Sending this again, since Harald claims here:
http://www1.ietf.org/mail-archive/ietf/Current/msg22200.html
That he is not censoring me, yet my 3rd post on this thread:
http://www1.ietf.org/mail-archive/ietf/Current/msg22198.html
Has already appeared (thank you Harald) while after some hours, the 2nd post I
made (this email) has not.
I will also be emailing Harald privately to find out why my repeated attempts
in last few hours to unsubscribe (so I would not be provoked by new posts) are
not being recognized by Majordomo (still receiving email from list).
Already been done, and better - Consider a virus that installs an open proxy
for spammers to use. Do the lit review yourself if you can't name which
one(s)
did this (yes, more than one has).l
Valdis what you describe is not the same as what I described. The virus must
run autonomously in order to have the effect at the scale I described. Such a
virus would want to be on the order of > 1 million computers sending > 10,000
random spams per day (or any combinations of the same product, e.g. 100,000
sending 100,000). If you assume that 10 billion spams are sent per day now,
and that DCC, Bayesian, etc catch 90% of them, then 10 billion undetected spams
would give inversion of performance (from 90% detected to 10% detected).
Do the lit review for which famous viruses created havoc by sending around
other attachment at random off a person's disk.
Viral attachments are easily to block, so you would not want an attachment in
the outgoing spam. Reread what I wrote last post about the potential virus
could spread itself orthogonally to it's spam function.
However, keep in mind that the spam can't be TOO randomized and still
convey a message
Conveying a message wasn't what I suggested a the virus could do. I proposed
it would simply disrupt antispam systems and wreck havoc on the email system.
It is a macho thing, such as the "ILOVEU" virus from the Philippines. I should
disclose that I am currently visiting the Philippines for a conference (field
research) on this (check my IP address). With a truely random content (except
normal words and normal word distribution, "normal" in uniform statistical
distribution sense), and with a huge volume, you need not care if any one reads
it. The only point would be to get past the antispam systems and users who
were formerly getting 90% antispam would be seeing more like 10% (missing a
zero in my previous post) antispam and 90% spam.
I wonder how actual field research other "experts" actually do on virus/spam
havens.
Already being done: Consider the following obfuscations seen in today's spam
No I meant truely random ordering of *normal* words. I usually mean what I
write. The *normal* words are needed to avoid Bayesian filters.
We're quite aware of the architectural problems. We're also aware of exactly
what it would take to deploy a solution....
Nice boast but imo you have proven otherwise in the way you handled my posts,
which is going to be quite clear to independent (outside) observers, when/if
the "mega spam virus" I described herein hits the world.
Lastly I have done the full background search at ASRG (IRTF), and I did not
find prior art for either the proposal I made to legitimize bulk email by
moving it to "pull", nor the prior art for our soon to be patent-pending
anti-spam algorithm.
Your search was incomplete, and here's some prior art.
The one you quoted is referring to "RSS" which is not entirely correlated with
what I proposed. I proposed using POP (or what ever the receiptient prefers)
which does not require a "complete overhaul of email clients" as quoted as one
of the complaints in thread you mention. I have long ago in the this list
readily admitted that message pull has existed for a long time, such as our
past discussion of usenet. Also the one you quoted does not discuss the
benefits I proposed, such as the ability to define spam at other nodes in the
channel than the pyschology of "unsolicited" and the benefits that follow such
a logic.
BTW, I noticed there were no reasonable objections in the thread you quoted
regarding overall concept of email pull.
Make sure that the
claims on your patent don't cover anything in this message, as that would
of course be a big no-no.
You are confusing 2 different things. Please read my posts more carefully.
The proposal I made here has nothing to do with the antispam algorithm we
developed. I stated "time-domain analysis" (idea only, no details found) as the
closest thing found at ASRG, but not substantial or correlated (just an idea,
no details) enough to be prior art, as Vernon and you claim.
Shelby Moore
http://AntiViotic.com