[mercury-users] More supporting material

[Luke Evans]

Hi all,

This could quite easily seems like a negative criticism - it isn't.  I know
this has been discussed before, but I continue to feel that Mercury is under
represented in terms of supporting material - tutorials, documentation,
examples etc.  Now, it _does_ have each of these, but I recently had a look
around at what was offered for Haskell and Clean, and there is so much more
supporting material - particularly for the beginner or those converting from
an imperative background (I'm not counting formal language specifications
here).  

Before I continue, here's a little diversion...

Personally, my background is in imperative programming, with an interest in
Prolog gained from a minor in AI.  I'm a software architect with Seagate
Software and very familiar with things in the OO world.  I'm convinced of
the benefits of the logical/functional paradigm once the concepts are
packaged in ways which streamline development in large, commercial projects
(rather than being interesting academic tools).  I believe this is starting
to happen.  Mercury interests me because I see the value in the concepts it
endues, but also because I believe the intent is to (ultimately) deliver a
_practical_ general purpose language with a useable set of tools from
Melbourne to bootstrap interest in the language.  The provision of a useable
debugger is important here - and we now have this tool.

... now it seems to me that Mercury is, to one extent or another, going to
be competing in the 'functional' programming niche (what I'm really saying
here is that it is part of the declarative set of languages which may become
popular).  Assuming this niche gradually makes it's way into the mainstream
(through more and more developers being appraised of its benefits), the
successful languages will obviously be those that have attracted continual
and growing attention over a long period of time.  At some point, these
languages will bridge the chasm and become a gorilla in the 'marketplace'
(to borrow heavily from Geoffrey Moore).  I'm convinced that the collateral
material is a critical part of attracting the support which will help propel
the language into the mainstream.  A few of the other languages in the
functional/declarative domain already have books in print and these will
help fuel interest.

I imagine that one of the problems with Mercury is that it is the product of
so few people.  Haskell is much more of a collaborative effort involving
many more resources.  This makes it hard for the core team to concentrate on
the important aspects of moving forward to a 'version 1' of the language and
tools, whilst also bearing in mind the supporting material.  I'm not sure
how big the Mercury developer community now is (does anyone have an idea?),
but it seems to me that we all need to contribute with the supporting
material if there simply aren't the resources for this in the core team.
Ralph Becket has made an excellent start with his tutorial, but I think we
need to build on this.  

I can't (yet) count myself amongst the cognoscenti when it comes to Mercury.
To date I have but scratched around with the language, and many of the
concepts remain new to me.  However, I'm about to embark on some new
research within my company, and I've a strong feeling that Mercury will
feature :-).  Suffice it to say that once I develop a reasonable confidence,
I would be more than happy to contribute to informal documentation,
tutorials and examples.  Assuming there are others engaged in projects
involving Mercury, there must be a growing community of users out there
which might be willing to contribute in this regard, with examples, case
studies, hints and tips and other sundry collateral material which would
surely bolster the uptake of Mercury.  Maybe all that's needed is some form
of moderation for material offered - an editor in chief - and a channel on
which to present new material?  Would a sort of 'open development'
documentation effort be feasible?

Luke

--------------------------------------------------------------------------
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
--------------------------------------------------------------------------