I ran an extensive set of benchmarks to compare the relative performance of storing mail in mboxes versus maildirs. I have prepared a white paper with the results.

I am looking at having a few people look over my results, from a devil's advocate point of view, before I publish it. I want to make sure that my results can withstand the inevitable criticism; and that I have not made some silly mistake that makes all the results meaningless.

If you would like to look over my results, contact me directly. I must say in advance that this is a very lengthy and technical document, it's more than just a spreadsheet with a bunch of numbers. If you do not hear from me -- sorry, but someone probably beat you to it. I don't need more than a few folks to give me a reality check, and my preferences lean towards folks with extensive technical background who fully understand the underlying technical issues of something like this.