This was a pretty awesome book showing some of the cool things that you can do with set-based programming and SQL databases: medians, efficient tree-storage, and even string parsing!

The edition I read had a lot of typos, but they are generally pretty obvious. The author is also rather active in newsgroups, so there is a lot of supplemental information and the ability to ask questions.

Expecting volunteers to carry 2/3 of the load for Fedora seems a bit much.

We don't *expect* volunteers to maintain 2/3 of Fedora's packages. That is just what the numbers ended up being, and they flux often as different people either adopt or orphan packages depending on their circumstances.

I was thinking "I'll just check Slashdot and see what's new today" and all of a sudden I saw my name on the front page. I thought "what on earth am I doing on the front page of Slashdot? I haven't given an interview in quite a long time."

I realized that the interview was from Linux Format UK (at the end of the article), and checking my records, I can see that it a summary of an interview I gave at LUG Radio Live back in July.

It's been eight years since the Clinton administration. This is 4x the doubling period based on Moore's Law. While Moore's Law relates to transistor density, Wikipedia says that it's roughly similar to gains in disk storage. So in the last eight years, we could estimate disk storage gains of 2^4 = 16x. This doesn't get you all the way to 50x, but it cuts out a big chunk of the gains.