Ed Willink responded below, but his message bounced because of an address change.
Ed's message is below -Christopher > -----Original Message----- > From: shirin mohseni [mailto:[EMAIL PROTECTED]] > Sent: Wednesday, January 01, 2003 12:25 PM Hi Shirin > Anyone knows about WDL. I am the (primary) author of the original WDL proposal. The proposal work was carried out under a UK defence contract, and while the result exceeded the customers expectations, competitive tendering for the Phase 2 development work resulted in the associated demonstrators dominating the costs and so the contract moved out of my influence. I understand that the idea of a definitive specification language has been abandoned in favour of a pragmatic babel of pre-existing 'languages'. Since these lack formal definitions or intregration, it is unclear how the goal of predictable and portable behaviour will now be achieved. The Phase 2 work is not available on the web although the programme is 'open'. > There is some documents about it, Indeed: http://www.computing.surrey.ac.uk/personal/pg/E.Willink/wdl/wdl.html > but I want to know if any compiler for WDL exists? No. Though there is a yacc grammar for the language, which you could have but without any other tool support it is useless. Ptolemy II is probably the closest you will get today. But there is quite a gap between the degree of abstraction available in Ptolemy and that advocated in WDL. > and any sample file? http://www.computing.surrey.ac.uk/personal/pg/E.Willink/wdl/documents/Fm3tr. pdf is a substantial example of the way that a hierarchical behavioural decomposition into statecharts and message flow diagrams ultimately hitting a leaf definition lantuage works. All readers have felt that the example is much more readable than the original, as well as being potentially formal and executable. WDL gathered together a number of research ideas: Arbitrary hierarchical decomposition has long been a UCB research area. Recent successess with modal models demonstrates the viability of this. A definitive leaf specification language is now being pursued by the Caltrop and Compact activities. The Caltrop language has many similarities to the WDL leaf language but is rather better researched. WDL used heterogeneous composition, allowing events and tokens to be used in a WDL specification domain that abstracted away from the concrete Ptolemy implementation domains. Heterogeneous composition is a powerful extension providing major opportunities for intellectual spaghetti. A recent ptresearch thread highlights the restristiveness of strictly homogeneous composition. WDL was originally envisaged as an abstract domain that could be translated into concrete domains as part of the Ptolemy composition process. The difficulties of producing production quality code within Ptolemy II and the flexibility with which XSL allows XML notations to be transformed, leads me to believe that a WDL-like approach should be resolved in front of Ptolemy so that all the many transformations from abstract specification concepts can be translated away as the application is reshaped into a target(s)-specific form. Use of a Ptolemy simulation is just another target, but one for which generation of application-, domain-, type-specific composite actors can dramatically reduce simulation times. You may find my joint paper on Caltrop at the OOPSLA MDA workshop relevant. http://www.softmetaware.com/oopsla2002/willinke.pdf I hope that the differing actor-oriented and class-oriented perspectives on transforming specifications into implementations can be abstracted to support a more general tool that support transformation of the eccentric characteristics of almost arbitrary domain-specific languages via relatively neutral expression and statement representations to effective implementations. WDL and its successors will then just be domain-specific languages. Regards Ed Willink ------------------------------------------------------------------------ E.D.Willink, Email: mailto:[EMAIL PROTECTED] Thales Research and Technology (UK) Ltd, Tel: +44 118 923 8278 (direct) Worton Drive, or +44 118 986 8601 (ext 8278) Worton Grange Business Park, Fax: +44 118 923 8399 Reading, RG2 0SB ENGLAND http://www.computing.surrey.ac.uk/personal/pg/E.Willink ------------------------------------------------------------------------ (formerly Racal Research and Thomson-CSF) As the originator, I grant the addressee permission to divulge this email and any files transmitted with it to relevant third parties. ---------------------------------------------------------------------------- Posted to the ptolemy-hackers mailing list. Please send administrative mail for this list to: [EMAIL PROTECTED]
