UML + some UI

Martin Naef EMAIL HIDDEN
Wed Apr 15 12:06:53 CEST 2009


I think the take-home message of this thread is that it's worth to think 
about methodologies, and match them to the project at hand.

Before you start *any* project, you should do a bit of thinking. Laying 
out your project in UML can be a very useful way to do this.

Only once you understand what you're doing, you should be writing actual 
code. But there are different Schools of thought - the traditional 
methods essentially assuming that the problem is completely understood 
before progressing to the next step, vs. the "agile" methods splitting 
the project into many small iterations assuming that it's more efficient 
to learn as you go. But while the agile methods might be mistaken for 
"let's get things done and think later" (which I'm sure happens), they 
actually also assume that you understand the problem for the small 
iteration before you actually do it just as well.

BTW: It's well worth reading Brook's "Mythical Man Month" book. It was 
written based on his experience managing the development project for the 
old IBM mainframe OS, but a lot of what he said is just as valid today 
(there's a 20 years anniversary edition where he actually reflects on 
what he wrote in the first edition). One of his key statements still is 
"there is no silver bullet" - meaning, there's no magical solution that 
suddenly makes everything better or easier. Whatever methodology you 
choose to plan and execute your projects, you still have to understand 
what you're doing and manage the complexity well.

Martin

-- 
http://www.navisto.ch
http://www.myspace.com/navisto



More information about the music-bar mailing list