACE is not used at all in industrial context..."
and asked me why that was the case.  I thought, who better to ask than
Norbert Fuchs and Rolf Schwitter?

Rolf thinks that lack of good tool support is a large factor in the lack of
widespread adoption in industry.  I agree, and it's great to see the work
that you, Tobias, Kaarel, and Rolf are doing to create powerful and
easy-to-use tools.

It's also great to know you are still interested in executable
specifications.

I have a few more questions:

   - I have, for quite some time wanted to read your thesis, but I don't
   know German.  I have tried several of the free translation tools available
   on the web, but none seems to do a very good job of translating from German
   to English.  Do you happen to know of a translator that does a pretty good
   job of German-to-English translation?
   - Are you planning to include a Java API with you upcoming release of
   RACE?
   - What advantages does RACE have over ACE -> TPTP -> theorem prover?

Thanks,

-- Ken

On Sun, Oct 31, 2010 at 11:08 AM, Norbert E. Fuchs <fuchs at ifi.uzh.ch> wrote:

>
> On 30 Oct 2010, at 22:35, Kenneth Jones wrote:
>
> > ACE was initially created as a language for the expression of executable
> software specifications.  You have since turned your attention to other
> potential applications of ACE.  Why is that?
>
> Ken
>
> You are right. Before we started developing ACE in 1995, I worked on
> executable software specifications. See for example
>
>
> http://www.ifi.uzh.ch/rerg/fileadmin/downloads/publications/papers/Executable_Specifications.pdf
>
> During this time my former PhD student Markus Fromherz and I developed a
> graphical specification system where the functionality was expressed in
> Prolog – thus defying our intention to hide formality from the casual user.
> Rolf Schwitter, another former PhD student of mine, and I decided to express
> the functionality instead in natural language – which was the start point of
> ACE as a software specification language. A typical early paper is
>
>  web.science.mq.edu.au/~rolfs/papers/EMISA96.pdf<http://web.science.mq.edu.au/%7Erolfs/papers/EMISA96.pdf>
>
> Working on ACE fascinated us so much and took so much of our time that the
> original graphical specification system was left behind. Later we were
> involved in, and most importantly got funding for, projects concerning the
> semantic web and medicine which led to so many extension of ACE that in the
> end it developed into a general knowledge representation language. My former
> PhD students Kaarel Kaljurand and Tobias Kuhn took fancy to the semantic web
> and developed the tools ACE View and AceWiki.
>
> But executable specifications are not forgotten. During the workshop CNL
> 2010 I gave a tutorial entitled "First-Order Reasoning for Attempto
> Controlled English" where I presented my reasoner RACE (
> http://staff.um.edu.mt/mros1/cnl2010/TALKS/fuchs.pdf). On the last slide I
> described future work among which you also find "executable specifications".
>
> > Also, As far as I know, ACE has not been used in industry to improve the
> quality of specifications.  Why do you think that is?
>
> I do not have complete information for which purposes people have used ACE
> so that I do not know whether your statement is correct.
>
> Regards.
>
>   --- nef
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ifi.uzh.ch/pipermail/attempto/attachments/20101102/c742244e/attachment.htm>