Re: how does nodupmail work?

[Frank T. <ft...@ne...>]

Robert Brown, at 10:41 -0500 on 2002-02-22, wrote:

> Hi.  When detecting duplicates in the body, how is this implemented?
> Does it use a hash?  If so, is it a fuzzy hash, like nilsimsa?

It uses a cryptographically strong hash of the message, with some
canonicalization performed.  It's what I thought of at the time.

I haven't looked into using fuzzy hashes, but I looked at nilsimsa's page
and it seems like a better approach.  I'll have to look into implementing
it.  Thanks for the heads-up.

-- 
Frank Tobin		http://www.neverending.org/~ftobin/