Why UML Fails to Add Value to the Design and Development Process

Posted by astrec on Wednesday October 29, 2008 @06:46PM
from the lispy.wordpress.com dept.

http://lispy.wordpress.com/2008/10/29/why-uml-fails-to-add-value-to-the-design-and-development-process/

( 4 comments | 20 points | on HN )

Re:Why UML Fails to Add Value to the Design and Development ... by systems on Thursday October 30, 2008 @04:01AM (#348544)

In my personal experience, the biggest problem I had with UML is tool support. Most modelling tools are terrible and/or very expensive.

Modelling is very tool dependent, without a good tool that help you view, edit, print and manage your models; you will just quit it.

A mature, FOSS tool, can take UML a long way!

[ Parent | on HN ]

Re:Why UML Fails to Add Value to the Design and Development ... by DanielBMarkham on Thursday October 30, 2008 @08:39AM (#348658)

UML is a language to facilitate conversations about design and architecture. It's about collaboration, communication, and consensus.

People try to make it about another C -- Control. This is the whole MDA, MDD, etc scenario. In my opinion, it's not there yet -- for some of the reasons outlined.

Instead of having such verbose criticisms, it might be better just to use the language for what it's good for and ignore the rest. Life is short.

[ Parent | on HN ]

Re:Why UML Fails to Add Value to the Design and Development ... by learninglisp on Thursday October 30, 2008 @05:04PM (#349210)

I don't think it was verbose criticisms. I was focusing more actually on an architecture for increasing the level of abstraction.

[ Parent | on HN ]

Re:Why UML Fails to Add Value to the Design and Development ... by DanielBMarkham on Thursday October 30, 2008 @06:08PM (#349306)

Increasing the level of abstraction for what? It's important to outline your goals here: audience, desired end result, level of abstraction, etc.

[ Parent | on HN ]