The Deadline

The Deadline: A Novel About Project Management by TomDeMarco ISBN 0-932633-39-0

From the website (http://www.dorsethouse.com/books/dl.html): A project management novel that vividly illustrates the principles -- and the outright absurdities -- that affect the productivity of a software development team.

De Marco lays out the premise of the book in his Dorset House interview (http://www.dorsethouse.com/news/inttdm.html):

Poor Mr. Tompkins, you see, doesn't actually accept the job; he gets kidnapped. And the motivation provided from above is starkly effective: He is to finish the job on schedule or pay for it with his life. So we have a manager who has his very survival staked on a project deadline.

Yes, but "I'll pass on this job," or "time for me to bail" wouldn't have made for much of a story.

From TomDeMarco's page: http://www.atlsysguild.com/GuildSite/TDM/Deadline.html

Preface and Chapter one (a teaser!): http://www.systemsguild.com/tdmbooks/DeadlineSample.pdf


I just got this book and am reading it cover to cover, marking up the pages as I go. Most things I agree with, some are superbly stated, such as, Management is hiring good people, task-matching them to suitable work, building a team, helping it jell, keeping it healthy... all the Gantt/Pert/timecard stuff is administrivia. A few I find provocative, as in the quote in BigDesignUpFront. I suggest every team leader and manager read, underline, and compare notes on this book with some other people. -- AlistairCockburn


The book is full of great observations and insights. It is clear that TomDeMarco is quite the advocate of BigDesignUpFront. One of the aspects I really liked was the point that use cases (function points, user stories, take your pick) are what "flow" through the development process. It's obvious, but it's nice to have an attribution. The application of simulation to the software development process was also interesting - I've always thought that software development is way too stochastic to simulate - there are too many variables - but the chapter about it in TheDeadline illuminated some good first approximations. -- RandyStafford


A little Wiki contest: NameThePeopleInDeadline.

-- ThaddeusOlczyk


One obvious plot hole was that you had a team working to clone every popular early 90s piece of software, finishing just in time for the Internet boom to make desktop software largely obsolete. Sure people still buy desktop software, but you don't base an IPO (particularly a country-wide one) on it. BigDesignUpFront strikes again.

On the other hand, I loved the story as a whole. It prepared me for a lot of XP concepts, most notably that you don't make a process faster by adding steps.

-- JohnBrewer

The section that jumped out at me was towards the end, where it is suggested that more successful teams spend more time doing "design", and less time writing code. But they do this "design" in some kind of DeusExMachina design language that is almost 1-1 with code. And then they review and desk-check it. Weird stuff indeed. When was the last time DeMarco worked on an actual development project? -- KeithBraithwaite

He's gone over to the dark side: 80% of project doing design, then bring in a horde of programmers and code it all.

Now how are you going to know that your coding time estimates are right when you're 80% through the project time, and have yet to do any coding? And what objective measures can you have on how well different team member's designs will translate into code? -- JeffGrigg

Actually, DeMarco is now a convert to XP. See XpImmersionThree for details. -- JohnBrewer

How very nice for him. To repeat the question above "When was the last time DeMarco worked on an actual development project?"

A very good question!


Several things that failed to seem credible in my eyes:
  1. Thompkins marries the wrong woman -- he should be falling in love with the woman he spends the most time with [otherwise, he'd need to exhibit more attraction to drugs and other things of which I will not speak],
  2. he never has to deal with SNAFUs, FUBARs, prima donnas, and only one manager with personality problems -- everyone else is unrealistically happy to work together and for him,
  3. lying to Belok-type managers doesn't work [they'll walk around the cubicles on evenings and weekends to see who is really working],
  4. I really wanted to see a realistic approach to dealing with a Belok-type manager, rather than the 'below the belt' tactic that was actually used.

I found the characters of 'Harry Winnipeg', 'Great Yordini', and (someone else?) very amusing.


CategoryBook

EditText of this page (last edited November 1, 2010) or FindPage with title or text search