1421Re: [scrumdevelopment] Digest Number 327
- Jun 5 7:20 AMI think the consus of this group would be the same as I believe ACM has
taken. SWEBOK is attempting to promote processes on which there is no
consensus and where substantial data exists to demonstrate that many
proposed approaches do not work.
The ACM and leaders in software development today do not support SWEBOK
does not accurately reflects the current consensus.
The June issue of IEEE Computer has an article which documents the failed
DOD standard promoting the waterfall method that cost the military about
$100B in abandoned or broken projects. The author of the standard said he
used consultants and textbooks to draft the standard. He had no real world
experience in building production software and would have proposed
iterative incremental development, had he known what was really going on at
NASA and IBM. It has taken more than a decade for a review committee led by
Fred Brooks to repair the damage.
It has also led to curriculum development in universities which has been
largely irrelevant to production software development. I gave a talk at
Brown University a few years ago on modern software techniques used in real
businesses today. A Ph.D. candidate got up and stated to the entire
conference that virtually everything he had learned about software
development at Brown was largely irrelevant to modern methods.
I don't want to fault Brown because I was on an advisory committee to the
University of Massachusetts and they were even worse. They agreed with the
industrial advisory committee about what to do about the problem and then
failed to implement it teaching the same old things that were not relevant
to product development.
At 01:55 PM 6/4/2003 +0000, you wrote:
> Date: Tue, 3 Jun 2003 07:30:37 -0600
> From: "Alleman, Glen B." <glen.alleman@...>
>Subject: RE: IEEE SWEBOK Is Looking for Reviewers--They Don't Even Mention
>XP, Agile, etc.
>As a reviewer of SWEBOK this is a problem in a larger context. SWEBOK is
>an academic framework for the teaching of Software Engineering. The
>first thought is that XP and agile method are transitory in the
>engineering world. Meaning they are the "current" methods and describing
>them in a "body of knowledge" is risky. Not because of their value to
>the software world but because they are too far down in the food chain.
>It would belike teaching GIS methods for waste water remediation in an
>environmental engineering Body of Knowledge - critically important but
>only so once you reach the field.
- Next post in topic >>