ANN: New technical paper from Ada Europe 2004 now available

[rod.chapman@praxis-cs.co.uk (Rod Chapman)]

I'm pleased to announce the availability of two new technical
papers that were recently presented at Ada Europe 2004.

Both are available now in PDF at www.sparkada.com

Note that the first of these joinly took the best paper award at the
conference.  The latter was awarded best presentation.

"High-Integrity Ada in a UML and C World", Peter Amey and Neil White

Abstract
The dictates of fashion and the desire to use "hot" technology not
only affects software developers but also influences potential
customers. Where once a client was just content to accept something
that worked (actually, would be delighted to have something that
worked) now they are concerned about the means by which it was
constructed; not just in the sense of was it well-enough constructed
but in the more malign sense of was fashionable technology used. This
paper shows how the customer's desire to use de facto standards such
as UML and their wish to use language such as C?perhaps to support a
small or unusual processor; to integrate with other subsystems; for
the perceived comfort of future portability; ot for other,
non-technical reasons?can be aligned with the professional engineer's
need to use those tools and languages which are truly appropriate for
rigorous software development.


"High-Integrity Interfacing to Programmable Logic with Ada"
Adrian J Hilton, Praxis Critical Systems Limited and
Jon G Hall, Open University.

Abstract
Programmable Logic Devices (PLDs) are now common components of
safety-critical systems, and are increasingly used for safety-related
or safety-critical functionality. Recent safety standards demand
similar rigour in PLD specification, design and verification to that
in critical software design. Existing PLD development tools and
techniques are inadequate for the higher integrity levels. In this
paper we examine the use of Ada as a design language for PLDs. We
analyse earlier work on Ada-to-HDL compilation and identify where it
could be improved. We show how program fragments written in the SPARK
Ada subset can be efficiently and rigorously translated into PLD
programs, and how a SPARK Ada program can be effectively interfaced to
a PLD program. The techniques discussed are then applied to a
substantial case study and some preliminary conclusions are drawn from
the results.