good chemistry is complicated,
and a little bit messy -LW
Developing in Processby drfrog (Deacon)
|on Dec 25, 2003 at 18:56 UTC||Need Help??|
This is my first meditation so bear with me here
Lately i have been doing some consulting, and getting some pretty good jobs, I am always running into the idea of process, of the value of preproduction, and well thought out projectsmake higher quality ones. As I am sure most of you know most companies do not see this as a luxury, that specificationas and understanding that goes along with it is secondary to accompishing the task, they hired you to do a job, and that job must be done now.
Well Ive almost given up on trying to figure out how to make a process that works in accordance with both the clients goals and mine, as a consultant the fast i can understand and accompish the goals, the less cost for the output
I have this rant on UML and perl. what i was doing there is inquiring into the notion of process, the ideal of ramping up from preproduction, UML just gave me the insight, and i dont think i was being clear enough. Anyways going from diagrams and documents into actual production is an intriguing mediation. in someways h2xs and make maker , as well as ummf, the uml to perl translator, all have aspects of what i am point at/meditating on, the idea that since the spec drive the project, how can one take all that hard work/meditating done in pre production and make it into at the very least a very strong starting point for a project?
here are a few things i think a system like this should be able to do:
o able to take uml xmi file and produce a structured code filesystem
o cpan style
o test suites, either perl or mod perl
o able to insert documentation into the right perl module perldoc
o able to file in copyrights, and author details
what other things should be put into this ?
I'd love to here what you other monks think of this concept!