Re: Lisbon meeting -> schedule of work!

["Peter D. Mosses" <pdm@daimi.aau.dk>]

Andrzej wrote:

> Dear Semantics People,
> ...
> Schedule
> ========
> 
> The main goal is to produce semantics for CASL. As you all know, CASL
> v.0.99 is to be officially released any minute now, 

:-)  That seemed almost like an invariant these past days...

> with a hope that there will be only very minor adjustments (if any
> at all) to obtain v.1.0.

The syntax and semantics of views are still quite tentative, since the
changes agreed in Lisbon had an impact on the way that generic views
had been provided.  (The discussion of the detailed design better be
on the cofi-language mailing list, I suppose.)

> Therefore, we should aim at producing a complete semantics for
> v.0.99 as soon as possible. A rather tight schedule has been agreed
> upon in Lisbon; I allow myself to change the dates taking into account
> the change of date for the v.0.99 release:
> 
> - as soon as v.0.99 is officially released, Don will split the text up
> and will send the appropriate chapters to the subgroups responsible
> for the semantics of particular parts of the language.

[I will do the splitting up and sending out today if I manage,
otherwise tomorrow.  -- DTS]

Meanwhile, I'll try to get casl.sty ready for release, as it'll be
needed for formatting the split parts (unless one removes all the
concrete syntax...)

I've made a few changes to the names of commands and environments used
in the Summary, and removed some of the relics of the days when
Hyperlatex didn't know so much about Latex.  I'm not sure to what
extent the affected commands and environments were used in the formal
semantics, but from a quick look it appears to be minor.  If I'm wrong
on that, it might be worth updating S-6 to use the new names before
splitting.

[I think it would be a distraction to update S-6.  The authors might
have to be a little careful about adapting the text in the previous
semantics to the current version of the summary. -- DTS]

I haven't updated cofidoc.sty recently, but everyone should get fresh
copies of cofibib.bib (just) before running bibtex on the Summary.
[ftp://ftp.brics.dk/Projects/CoFI/Bibliography/cofibib.bib]

> - we give ourselves about three weeks to produce the semantics for
> each part; by May 16th each part of the semantics should be sent to
> the assigned kibitzers.

I leave for Japan on Friday morning, get back on 4th May (evening),
leave again for England on 7th May and return on 13th May.  I won't
have much e-mail contact at all during that period - nor much spare
time.  But the last two weeks of May should be relatively free, and
not much has changed with the design of libraries, so I hope that any
delay to my part won't upset the schedule too much.  (I should of
course take account of the valuable comments made on the previous
version.)

> - by May 23rd we expect that the semantics of each part will have been
> studied and commented upon by the kibitzers.  The authors have a few
> days to take these comments into account before sending their parts of
> the semantics to Don.
> 
> - by May 30th Don will have the whole text integrated.
> 
> Let me remind you once more of the assignments:
> ...
>   Libraries: Peter (kibitzer Hans-Joerg)

OK for me - but as Till is so interested in libraries, perhaps he
might help or take over from Hans-Joerg?

> ...
> Institution independence
> ========================
> 
> It would be desirable to make the semantics of CASL as institution
> independent as possible. This was one of the points that the IFIP
> WG1.3 referees stressed. It seems that a notion of institution with
> identifiers (proposed by Till, close to that of institution with
> syntax of Sannella & Tarlecki'86) should provide an appropriate
> abstract framework for this. Most roughly, an institution with
> identifiers is an institution additionally equipped with a functor
> that extracts from each signature the set of fully qualified
> identifiers that occur in it, and with some possibility of lifting a
> partial map on such identifiers to a "less partial" map, intuitively
> capturing extension of a map on identifiers to compound identifiers.

You may have noticed that Ch 1 of the Summary now mentions |Sigma|.  I
call "fully-qualified identifiers" simply "symbols" - I hope that
doesn't cause confusion?

> Wishing everybody good work!

Ditto! - but I'm quite busy now preparing for my trips, so don't
expect much further input from me for the next 3 weeks or so - apart
from casl.sty, and fixing any problems with adapting S-6.

Best regards

Peter
_________________________________________________________________
=========   Peter D. MOSSES              Associate Professor, PhD
==== ====   BRICS                        mailto:pdmosses@brics.dk
=========   Dept. of Computer Science    == pdmosses@daimi.aau.dk
==== ====   University of Aarhus         http://www.brics.dk/~pdm
==== ====   Ny Munkegade, bldg. 540      Telephone: +45 8942 3364 
==== ====   DK-8000 Aarhus C, Denmark    Telefax:   +45 8942 3255