Re: [scrumdevelopment] Government funded projects
This is why Jarzombek reported in 1999 that out of a total cost of $37 on
DOD projects analyzed, 75% failed or were never used and only 2% were used
without extensive modification. DOD changed their spec to dump the waterfall
method and move to iterative development over this.
It also reminds me of Goldratt's book on Project Management where the
students in the story visit a construction business to find out if they have
project overruns as is common in software development. The answer was, "We
always have cost overruns. We never make any money on the initial bid
because we always lowball it. All our money is made on overruns and we would
be out of business without them."
So the "can do" attitude is really a smoke screen for "let's rip off the
government for as much money as we can get."
The argument for these people is that SCRUM would allow them to get stuff
done more quickly and still stiff the government. I once worked for a
department chairman that had a very effective strategy of completing his
next grant's work on the current grant's money. He then could write a very
compelling proposal for the next grant (which was already done) and always
had plenty of extra money for his research projects. This is probably
illegal, but in the larger scheme of things is more moral that what the team
you describe was doing.
Date: Tue, 29 Apr 2003 10:07:21 -0500
From: "Mary Poppendieck" <mary@...>
Subject: RE: Culture clash
I recently ran into yet another reason why people do not raise a red
flag when things are not going well. I was at a site where extremely
time-critical programs are government funded, and frequently
under-funded. I was extolling the key virtue of SCRUM - that you will
know very early if the project is 'doable'. The most senior person in
the room responded: "But we don't WANT to know if a project can be