At 21:16 -0700 3/9/03, Vernon Schryver wrote:
Fine, but "forged" is wrong. In English "forgery" means "an act of
forging; especially the crime of falsely and fraudulently making or
altering a document (as a check)" (See http://www.m-w.com/ )
It is wrong to say all 96.6% of those correspondents are guilty of
"falsely and fraudulently making or altering" SMTP headers.
This matters because people in this mailing list have been saying
"because 99% of spam is forged, spam defense X won't work/is required."
They are using "forged" to mean "falsely and fraudulently making or
altering" SMTP headers, but I suspect that their numbers are less
convincing than your's. Your numbers are not guesses or intuition,
although they have nothing to do with "falsely and fraudulently making
or altering."
So you've finally come around to the point you've apparently been
trying to make for these past many days without actually saying so.
Can't we just accept that the phrase "forged header" has long been a part of
the e-mail vernacular, as a signifier of message metadata whose contents do
not point back to the true sender, point of origin, and/or subpath by which
the message traveled. Such "forged headers" may point to non-existent senders,
non-existent points of origin, and so on.
Perhaps the phrase should be "forced header," as the common method of
production of
such "forgeries" is to impose the inaccurate header data over that which would
ordinarily be placed on the message by properly-configured clients and servers.
And if we're going to be pedantic about it, it's worth noting that "forge" may
also mean "to give form or shape to" so you might say that all e-mail headers
are "forged" by server client software. But then I suppose if it's all forged,
there's no point in talking about forged messages at all.
On the other hand, it does seem that when message headers are
_forged_ by spammers,
they're often _forced_ to bear information that is not useful in finding the
sender or other responsible parties around the sender.
I respectfully request that replies be redirected to asrg.linguistics
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg