Older blog entries for bcully (starting at number 43)

Looks like the spam attack on advogato is evolving. Is codemastr the first sign of the spam singularity?

@Omnifarious: I completely agree about your ext4 rant!

@Omnifarious: There's an interesting thread about that ocaml book on the caml mailing list here, suggesting it may have been plagiarized from this excellent draft. I haven't seen the book myself though.

davidw: If you're put off by git, why not give mercurial a try? It's much more intuitive, especially for people familiar with subversion.

robertc: for persistent hashes, I have been pretty happy with qdbm. I can't say whether it fits all your requirements, but it has been faster and more stable than bdb for me. Its successor, which I haven't used, is called tokyo cabinet and there are python bindings floating around for it. It might be worth looking at if you haven't already.

jch: In case you haven't already read them (though it seems unlikely), I believe the SEDA and Capriccio papers both contain similar evaluations of web server latency, SEDA from an event-oriented perspective and Capriccio showing what can be done to improve the performance of a thread-based model.

rillian: 44 nano? nano?!

rufius, as a Mercurial developer I would be interested in hearing what went wrong in your trial.

ncm: I may be biased as a New Yorker and one of the more than one hundred thousand people who publically protested the war in New York City before it started, but I doubt you have any idea what "most Americans" think. Maybe you find it comforting to believe that the government wouldn't behave the way it does without the support of "most Americans"? I can see why, but it's unfortunately a bit naive. If you were American, what would you have done to prevent the invasion?

ncm: are you being deliberately condescending?