MiSTed: CPSR Clipper Petition test results
Steve Brinich
Gypsy: Mike... Mike...
[Mike stirs, then wakes up, rubbing his eyes.]
Mike: Wha...?
Gypsy: The Mads are going to be calling shortly.
Mike: So? I've had my Invention Exchange finished for weeks. It kept
getting put off for weeks because of something happening at Deep 13:
the Mad Scientists' Award, the Mad Scientists' Conference on Oahu,
the lab explosion....
Gypsy: Yes, but you want to *look* like you've been working when they
call, don't you?
Mike: Maybe you're right. A bogus facade is worth two good solid
arguments. That's what I learned in college.
Gypsy: The college professors taught you *that*?
Mike: Well, yes and no. What I mean is, they didn't *mean* to teach
me that, but they did anyway....
[Commerical sign lights]
Mike: We got commercial sign. [Hits button]
[SOL -- Gypsy, Crow, and Tom Servo are listening to Mike talking. Mike
is still lounging in his hammock.]
Mike: ...and the key is, figure out your conclusion first, and you
can always come up with the arguments later.
[Incoming Call light flashes]
Mike: Speaking of arguments, Lincoln and Douglas are calling. Give
me a hand with this, will you, guys?
Tom: You gotta be kidding.
Mike: Oh, right. Never mind.
[Mike gets out of the hammock, stretches, and takes down the hammock.
He then steps over and hits the incoming call button.]
[DEEP 13 -- Dr. Forrester stands behind the console. TV's Frank is
seated on some sort of excersize bike with various attached gizmos.]
Dr F: Oh, there you are, labrat. I was starting to think you'd stepped
out for a minute. Are we ready for the Invention Exchange?
Mike: The what?
Dr F: [annoyed] The Invention Exchange. We demonstrate one of my
inventions to you, and you show us whatever it is you managed to
tinker together.
Mike: Oh, that. It's been so long, I just about forgot the whole thing.
Dr F: [speaking through gritted teeth] Oh, did you? [hand moves toward
the Shock To The Shammies Switch]
Mike: [quickly] But I do have something ready.
Dr F: [pauses just short of the switch, looks up] Oh?
[Mike pulls out his boot with various attached suction cups, Velcro pads,
pitons, etc.]
Mike: Another thing I can hardly remember is the bad weather you
get on Earth sometimes. In fact, when it's really bad, I'll bet you
wish you were here and I was there.
Dr F: No. I don't. Get on with it!
Mike: Anyway, as you can see, this shoe is designed to maintain its grip
in any type of inclement weather. The suction cups cling to wet
surfaces, the cleats and pitons provide traction in snow and mud,
and the blades help you coast right over ice.
Dr F: Interesting. What does the Velcro do?
Mike: That helps.... [pause] That helps hold on the suction cups,
cleats, pitons, and blades!
Dr F: [sneering] Oh, *that's* really advanced engineering! Now, let me
show you something that'll really keep your feet moving. [To Frank]
Frank, let 'er rip!
[Frank begins pedalling the excersize bicycle.]
Dr F: At first sight, this appears to be an ordinary excersize machine.
However, I've installed a few additions to overcome the main problem
with gym equipment.
Crow: You've covered it with deodorizer?
Dr F: No! Don't be ridiculous! I mean the problem of backsliding.
For instance, I bet that Franks would like to just quit right now,
wouldn't you, Frank?
Frank: Yeah!
Dr F: But, you see, he can't --
[DEEP 13: New view from further back, showing a large blade set to
swing at Frank, a la "The Pit and the Pendulum", held by an
electromagnet and tied in place with twine.]
Dr F: because once I cut this twine [he pulls
a scissors out of a lab coat pocket and does so], the only thing
holding back this blade is this electromagnet, which is powered by
his exertions.
Frank: How long do I have to do this?
Dr F: Until I tell you to quit! [to Mike] Now, today's experiment
is a little excersize in statistics.
Tom: You mean like numerology?
Dr F: Not quite... but close enough, in this case! Ha Ha Ha! [to
Frank] Start the experiment, Frank.
[Frank stops pedaling and gets off the bicycle. The swinging blade
misses him by inches as he stands, swinging toward Dr. Forrester.
It grazes him, ripping his lab coat.]
Dr F: FRANK!!!
[Frank quickly steps around Dr. F and pushes the Send button.]
[SOL -- Experiment lights and alarms go off.]
Mike: We got number sign!
[DOOR SEQUENCE]
>Article: 2770 of alt.privacy.clipper
>Newsgroups: comp.org.eff.talk,alt.privacy.clipper,alt.security.pgp,
>talk.politics.crypto,alt.security.ripem,alt.fan.david-sternlight
>From: strn...@netcom.com (David Sternlight)
Crow: Sterno!
>Subject: CPSR Clipper Petition test results
>Message-ID: <strnlghtC...@netcom.com>
>Followup-To: poster
>Organization: DSI/USCRPAC
>Date: Sun, 27 Feb 1994 06:36:22 GMT
>Lines: 93
> CPSR posted a notice asking for "votes" against Clipper in the form of
>e-mail messages saying "I oppose Clipper".
Tom: What do you expect them to do? Ask for votes in the form of
E-Mail messages saying, "Elvis Lives"?
> Within about a week they
>claimed "10,000 have signed." I questioned whether these were genuine
>signatures--
Crow: Because, after all, there's no way the Net could have possibly
handled 10,000 messages in only a week.
> in particular whether each was a real, and distinct person.
Mike: I guess that leaves out bots....
Tom, Crow: [angrily] Hey!
>I also sent them a message asking how they checked.
>
> That message was never answered. Meanwhile they claimed 20,000,
>then 30,000.
Crow: Well, it'd be pretty odd if they claimed 30,000, then 20,000...
>
> Given that some here are cypherpunks skilled at creating bogus
>identities on the net,
Tom: I can deny ever writing this stuff.
> it would be a simple matter to fake many of the
>"signatures".
Mike: And an even simpler matter to make unsupported allegations.
> Thus some test is needed before they can be accepted at
>face value, particularly given the strong feelings and heat
Crow: Generated by Sterno Flames.
> associated
>with this topic.
>
> As part of my message I pointed out that there were only 10-20 who
>were very vocal here on this topic.
Mike: Huh? He thinks only people who post regularly are real?
Tom: [announcer voice] Usenet. Real talk for real people.
Mike: But what about the thousands of people who read the Net
without posting?
Crow: They're not real.
Mike: But I met a lot of them on Earth....
Crow: Forget it. They guy you thought was liveware is another
techie hack.
> Thus there would have to be a large
>silent majority"
Crow: Of people who support President Nixon.
> for the counts claimed by CPSR to make any sense. One
>reader posted a message urging those "silent" members who had signed to
>e-mail me, so stating.
Tom: [Dragnet announcer voice] This person has been sentenced to
5 years of public ridicule for incitement to mail-bomb.
>
> I decided to make this the subject of an experiment.
Mike: Hey! This guy sounds like one of the Mads!
Crow: I don't think he means that kind of "experiment".
Tom: Even so, he *still* sounds like one of the Mads....
> Originally I
>decided to use the Statistical Decision Theory approach: assume
Crow: When you "assume", you make an ass of you....
[pause]
Mike: "And me."
Crow: Yeah, you, too.
Mike: [slaps forehead] D'OH!
Tom: Good one, Crow!
> the
>CPSR signatures were genuine and evaluate the probability that I'd receive
>whatever number of signatures I would, given that prior assumption.
Crow: Let's see.... probability that any messages that even mention
Sterno get through the kill file, times probability that the newsgroup
is accessed in time, times probability that the reader thinks Sterno
is worth the keystrokes it takes to send him E-Mail.... Darn!
Mike: What happened?
Crow: That last calculation locked up my calculator. What does
"underflow" mean?
>If the probability was at all reasonable it would support the hypothesis.
Tom: That thousands of Usenet readers have no lives.
>I expected some reasonable number of responses roughly related to the
>10,000 and was prepared to be very loose (give CPSR the benefit of
>every doubt) in my evaluation.
>
> In order to give them the benefit of every doubt, it was my intention
>to accept every confirmation sent to me as genuine
Tom: Send me E-Mail, no questions asked! Please!!
Mike: I guess it's lonely when nobody thinks you're worth talking to....
> unless it was blatantly
>bogus on its face. Given that, and given the possibility of spoofers,
>it was important to be vague about the methodology and criteria so
Crow: That I could change it as required to make the results come out
the right way.
Tom: Ain't science wonderful?
>spoofers didn't flood my mailbox. Though I didn't say so, some assumed
>I was going to check the mail addresses carefully,
Mike: While others assumed I would just go off half-cocked as usual.
> check for bogus
>origins, etc. Since I planned to give CPSR the benefit of every doubt,
>I didn't plan to do that. But I wanted that impression to remain so
>that spoofers would not be tempted.
Crow: [sarcastic] Yeah, corresponding with this guy is a major
temptation, right up there with junk food, booze, and women!
>
> In the event, each message I received was a "human" message--that is
>there were intelligent and literate comments with each one,
Tom: [stereotypical dumb surfer voice] It's, like, so bogus how they
don't make us bots smart like humans!
Crow: [stereotypical Valley Girl voice] Rilly! It is *such* a bummer
getting stuck in remedial English every year!
> and each had
>a somewhat different style. I am persuaded on inspection that they are
>all genuine and I count every one.
>
> So how many did I get. 5000? 1000?
Crow: He really expected to find thousands of people out there with nothing
better to do than send him E-Mail??
> 500? 100? 50?
Tom: [auctioneer voice] OneHundered!DoIHearFifty?Fifty!FiftyGoingOnce...
>
> No. Over a week I got 35 messages.
Tom: [auctioneer voice] SOLD! To the man in the mismatched socks and
fish tie for Thirty-Five!
Mike: Gee, think about the boring existances of those 35 poor people....
>
> The result seems so dramatic that no amount of "explaining away"
Crow: Can disguise the fact that most people dismiss me as an
insignificant pest.
> (and
> I expect there will be many attempts by the anti-Clipper crowd) can
>convince me
Tom: [snooty voice] Ooooo, we must *convince* the Great and Powerful
Sternlight!
> that CPSR got anything like the number of genuine and
>different people responding comparable with their claims. The result
>is so dramatic that there's no need to construct any Bayesian priors,
>and distributional assumptions about the response process, and do any
>posterior probability calculations.
Crow: Calculate the probability of this posterior! [Gets up and
turns around as if to "moon" the screen]
Mike: Crow!
>
> Though CPSR isn't the hoaxer but the victim of the hoax, it's a hoax,
>people.
>
> If CPSR wants to contravene this conclusion the only way they can
>convince me
Tom: Yeah, like that's their Number One priority!
Mike: I don't think it's even their Number One Thousand Four Hundred
and Seventeenth priority....
> (dunno about others) is to check each signature's mail path
>and "from", and actually investigate a random sample (speaking
>professionally) of the signatures. I'd suggest at least 500 randomly
>selected replies, using standard random sampling to pick the 500.
Crow: Yeah, like a guy who draws conclusions from his incoming E-Mail
knows about "random sampling"! Bite me, pal!
>
> I expect this message to be widely attacked
Tom: Oh, good. At least he can't accuse us of sneaking around
behind his back.
> and all sorts of "logical"
>arguments attempted to refute my conclusion. It won't play, fellas. 35 out
>of 30,000.
Crow: ...people believe they are in regular contact with space aliens.
> No way in hell.
Mike: Will my net.reputation ever recover from this post.
>
> The experiment is now over,
[Tom and Crow get up to leave.]
Mike: He means *his* experiment, not the Mads' experiment.
Tom: Sorry. I got them mixed up.
Crow: I think the Mads are closer to real science....
> so any further responses will be ignored.
>CPSR posted a one-shot call for messages of support in the relevant
>newsgroups, and the individual who suggested this set of responses posted
>a one-shot message asking for them.
Mike: And both messages carry exactly the same weight with the net.public.
> Now that this message will become a
>cause celebre,
Tom: [laughing] "Cause celebre"? Listen to Mr. Humility here!
> it may tempt more respondents, but since CPSR didn't
>repeatedly ask for signatures in the same newsgroups, responses stimulated
>by this message wouldn't count anyway.
>
> I'm done with this topic,
All: YAY!!
> and don't plan to reply to any messages
>attacking me, the approach, the data, or the comments here.
Mike: Nobody replies to these MiSTings anyway.
Tom: Well, there *was* this one guy who got mad and threatened
to sue....
Mike: [alarmed] What?
Crow: Before your time. Don't worry; it'd get laughed out of court
anyway.
> I do not
>intend to enter an argument about this. I am satisfied, and if others
>aren't, they're certainly entitled to their opinions.
>
> Note than in an attempt to keep hysterical responses out of the
>newsgroups.
Tom: I have written this sentence fragment.
> I've placed "poster" in the Reply To: Field hopefully to
>create a pause for thought. Responses will go to me as e-mail
Mike: Wow! This guy *is* desperate for E-Mail!
> unless
>the respondent edits the header field of his reply to make it post to
>the various newsgroups. As I've said, I don't plan to reply to these
>responses,
Tom: Promises, promises....
> though I do read my mail.
>
>David
>
>--
>David Sternlight If you want to get somewhere, it's easier to move
> your sail than to try to change the wind.
[DOOR SEQUENCE]
[SOL -- Tom Servo is working on calculations at the blackboard]
Mike: What are all those figures?
Tom: I did a statistical analysis of the people who responded to
this petition, taking into account the number of places it has
appeared and a psychological profile of the net.public, and
reached the following conclusions:
47.3% -- Never saw the request to send E-Mail to
Sternlight, because they heard about the
petition somewhere else.
23.8% -- Never saw the request to send E-Mail to
Sternlight, because they long ago set up
bozo filters to reject any message that
even mentions his name.
12.7% -- Saw the request to send E-Mail to Sternlight,
laughed themselves silly, and forgot
the whole thing.
10.4% -- Saw the request to send E-Mail to Sternlight,
thought about it, and decided that it
was wrong to letter-bomb, even a net.bozo.
5.7% -- Saw the request to send E-Mail to Sternlight,
thought about it, and decided that they
have more important things to do, like
optimize their hard disks.
0.1% -- Saw the request to send E-Mail to Sternlight,
had nothing better to do, and did it.
Mike: That's pretty impressive. How did you get those figures?
Tom: I used a Statistical Decision Theory approach.
Mike: What does that mean in English?
Tom: I guessed.
Mike: Let's see what the Mads make of all this....
[DEEP 13 -- Dr Forrester and TV's Frank are tinkering with the excersize
bicycle.]
Dr F: That should speed up the swing so the user can't sneak off.
Frank: OK, Steve. Tie the blade in place so we can test it again.
Dr F: What? *Me* tie the blace in place? That kind of menial task
is what I hired *you* for! Go tie the blade in place!
[Frank walks over and ties up the blade. Dr F gets on the bicycle and
starts pedaling. Frank cuts the twine, ambles over to the console,
and lounges back with a bag of chips and a soda.]
Frank: Oh, hi, guys. We're starting the long-term test of our invention
now, so I'll have to sign off. [Frank pushes the button.]
Dr F: [voiceover] Long-term test?? FRANK!
MiSTing by Steve Brinich
\ | /
\ | / Mystery Sceince Theater 3000 and associated characters
\|/ are the property of Best Brains, Inc. and used here
----O---- for satirical purposes only.
/|\
/ | \ This post is not intended as a personal attack upon
/ | \ the original author, and is meant only as entertainment
and commentary on the content of the original post.
> As part of my message I pointed out that there were only 10-20 who
>were very vocal here on this topic. Thus there would have to be a large
>silent majority" for the counts claimed by CPSR to make any sense. One
>reader posted a message urging those "silent" members who had signed to
>e-mail me, so stating.