Dodging (Silver) Bullets

In this article at ACM Queue, Alex Bell, a software architect at Boeing, pans the latest palatives in software development -- XML, UML, etc. -- and recommends what might be called reality-based software construction:

The challenges of software development are difficult enough without also having to endure the ricochet of silver bullets strafing supporting organizations. For example, many systems engineers have discovered that using the UML greatly simplifies the efforts that their predecessors unnecessarily struggled with in pre-UML days. As opposed to having to devote significant efforts to the consideration of constraints such as network bandwidth, processor speeds, and the speed of light when developing system architectures, such annoyances are now overcome by creating reams of UML diagrams containing very vague entities and equally vague navigation between them. Unfortunately for software engineers, the UML silver bullets whizzing through the systems engineering space translate into much more work for them.

[...]

It is ironic that a reverence for silver bullets usually brings more pain to software organizations than if they had simply rolled up their sleeves, focused on traditional engineering activities, and accepted that there is no such thing as a free lunch. Postponement of effort, production of useless artifacts, and illusion of progress are just a few of the factors that result in big price tags when reality eventually trumps denial. Even in the absence of Fred Brooks's well-known admonitions [there are no silver bullets that solve everything], it is startling that the failure of past silver bullets to hit their targets is not enough to make us very wary of them today.

Link via Slashdot, where comments will, with near certainty, turn into one or more flame wars.