Following up on HowDoYouRead I thought about the costs of reading. Not the financial, but rather the mental. Essentially, I have two: 1) context switching to verify or extend info, and 2) fitting the new info into my understanding of the universe.

The benefit of validating info pays for the context switching costs, as far as I can tell (over a lifetime, we seem to get some junked up knowledgebases). Fitting the new info into the tree (or heap) sucks because it involves comparing new and previously learned info. It seems to be at least O(n^2) comparisons, but each comparison is bounded only by the size of the entire tree as a hugely explanative idea could impact all previously gained knowledge (my update for TheSelfishGene ran for days).

The tree search to find all impacted previous knowledge sucks because I don't have an ACID complaint database (yes, I realize there is a joke in there somewhere ;), and previously established links fade with disuse (failing Durability). So info that needs to be updated may not be (I found unupdated nodes impacted by TheSelfishGene over a year later).

Slow and prone to error....

Seems like a strong argument for getting lots of sleep and taking long walks. Gotta have time to integrate all that knowledge. Also, kind of makes you realize how important it is to read high quality material. If you're going to spend all that energy trying to integrate your knowledge better make sure you're integrating quality content. -- David W.

I wish I had two db, one ACID-compliant for prioritized data, another lossy db for everything else. -- Patrick.

This is why I stick my head in the sand and rely on cliche rather than profound understanding. - Nathan