OXFORD UNIVERSITY  COMPUTING LABORATORY

Hello,

If you're reading this, my mailing list set-up has worked.  Welcome to
the CZT (Community Z Tools) mailing list.  A separate message contains
administrative information.  I've added you to the list because you
expressed interest in the proposal.

Responses

Thank you to those who have responded: it's not a torrent, but
hopefully there's enough interest to achieve something useful.  Here
is a summary of the emailed responses.  [There are a few other
supportive comments I've received too, which I'll not include.]

  •  Jin Song Dong expressed a willingness to be involved for him and his
      students, based on their work with Z and XML.

http://nt-appn.comp.nus.edu.sg/fm/zml/

  •  Rob Arthan said

I am intending to make ProofPower available open source in
the not too distant future and the Z proof support would provide one
option for [proof tool component]

  He said that support for Unicode characters should be included (the
  Standard now defines them for Z).  He also raised the issue of the
  choice of software licences (see below).

  • * David Alex Lamb offered effort and experience with Java,
      tool interoperability, data interchange, etc.  He's especially interested
      in getting the design right.
  • * Don Sanella wished us well, remarking on the similarity with the CoFI
      project in the algebraic specification community.  He also said that
      "coordinating this kind of thing is a lot of work"!
  • * Ted Pearson is working on integration of Z and UML, and tools.  He
      has a prototype tool and is interested in participation with a view
      to improving integration of formal methods with UML and best
      Software Engineering Practice.
  • * I have one or two students who might work on aspects of the tools in
      future.

Live Issues

An early issue to resolve is the one of copyright.  I am evidently out
of date in my understanding.  The GPL may not be suitable, the BSD
licence has much to commend it.  Rob suggested:

   In practice, it may work out OK to let contributors choose their
   own licence terms subject to Open Source Initiative approval and a
   preference for terms which don't restrict selling on in the way
   that GPL does.

Another thing to do is to identify existing components to build on.
Besides Rob's mention of ProofPower, and Jin Song's XSLT scripts,
another project worth looking closely at is ZETA:

http://uebb.cs.tu-berlin.de/zeta

I have tried to contact the project team, but heard nothing back so far.

What next

1. Perhaps you have had further thoughts since you wrote previously;
   perhaps I have badly summarised your response above.  Please
   discuss on this mailing list.
  

2. How many people does it take to make a viable project?  It would be
   good to get a few more people involved.  I will post a follow-up
   article on comp.specification.z mentioning this mailing list.  Who
   else should be contacted?  Are there mailing lists that should be
   included?

3. Managing this project sounds ... a little daunting.  If you know of
   anyone with experience of a similar project, I'd be interested to
   get their advice.

4. A project plan and outline design/specification/requirements document(s)
   are the next things to write.  Who would like to help draft these?

Andrew.

Random Image
Random Image
Random Image