“Large-system programming has over the past decade been such a tar pit, and many great and powerful beasts have thrashed violently in it. Most have emerged with running systems—few have met goals, schedules, and budgets. Large and small, massive or wiry, team after team has become entangled in the tar. No one thing seems to cause the difficulty…….But we must try to understand it if we are to solve it.”
The Mythical Man-Month, Essay 1 – The Tar Pit
Is today any different?
Nearly forty years later it still rings true of today’s software development landscape.
You could think that software development has come a long way since 1975. We have many, many new technical tools that were unimaginable back then.
We have new project management methodologies. We have more university-trained programmers and more tech-savvy users. And yet, the failure-rates of software projects are still high, both in comparison to engineering and construction projects, and high in comparison to the casual expectations of most people involved in IT*.
We are still stuck in the tar
Maybe part of the explanation for this is that tar just is sticky no matter what tools you have available. Software development is a difficult thing to do and all of our accumulated tools and experience should not lead us to expect it to be otherwise.
Its actually not at all easy to put a finger on what is so difficult about it. In my Software Engineering Notes article I try to give an answer by means of a comparison with logical problems and puzzles. Much like with logic problems, there are no reliable and repeatable techniques that will always enable us to solve the problems (deliver the projects) we want to.
More often we have to develop new techniques or apply old techniques in novel ways.
Anticipation is everything
In my article, I focus on accepting the tar pits for what they are but emphasise that it’s always easier to get out of a tar pit if you can learn to resist false hopes of walking across it. Be prepared for the black sticky stuff!
Have a read of the article and let me know what you think. Do you have any tips for wading through the tar?
* The Standish Chaos study arrived at a 30% failure rate (http://net.educause.edu/ir/library/pdf/NCP08083B.pdf) and something around 30% is the figure most usually quoted. It has more recently been claimed that if overruns and significant scope reductions are taken account of then the rate goes up to 68% (http://www.zdnet.com/blog/projectfailures/study-68-percent-of-it-projects-fail/1175).