Pre MEP discussion

[Bob Calco]

I'm thinking of formulating a few MEPs to get Mozart-Oz to an
Enterprise-Ready status for common business applications (or I may
never be able to convince management to let me use it!).

Specifically, I'd like to see (and contribute to) immediate
improvements in the following areas:

	1. Native database support. Need SQL Server, Oracle, DB2,
MySQL and PostgreSQL support out-of-the-box.

	2. Library support. I've become quite cozy with the MS
Enterprise Library for .NET and I like the idea of creating a
mozart-entlib implementation of various "Application Blocks" (i.e.,
chunks of interdependent yet orthogonal enterprise functionality, such
as XML-based configuration; database access; exception handling;
logging and performance monitoring; cryptography; security; etc.).

	3. Windows platform support. Specifically, COM and possibly
steps in the .NET interop direction. Being able to automate tasks on
Windows, such as creating Excel spreadsheets or perform adminstrative
tasks, is a must. Would be nice to be able to create business object
COM+ components that can be used form .NET, too. I think Oz would make
a great middle-tier business logic language, but the technology
infrastructure for integrating with common platforms like .NET (and
Java, frankly) is lacking. I like what some smart folks have done for
Ocaml (http://tech.motion-twin.com/ocamole.html). Python, Perl and
Ruby all support OLE on Windows, too.

Eventually I want to realize my dream of a Windows-based development
environment, an integration with VS2005 or something, too, but for
starters, Mozart's library for the above enumerated uses must be
beefed up.

On one level, it seems to me a ground-up .NET implementation would be
desirable. But perhaps simply being able to interoperate at the COM
level would be great. In that case, Oz needs to be both a COM client
and a COM server. The latter may require getting Oz to compile using
the MS or Intel compilers. Maybe this too is a possible MEP.

I just wanted to start the discussion and rededicate myself to
contributing to all of the above. I have a soft-spot in my heart for
Mozart's multi-paradigm programming support, but until and unless it
is able to compete with other technologies for mindshare in an
enterprise development environment, I don't have much hope of getting
to use it anywhere anytime soon... Unless I win the lottery and can
dedicate myself full time to pet technology projects... But I don't
play the lottery, so my chances are pretty nil... ;)

- Bob Calco