Caught author interview with Scott Rosenberg, about his new book about the Chandler project and software development. I like Rosenberg's writing, but I haven't read the book yet.
From the interview, Rosenberg sees Chandler's failure-to-thrive as a cautionary tale about all software development. However, Chandler actually had a distinctively awkward set of initial conditions:
* architecture driven. They had a grand vision of a message-based storage model that they needed to get perfect before they did anything
* clearer vision of architecture than application. Reading Chandler's material, there was no clearly articulated goal beyond a free clone of Outlook (though that alone wouldn't have been a bad thing)
* infinite budget. Open source, with a wealthy funder. No economic constraints or time pressure to keep them on the straight and narrow. No personal itch-scratching, unlike the classic open source story.
Plenty of software projects fail because they don't adhere to the logical set of constraints. Chandler started without the constraints.