doing dccproc -t many

Vernon Schryver vjs@calcite.rhyolite.com
Sun Apr 7 16:28:01 UTC 2002


> From: "Tony L. Svanstrom" <tony@svanstrom.org>

> ...
>  Why even have a function to report a message as "spam" in a system like DCC;
> shouldn't it focus only on actually "reporting" messages sent to more than one
> person... then why accept "fake" values based on if someone doesn't like a
> certain message?
>
>  Not trolling, just trying to understand the thinking behind it...

In fact there is no way to mark a message as spam for the DCC (in the
normal mode of running DCC servers) and there are only reports of checksums
with a number targets and a total for the numbers of targets.  "Many" is
merely the "infinity" or limit at which the total count would overflow.
`dccproc -w many`
`dccproc -t 9000000; dccproc -t 9000000`
and `repeat 17000000 dccproc -t 1` have the about the same effect
inside the database.  Both cause the DCC server to say those
checksums have counts of "many".

Part of the idea of the DCC is to accept reports from anonymous DCC
clients.  Anonymous DCC clients can include the bad guys as well as
confused good guys.  That implies that DCC servers must tolerate odd
things such as `repeat 17000000 dccproc -t 1`.

"Many" should not be seen as marking "spam".  The DCC can't know
about "spam" or "unsolicited bulk," but only about "bulk" unless
you radically restrict DCC clients.  Unless you run your DC servers
to accept reports from only a very few people who you trust to not
only be good guys but to not make mistakes, there will be mistakes or
malicious reports.  Instead, "many" should be seen as saying "at least
one and possibly more DCC clients said this checksum was awfully bulky."


Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.