[GNU-COBOL] Thanksgiving notes.

[Laura Tweedy <tweedy@umr.edu>]

Okay!  Here we go.

I know I said that I would give you design docs or interfaces to write to.
But, after discussion, it's too early to solidify stuff like that.  We feel
that it's better to keep the design fluid until we've examined some different
approaches.

I'm sorry this isn't as formal as projects you might see in a true job
environment, but this is, after all, a volunteer project being run by amateurs.
:)  I'm trying to work in the interests of the project by not excluding new
ideas early, which I think giving a hard and fast design would do right now.

Here are some things we did come up with:

1) design is an ongoing process.  We have some direction because of our last
	Cobol 2 C project, but there are new issues we've never examined before.
2) customer-producer relationships.  It would be good if everyone can use this
	mindset.  As a programmer, you are a producer and you have to meet
	your customer's needs.  Your customer may be vague, as in real life.
	Ask your customer what he needs.  Write what you think they want and
	ask for clarification if you're not sure.  As a customer, try and be
	specific and complete in your request.
3) runtime is chad's responsibility.  If you are writing runtime code, he is
	your customer and the final authority.  Please don't ask me about what
	Chad wants, only he knows that.
4) programming tasks.  The creation of "modules" has been artificial so that
	I could give people something to do.  This was wrong.  Chad will have
	a list of tasks out soon.  NOTE: a task may include research, 
	prototyping, etc. and may or may not be included in the final project,
	but they are necessary tasks.
5) user stories.  Glen, you and I will need to come up with these.  User stories
	describe, in English, what a piece of Cobol code does, inputs, outputs,
	side-effects, semantics.  These are used to document what the final
	program should do.

That's pretty much it for general descisions/conclusions.  Now, on to some
specifics.

Matt - functions to write:
	trunc, getString, putString
	You define the interfaces.

Mark - pull out accept and display code
	do not modify beyond replacing %s's etc.
                - can remove __'s and make variable names more readable.
        create interface from var args to fprintf in the old code, vars
                become arguments to the function
        content and structure are correct as they are
        interfaces are fluid and subject to change

Justin - file io  We had a bunch of stuff, but since then, we thought of some
	new wrinkles.  So, ask Chad what you need to do.

Glen - I will send you the latest copy of the source tree.  We are still 
	working on getting anonymous CVS working.
	You and I need to get together and discuss what to do on the grammar.
	I will mail you with times that are good for me.

Whew!  If you have any questions, I'm planning to be on the moo whenever I'm
at a computer.  Everyone should have an account, if not, login with "connect
guest" and complain.  The moo is at acm.cs.umr.edu 31337.

Thanks everybody!  Good luck with finals.

laura

-- 
   .. Laura Tweedy .. tweedy@umr.edu ..

--
This message was sent through the gnu-cobol mailing list.  To remove yourself
from this mailing list, send a message to gnu-cobol@acm.cs.umr.edu with the
words "unsubscribe gnu-cobol" in the message body.  For more information on
the GNU COBOL project, send mail to gnu-cobol-owner@acm.cs.umr.edu.