Español
Subscribe Ads
Links
Ivar Jacobson reflecting on the evolution of the UML
Ivar Jacobson (one of the three initiators of the UML back in the early 90s) has written a blog post where he reflects on the evolution of the UML.
I could not agree more with his last sentences: "UML is not the silver bullet it was sold as ten years ago. Nor it is as bad as academicians, agilistas and competitors claimed five years ago" and "Still, UML has become complex and clumsy. For 80% of all software only 20% of UML is needed".
As we have discussed several times, the key is to find the right amount of modeling for every kind of development project (depending on many factors, as the expertise of the team, the criticality of the system to be developed, its complexity and so forth).
If you enjoyed this post you can subscribe to this Software Modeling blog and/or follow me on twitter . And if you really liked it help me pass it on to others by bookmarking and sharing the post using the links below:
Delicious
Digg
StumbleUpon
Reddit
Facebook
Google
Yahoo
Technorati
No deep discussion about UML vs DSL so far
Jacobson's post is very interesting indeed, mainly to get a grasp on the UML short history from one of its parents.
As an advocate of MDD/MDE, I often lack in that kind of "UML after X years" articles a review of such a standard with respect to the advent of ideas such as DSL (Domain Specific Language), ADL (Architecture Description Language) and Architecture Frameworks, with their viewpoints/views. (I mean, something deeper from the rapid answer "UML Profiles are good enough to give support to DSLs, and UML different diagrams are good enough to any architecture framework ;-).
IMHO a deep discussion about the role of UML among those "new" software engineering tools is more than necessary, letting us assess the real contribution and health of UML to modern software development.
In this respect, I fairly agree with Pedro Molina's "20% of UML is useful to describe the 35% of the design. (For the restating 65%, other models are still needed)" Not sure about the figures though ;-)
A related blog post
Pedro J. Molina has also written a blog post commenting Jacobson's article . Definitely worth reading