[mercury-users] Some remarks and questions

luchp at xs4all.nl luchp at xs4all.nl
Tue Apr 29 23:47:55 AEST 2003

On one of the pages I read: "Mercury is intended for the production of
large scale, reliable software systems by teams of programmers". Some
remarks on this:

I don't think it is possible (or even desirable) to have one language that
fits all situations. Existing highly flexible languages (C++/Perl), lead
to low productivity for large projects because programmers are very good
in finding creative solutions that contain suble bugs. Al these creative
solution also lead to maintainace problems.
Furthermore, end-users experience commercial application as frustrating
complicated and inflexible . They quickly find out that they still need
lots of domain-knowledge to be able to use the tool, and even if they are
a domain-expert, the tool requires them to exactly follow the programmers
way of thinking to perform common tasks.
In my view, these two points are major ingredients for the current
software crisis.

I'm exited about .Net however, it is more safe then binary code and it
offers nice language integration. This enables us to build each part in
the most suitable (safe and restricted) language.  Mercury could be used
to implement the functional part (represent domain knowledge) and provide
the backend for a more flexible, interactive, AI like user interface
('smart' wizards and help).

Getting Mecury accepted as a main-stream programming language will be
benificial, and integrating it in .Net is a good first step.  There is one
major problem though: The average programmer is not of academic level and
has no training in discrete mathematics or logic. In our company only 30%
of software engineers are of academic level and they are responisble for
technology roadmaps, architectures, top-level designs, requirements, etc.
The actual detailed design and implementation is done by engineers on
(polytechnical) college level. Tutorials on Mercury (and existing logic
programming languages) are very much targeted on academic researchers or
students and not on the average professional programmer. Just grab a
Visual Basic book from the shelf and compare it with the Mercury tutorial,
and you know exactly what I mean.

So I have two questions:
1) Will a Mercury plugin for Visual Studio be developed, and if so, when
can we expect it to be released?
2) Are there any plans for a more elaborate tutorial? Maybe you can point
me to some good books.

Luc Holtkamp
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe

More information about the users mailing list