My problem with benchmarking is that it's tedious (and finicky). Goodbenchmarking mostly consists of twiddling my thumbs while my tests run,often yet again after some minor tweak or improvement or additionalcase. I can kill days running tests and staring at numbers and not reallyget anywhere (and sometimes I've had to).

All that said, after my entry on some regular expression performancesurprises , I got curious enoughto do more detailed micro-benchmarking. With Daniel Martin 's assistance I was

Irony is a Referer spammer spamming my entryon how affiliate marketing is undead forsomething that sure looks like an affiliate marketing scheme. More ironyis that this is the first Referer spammer in a donkey's age; all the oldones seem to have given up months ago.

This just goes to show that I can find amusing things from readingmy server logs.

An analysis of the spammer

The spammer came from 217.15.96.18, an unremarkable DataStreamMalta IP address that

Now that I've automated almostall of the Hotmail spam report, of course it turns out we've had a quietweek, even more so than last week :

• no messages accepted.
• 22 messages rejected because they came from non-Hotmail emailaddresses.
• 54 messages sent to our spamtraps.
• 13 messages refused because their sender addresses had already hitour spamtraps.
• 5 messages refused due to their origin IP address (one in theSBL, one in the CBL, one from Nigeria, one from

I've written before about the advantages of scripts ,but my own recent experiences have reminded me of another one, whichI can snappily phrase as automation promotes action . Or in otherwords: when you've scripted something so that it's easy to do, you do itmore often.

I build Firefox from CVS myself, for various reasons, but I run it ona different machine than I build on. Bringing up a new build up on theright machine involves about three or four steps;