Re: (FC-Devel) Argo

[Jason Elliot Robbins <jrobbins@cadet.ics.uci.edu>]

Hi Danny,

>The impact of the decision to take Argo (as a whole) as a starting
>point has several major implications. However we will not know which
>until we studied and discussed it. Unfortunately the source code is
>not yet available. 

I agree that taking Argo as a starting point would have major
implications.  It would jump-start the FreeCASE project from starting
with almost nothing to starting with a fairly complete outline.  I am
sure that it will not be the outline you want in the end, but I think
it would be a great place to start.

I wanted to release the source code this week, but my advisor (my
boss) told me that he has one change that I must do before the
release, but then he went on a 4 day vacation before he told me
exactly what that change was!  Such is the life of a grad student.
Anyway, I'll use this time to write some overviews (see below).

Please don't evaluate Argo/UML as a final product, it is an ongoing
development and research project.  So if there is something that you
think is missing, we can work together to add it.

>I didn't plan to learn Java, but if I have to, I will.

Java is a great language to learn.  Especically if you know C or some
C++ already.  Having lot of example code will help.  I will gladly
help, and I am sure that many FreeCASE people would like to help teach
each other various things.

BTW, one cool way to learn a bit about Java is by using MindQ's
multi-media computer based training.  The full package is expensive,
but you can learn syntax for free.  I never thought having multi-media
would help in learning syntax, but it does.  The system is a joy to
use.  Sun's java developers site also has some good tutorial
information to get you started.  Try their Java Tutorial.

http://www.mindq.com/java/javatrain2.html

http://developer.java.sun.com/developer/onlineTraining/


>I think it's a promising idea, but for now it just puts uncertainty
>as to 'what next' and 'what can I do to help' into the project, just
>as it is getting momentum. 

That would be true if the answer to "what next" question was coding.
Rather than making it less certain, using Argo/UML and adopting my
project goals and scope similiar to mine (at least for now) would make
the question of what to do next _much_ more certain.  I have a pretty
clear idea of what Argo/UML will turn out to be, and I know several
places where I need other people to help.  So if you are ready to
code, Argo/UML is ready for you.

I think that a better answer to "what's next?" for FreeCASE is
requirements and architecture.   Also, one big contribution would be a
survey and comparison of existing CASE tools and what they do well and
what they do poorly.  The fact that FreeCASE is free is one imporant
differentiation, but that alone does not guide what to do next.  

>Maybe the contributors can give some
>pointers through this list to keep up the speed? 

Yes.  I'll be writing external documentaion and overviews of Argo's
features and code.  I'll put them on the Argo/UML website and just
post pointer on this list.  The first of these overview documents is
already on my website, and I have posted the JavaDocs.

http://www.ics.uci.edu/pub/arch/uml/v03/docs/

And I have also started to outline suggested enhancements to
Argo/UML.  

http://www.ics.uci.edu/pub/arch/uml/jobjar.html


I think it would be useful to decide whether (1) Argo/UML and FreeCASE
will totally merge, or (2) they will be different packages with
complementary goals.  I think that (2) might be the case because I
envision Argo/UML as primarily a single user application that is easy
to use and useful to making design decisions.  FreeCASE's direction is
still very much up to the developers to decide, but some early
indications are that it will have a much larger scope, be distributed,
support multiple users, with different clients, etc.

In any event, Argo/UML is there for people to look at and think
about.  Often it is easier to state what you want to do in terms of
how it differs from what is already there.  Also, there are some parts
of Argo/UML that are likely to be part of _any_ purely UML tool,
e.g. the UML meta-model implementation.

I have put about three years of work into the research and
implementation behind Argo/UML.  So I would really like to see it
used.  It would certainly make FreeCASE seem much closer to reality.

jason!