Release Cycles

by AJ

"Skyler's comment about how myspace keeps changing and growing organically, almost every day, is a passionate user's view of what the developer's call quick release cycles. Where software developers are typically on release cycles of 6 months to a year, the Threadless guys said that even two weeks was a little long. In fact, virtually all of the web 2.0-ish folks at the conference mentioned these quick release cycles as crucial.

There are a ton of issues, obviously, like what happens when a new release breaks something that previously worked. The Threadless guys said that happens, but only rarely, and they just do a rollback. Skyler said she's seen things break on myspace, but nobody seems to care much since they know it'll probably be fixed tomorrow."

--Kathy Sierra

I like the metaphor of "release cycles". I have the same view towards curriculum cycles. Most schools seem to change their curriculum infrequently. Occasionally, they may do a curriculum review-- typically a bureaucratic committee that plods ever-so-slowly towards a set of recommendations. These suggestions are then debated, modified, argued... till finally a new curriculum (or policy manual) results.

I tend to think of my curriculum as fluid... something I change constantly. I rarely go three weeks without tossing something out, adding something, or experimenting with something. Its in constant flux.

This approach, it turns out, beats the "change by committee" approach by a wide margin. Quick release cycles produce faster innovation. They allow for more field-testing. And field-testing is the way to go. In fact, I have no patience whatsoever for debates. Its a damn waste of time to debate the merits of a particular idea. Just try it, for goddsake, and youll know immediately whether it works or doesnt. Dont let the paralysis of analysis slow you down.

Theres another benefit to quick release cycles-- change is interesting. Students like it. They get bored with doing the same old thing every day. Some ideas might bomb, but its easy to discard them. And, as Sierra notes, sometimes you may break something that was working great. No problem, just rollback (ie. start using it again).

The biggest issue I have with traditional education has nothing to do with curriculum. My biggest beef is with the PROCESSES they use. Typically, they are slow, cumbersome, risk-averse, authority-driven processes.

Not surprisingly, cumbersome & unimaginative processes usually lead to boring and ineffectice practices. I have no doubt that teachers in such organizations are as smart as those in more effective schools. I have no doubt they could be just as creative, if given the right environment.

But traditional schools squash their creativity. They drain their energy with a thousand-and-one regulations. They sap their motivation with "catch them doing something wrong" management. They saddle them with big classes and dreary classrooms. They bind them with rigid syllabi... with standardized tests... with grades.

Very few teachers can be remarkable in such an evironment-- the school's processes are aligned against them.

The good news: none of this is necessary. As countless companies prove-- it is possible to be fairly large AND innovative. It takes committment. It takes a maverick leader (or organzied staff)... but its possible.

The best advice I have to offer to teachers is this: Shorten your release cycles. Try more and try it more quickly.



San Francisco, CA