1997-03-13 - Re: Is Graham-John’s inane spam robogenerated?

Header Data

From: dlv@bwalk.dm.com (Dr.Dimitri Vulis KOTM)
To: cypherpunks@toad.com
Message Hash: 972b479b66875cc64b245b46c2a1c48eeecf2a32efa2dcc6c6fdc4757071de4b
Message ID: <k3ci4D12w165w@bwalk.dm.com>
Reply To: <199703131139.FAA11708@manifold.algebra.com>
UTC Datetime: 1997-03-13 14:45:47 UTC
Raw Date: Thu, 13 Mar 1997 06:45:47 -0800 (PST)

Raw message

From: dlv@bwalk.dm.com (Dr.Dimitri Vulis KOTM)
Date: Thu, 13 Mar 1997 06:45:47 -0800 (PST)
To: cypherpunks@toad.com
Subject: Re: Is Graham-John's inane spam robogenerated?
In-Reply-To: <199703131139.FAA11708@manifold.algebra.com>
Message-ID: <k3ci4D12w165w@bwalk.dm.com>
MIME-Version: 1.0
Content-Type: text/plain


ichudov@algebra.com (Igor Chudov @ home) writes:

> Dale Thorn wrote:
> >
> > Igor Chudov @ home wrote:
> > > Graham-John Bullers wrote:
> > > > Vulis time to take your pills.
> >
> > > I am not a native English speaker, and am curious about something.  Mr.
> > > GB does not put a comma after "Vulis", although I think that a standard
> > > English practice is to do so. That makes his messages look kinda cute. I
> > > wonder whether there is any hidden meaning in omitting commas in this cas
> >
> > It was obvious to me (a native speaker) early on that the messages
> > from GB referring to Dr. Vulis were auto-generated.  It's fairly
> > easy for a human to type in the exact same response over and over
> > (doing so manually), but to vary the hand-typed messages, while main-
> > taining the complete lack of emotion and sense of robotic blandness,
> > would require a great deal of attention that these messages would
> > not justify.
> >
>
> dale,
>
> i am sure they are not machine generated. the content is different
> every time, plus it srt of depends on the context to which he is replying.
> like, if vulis's article is about sexual perversions, GB calls him a
> pervert.
>
> one still could do that in perl, but it is not likely to be the case

I've been writing a program (in C, actually, although perl might be
a good tool for strings and such :-) that would scan Usenet newsgroups
for trigger keywords and generate randomized follow-ups depending on
what's been said. It's a big project; I hoped to have it done by
April 1st, but it'll definitely take longer.

I hope that with enough tweaking my spambot will sound less robotic
than Graham-John('s?).

---

Dr.Dimitri Vulis KOTM
Brighton Beach Boardwalk BBS, Forest Hills, N.Y.: +1-718-261-2013, 14.4Kbps





Thread