Thanks for having me come to talk

[mjd-perl-pm@plover.com (Mark Dominus)]

I had a good time, and the discussion will be useful for my book.

I will get the slides and example code online as soon as I can, either
today or tomorrow, and post the URLs here when it is ready.

We had a brief discussion of MD5 checksum collisions, where two files
are different but appear the same because they have the same checksum.
I said that it was futile to worry about such things, and alluded to
an article in which I answered a (possibly rhetorical) question about
whether I would be willing to bet the safety of a nuclear power plant
against the chance of getting an MD5 collision.

The articles are available from:

http://groups.google.com/group/comp.lang.perl.moderated/msg/2bc5c9b2bd45cba6
http://groups.google.com/group/comp.lang.perl.moderated/msg/902ec02aa263d2f3

I think the articles are worth reading.  But I'll summarize: If you
have 100,000,000 different files, the chance of a checksum collision
is around 10^(-22).  The chance that the nuclear power plant will be
catastrophically destroyed by a direct meteor strike is around 38
million times greater, so:

        If you're losing sleep over MD5 collisions in the power plant,
        you should be losing about 38,000,000 times as much sleep over
        meteor strikes.

        It's a scary world, and MD5 collisions are among the least of
        your worries.

-
**Majordomo list services provided by PANIX <URL:http://www.panix.com>**
**To Unsubscribe, send "unsubscribe phl" to majordomo@lists.pm.org**