Failure

I mentioned in my last failure that I had been considering whether the flexibility of giving myself an extra day was a good or a bad thing. That was largely an experiment, and I think I've got enough data now to say it isn't helpful.

The problem is that the flexibility really just allowed minor failures to be hidden instead of exposed and dealt with. Effectively it lowered the bar so that certain failures qualified as near-misses instead. I now think that's the wrong direction to go; effective improvement requires feedback, and the shorter the feedback loop the better. Waiting until the small failures turn big enough might save face, but I don't think it helped me in the long run.

So my new commitment is simpler and stricter: a post each day by 23:59:59 UTC. Let's see how that works.