Sunday, November 03, 2013

KISS of Death

The Washington Post has the inside story on the mismanagement of the development of the website. Because it's a story from insiders, who were fighting with each other over how to manage the project, it's still slanted in favor of the program. But the managerial details ring true to me.

"Inside the Department of Health and Human Services’ Centers for Medicare and Medicaid, the main agency responsible for the exchanges, there was no single administrator whose full-time job was to manage the project."

Excuse me while I sputter, searching for, and failing to find, the right words to describe such an arrangement.

In 2010, David Cutler, "Harvard professor and health adviser to Obama’s 2008 campaign", wrote a memo to the White House economic team, warning them that the project set-up was falling "far short of what it will take to implement reform successfully." But, his advice was not taken.

The underlying theme of the WaPo story is the triumph of political fears over technological needs... until the needs bit back.

So... why wasn't the bite-back foreseen? After all, this administration has put itself forward as smart, savvy, capable of coordinating markets and bureaucracies with just the right nudges.

I keep scratching my head over this, this phenomenon of self-proclaimed techno-managers who can't manage the tech. Peter Greenfield offers a scornful explanation:

"Our technocracy is detached from competence. It's not the technocracy of engineers, but of 'thinkers' who read Malcolm Gladwell and Thomas Friedman and watch TED talks and savor the flavor of competence, without ever imbibing its substance. These are the people who love Freakonomics, who enjoy all sorts of mental puzzles, who like to see an idea turned on its head, but who couldn't fix a toaster."

Obviously, this is an outsider's view of what went wrong.

Bright non-coders have a particular affinity, in my experience, for requesting overly complex system designs that aren't really as thought-through as they at first appear. For such situations, the KISS principle is the antidote, but it can be a bitter pill indeed. Keep It Simple Stupid seems so... insulting and limiting. And, really, it's not very precise in its formulation, since very complex systems do get built successfully - after all the details have been truly worked out, so that the lowest level components really are simple. In this case, clearly, that did not happen. Perhaps because it just couldn't happen over the space of mere years.

In support of which thesis, let me mention,
that the law itself, in its two-thousand page glory,
defied comprehension.

And here we pause our story.

No comments: