In These Times magazine, August 7th

Safety Net

By David Futrelle


Marty Rimm learned early on the value of a well-placed statistic. Nearly a
decade and a half ago, as a 16-year old student at Atlantic City High School,
he caused a stir in the local media with a study purporting to show that 64
percent of the students at his school had gambled in local casinos. The study
was roundly criticized by the casinos themselves--and it didn't exactly help
the flamboyant high-schooler's credibility that he disguised himself as an
Arab sheik in an attempt (successful, he contends) to infiltrate the Playboy
Hotel and Casino. But the study received a great deal of play in the media,
and led the state legislature to raise the age of legal gambling to 21. 

Today, another one of Rimm's statistics has pushed itself into public
consciousness in a big way: his claim, in a Georgetown Law Journal article that
formed the centerpiece of a recent Time magazine cover story on "cyberporn,"
that some 83.5% of images on the Usenet newsgroups available through the
Internet are pornographic. The figure has been cited on Nightline; it's been
mentioned on the floor of the Senate. It's routinely invoked in newspaper
stories and television debates. In a media culture built on soundbites, it's
become the soundbite of the moment.

There's only one problem with the statistic: it's almost certainly wrong--no
more accurate or scientifically valid than Joe McCarthy's famous claim that he
had assembled a list of 53 known communists in the state department. There is
no question that there is a considerable amount of pornography available
online, mainly through commercial "adult" Bulletin Board Systems (BBS's) and
Usenet "binaries" groups, where images are posted in coded form. A quick
glance at the binaries groups reveals that
alt.binaries.pictures.erotica.blondes fills up much more quickly than
alt.binaries.pictures.furniture. 

But, as anyone who has seriously attempted to study the elusive Net will tell
you, basic facts about the beast are terribly difficult, if not impossible, to
pin down with any degree of precision. Even Brian Reid, a respected researcher
at Digital Equipment Corp. responsible for the development of sophisticated
statistical techniques for measuring Usenet readership, acknowledges that his
numbers--based on aggregate date from an extensive list of sites around the
world-- may be off by as much as a factor of ten. Rimm, who conduced his
research as an undergraduate at Carnegie-Mellon university, based his number
on a much smaller, and almost certainly unrepresentative, sample--a one-week
survey of 32 Usenet newsgroups available at his site featuring "binary" image
files. 
 

Since its prominent mention in Time's July 3 issue, Rimm's study-- baroquely
titled "Marketing Pornography on the Information Superhighway: A Survey of
917,410 Images, Descriptions, Short Stories, and Animations Downloaded 8.5
Million Times by Consumers in Over 2000 Cities in Forty Countries, Provinces,
and Territories"--has provoked hundreds of pages of commentary online and off,
including detailed, and devastating, line-by-line critiques from notable
academics and researchers, some of whom Rimm himself had cited approvingly in
his text. Much of Rimm's study of Usenet, for example, is based on the
techniques developed by Digital's Reid. Yet when he first saw Rimm' study,
Reid was stunned. ``I am so distressed by its lack of scientific credibility
that I don't even know where to begin critiquing it," he wrote in a review of
the study posted online. "In this study I have trouble finding measurement
techniques that are *not* flawed.'' (See accompanying story.)

Rimm has hardly emerged from the controversy unscathed. Stories detailing the
study's flaws have run in publications ranging from the online newsletter
Cyberwire Dispatch to the New York Times. Philip Elmer-DeWitt, the reporter
who wrote the Time cover story, has admitted in online debates that he
"screwed up" in his handling of the report, and in a followup article in Time
on the controversy he acknowledged Rimm's growing credibility problem.
Carnegie-Mellon is looking into the possible ethical problems involved in
Rimm's analysis of the computer activities and downloading patterns of some
3,000 students and staff who for the most part believed their records were
private. Some of those listed as "contributors" in the study's footnotes have
explicitly disassociated themselves from the project. And Rimm was dropped
from a list of witnesses testifying at the recent US Senate Judiciary hearing
on children and computer pornography. 

But to focus merely on the flaws of Rimm's study is to miss the larger point.
How is it that an inconsequential and incompetent study--one put together by a
university undergraduate and world-class eccentric--managed to land itself at
the center of national debate in the first place? In part, the blame has to go
to Time magazine, which transformed a minor media event--the study's
publication in the Georgetown Law Journal--into a major one. And it didn't help
that Time illustrated its report of the sensationalistic study with even more
sensationalistic graphics--including a cover image of a young child staring
terrified and transfixed at some presumably lurid image on his computer screen
and a full-page image of a man engaging in sexual congress with a computer. By
raising the issue of children, moreover, Time also managed to misrepresent the
focus of the study, most of which is devoted to a discussion of commercial
pornographic BBS's with access carefully restricted to adults, and not to the
more wide open and ungovernable Internet.  

That Time's editors decided to give this story a sensationalistic spin is
hardly surprising. After all, the media has been awash in reports of
"cyberporn" for several months now; the Senate, after a debate hardly
exemplary in its devotion to mere fact, recently passed Sen. James Exon's
(D-NE) dangerous and ill-designed "Communications Decency Act," criminalizing
not only pornography but even routine profanity on the Net. (The House is
expected to vote on the legislation before its recess.)  

And while it is surely distressing that such a shoddy piece of research should
gain wide public exposure, it should hardly come as a surprise. The debate
over pornography  is one that has always been short on facts and long on
melodrama, where apocryphal stories and apocalyptic rhetoric make more
difference than scientific surveys. Most discussions on sexuality--whether or
not they presume to be based upon scientific fact--rely on "evidence" no more
compelling than that found in Rimm's study. 

Indeed, Rimm's 83.5% statistic is not the only invented fact to fill the air
in the recent debates. In a discussion of the report on ABC's Nightline, for
example, Christian Coalition president Reed recounted the story of one
youngster whose parents decided to hook up to the internet. Leaving him alone
for five minutes with computer and modem, they returned to find him staring at
a photograph of bestiality he had somehow found online. 

A riveting story, but almost certainly untrue, because such a turn of events
would be virtually impossible. Smut--particularly of such a specialized
nature--does not simply pop up on computer screens unannounced. First, you
have to have the necessary software, and know how to use it. You have to
realize that somewhere on the vast conglomeration of Usenet newsgroups there
are pictures that can be downloaded. You have to wade through a list of up to
10,000 or so groups to find the right one. You have to be able to tell which
posts in the group are pictures and which are not. You have to know how to
download and decode the pictures you want, and how to view them. Even for
those who know precisely what they want and how to get it, it's likely to take
a lot longer than five minutes. 

The Net--and particularly the network of newsgroups known as Usenet--is
unregulated, and largely unregulatable. As opponents of the Exon bill and
similar legislation have made clear, there is software available that can keep
children from accessing most online smut. And some online services already
provide "parental filters" for concerned customers. Like most things involving
computers, the filters are hardly foolproof. But, short of draconian
regulations of speech, mass arrests, and the virtual dismantling or disabling
of much the vast worldwide network, there is almost nothing that can be done
to ensure that the Net remain always and forever free of improper images and
words. Free speech may not always be pretty, but the alternative is much less
pretty indeed.  

****************

Sidebar: 

Rimm shot 

"Marketing Pornography on the Information Superhighway" breaks nearly every
rule of academic research--and then some. The evidence cited bears almost no
relevance to the conclusions advanced; the study's abstract seems to have been
grafted in from a different report altogether. Its categorical assertions are
nowhere proven in the study--and many seem simply unprovable. 

Tellingly, Rimm seems not to understand the notion of the "representative
sample." His survey of the origins of pornography on Usenet, for example, is
based largely on a survey of the five Usenet newsgroups on which "the largest
selection of sexual imagery was discovered." This is roughly equivalent to
basing a study of the publishing industry on the five thickest books at the
local bookstore. 
Astonishingly, Rimm seems to believe that his survey of pornography downloaded
by students and staff at one particular university--Carnegie-Mellon
university, a technically-oriented school with a mostly male population--is
representative enough to stand in for all universities. In a footnote
explaining his (circular) logic, Rimm simply asserts that "there is no reason
to believe consumption at the university studied differs from that of other
universities from which pornographic Usenet newsgroups can be accessed." A
university is a university is a university.  

In some ways the most remarkable thing about the Rimm survey is how utterly
unremarkable its results really are. Even if all of its statistics were
accurate, "Marketing Pornography" would not in itself prove much beyond the
mere existence of pornography online--in forums, such as private "adult"
BBS's, that are specifically and explicitly devoted to porn. Much of the
report is given over to the simple task of categorizing images--many of them
taken from the Amateur Action BBS, a repository of notoriously hard-core
imagery that is hardly typical of the internet norm. 

If the research is dubious, the researcher himself does not exactly inspire
confidence either. Everything Rimm does seems characterized by an almost
willful sloppiness. His e-mail messages and Usenet posts are filled with typos
and misspellings and basic grammatical errors. He responds to critiques with
indignant bluster and transparent evasion. 

And, though his research has been eagerly hailed by the religious right and by
anti-pornography crusaders such as Catharine MacKinnon, Rimm is hardly your
garden variety bluenose. In a self-published novel of several years ago, for
example, he described life and lust in the casino subculture in embarrassingly
explicit detail. 

Strangest of all, however, is The Pornographer's Handbook, Rimm's
self-published guide to the effective marketing of dirty pictures via modem.
Drawing heavily on the research underlying the more respectable
"Carnegie-Mellon" study, Rimm is nothing if not specific in his advice. "When
searching for the best anal sex images," he tells potential pornographers,
"you must take especial care to always portray the woman as smiling, as
deriving pleasure from being penetrated by a fat penis into her most tender
crevice. ... The slightest indication of pain can make some men limp."

Like everything else in this case, even the simple fact of whether or not Rimm
has "published" this book has become a matter of debate. He almost certainly
has written something by this name. The book is listed in Books in Print, and
lurid excerpts from it--which Rimm acknowledges are authentic--have been
floating around the Net. Yet no one seems able to turn up a copy of the book
itself, which Rimm now seems to want people to believe was simply a hoax. Rimm
told the Press of Atlantic City that the book was a "satire on the pornography
industry which was never printed, published, distributed or sold to anyone."
To Internet reporter Meeks, though, Rimm described it as a "marketing book."
It's hard to tell. After all, almost everything Rimm writes reads like a
parody.