IP: CWD--Data Misbehavin'

[David Farber <farber () central cis upenn edu>]

Date: Mon, 9 Oct 1995 20:21:01 -0700
From: "Brock N. Meeks" <brock () well com>
To: farber () central cis upenn edu








CyberWire Dispatch // Copyright (c) 1995 //


Jacking in from the "I'll Take Two From Column A" Port:


Washington -- A study released by Simba Information, Inc.,
has left the company scrambling to pay for misappropriated
data under the threat of legal action and issuing public
apologies to two major universities for lifting copyrighted
data without permission.


On August 22, Simba issued a press release touting "an
important finding" for Maalox-guzzling executives trying to
figure out how to make a buck off the Net.


Simba said its new study, "On the Internet: User
Demographics and Trends," showed "most Internet users appear
willing to accept the Web as a viable commercial medium."
Yow-Za!  Finally, hard data that corporate suits could use
to justify dumping all that dumb money into risky Web
ventures.


Simba Vice President Tom Niehaus is quoted in the second
paragraph saying:  "Almost 80% of current subscribers say
they would use the Web as a commercial medium, provided
that, as with any other consumer marketing proposition, the
quality of information and services was of fair value to its
pricing."


Meanwhile, savvy Web-surfers were left scratching the
underbellies of their collective mouse pads.  You see,
months earlier they could have read the following from a Web
site holding research done by the Hermes Project at the
University of Michigan Business School:  "Almost 80% of the
respondents are willing to pay for WWW access and services.
However, this is conditional upon appropriate quality and
price."


You can almost hear Yogi Berra: "It's like deja vu all over
again."




The catch?  To read the Simba report you have to pony up
$995; to read the Hermes report costs you nothing but
time.


Double Bogey:  The Simba report draws heavily on Hermes data
for the body of its report, despite a specific disclaimer in
the Hermes copyright that prohibits the use of the data for
profit-making.


Triple Bogey:  Simba never bothered to contact the Hermes
project to ask permission to use its data.


The Hermes Project director, University of Michigan
Professor Sunil Gupta, only learned of the lifted data when
another researcher, Jim Pitkow of the Graphics and
Visualization Lab (GVU) at the Georgia Institute of
Technology, notified him by Email.  Pitkow, who had worked
closely with Gupta and had completed a similar study of Web
demographics, also discovered that his research had been
used by Simba, again without permission and against specific
copyright disclaimers that the data could not be used in a
for-profit venture.


This squalid little data-heist hit the sunshine during a
little dust-up in an open conference on the WELL.


After the study was published, it was plucked from the
information flotsam and jetsam by HotWired columnist David
Kline, who ginned up a trick multiple choice quiz for his
readers so they could test their Net-awareness quotient.
The quiz was well done and informative, and all was right
with the world (just ask O.J.).  Then Kline ran into a
buzzsaw named Donna Hoffman.


Hoffman, a Net-Celeb in her own right, is a researcher at
the Owen Graduate School of Management at Vanderbilt
University.  It was Hoffman, and her partner Tom Novak, who
helped expose the bogus research underlying the "cyberporn
study" done by former Carnegie Mellon University student
Marty Rimm, which Time magazine shilled on its July 3rd
cover.


After reading Kline's HotWired column, Hoffman admonished
him in a public forum on the WELL, taking him to task for
promoting stolen research.


She said that Simba had used data from Hermes and GVU
without permission, violating copyrights that specifically
stated that the information was not to be used "for profit."
And at $995 a pop, Simba was definitely pocketing some
for-profit coin.


Kline launched a blitzkrieg investigation.  Within two hours
of being tipped by Hoffman, he confirmed that Gupta's data
had been lifted without permission and had confronted
Simba's President Alan Brigish with "disturbing facts" that
alleged copyright violation.


Brigish tried to buy time, telling Kline that the report had
been written by a freelancer, Peter Clemente, and that it
was Clemente's responsibility to have obtained all necessary
copyright clearances.  Nevertheless, Brigish told Kline he
would "look into the matter."


Kline reported his findings back to the WELL, noting he
wasn't buying Brigish's explanation.


But in trying to fend off Kline's pointed questions, Brigish
conveniently left out another piece of the story:  fully 50%
of the report was actually data published in 1994, lifted
from a study by another marketing research firm, FIND/SVP.


FIND/SVP had given permission to Clemente to use a limited
amount of data.  But when FIND/SVP found out that half the
data in the Simba study was their own work, they demanded
payment.  Brigish balked.


A series of furious point-counterpoint faxes flew back and
forth.  Finally, under the threat of a lawsuit, Brigish
coughed up a payment to dodge a nasty public spectacle.
Under the terms of the payment, neither side is allowed to
talk about the suit nor divulge the amount of payment,
according to sources familiar with the internal wranglings
of the case.


While the lifting of the academic research had caught
Brigish off-guard, questions about the "keep quiet" payment
to FIND/SVP blind-sided him.


When asked to confirm the report that this company paid for
the FIND/SVP data -- after the fact -- Brigish bristled:
"I'm not going to discuss this further."   When I pressed
him on the issue, asking him for comments on both copyright
issues, FIND/SVP's and the universities', he said I was
"pressuring" him and that he was "going to hang up the
phone."  He left me holding the receiver, the words "I'll
get back to you" trailing off in my ear.


My next call went to Gupta.  He said the discovery that
Simba had appropriated research "really bothers me."  He
said that all along he had been telling respondents his
research was being done "for non-profit" reasons.  "Now we
learn there are people making big bucks off this," he said.


He worried that if people found out about Simba
incorporating their data in a for-profit publication it
would put the Hermes project's integrity at risk.


Gupta did acknowledge that he had received a voice mail
message from Brigish, notifying him that he was "looking
into the matter."


Gupta had a rough idea of what Simba would have to do to
quell the matter: "Somehow he [Brigish] has to make clear
that what happened ... was absolutely terrible, that's a
basic, basic, requirement.  A public apology, also, I think
s reasonable."  And after thinking for a moment he added
that Simba might "make a bigger donation to the charities,
that would be cool."


Charities?  Curious statement, until you realize that as a
way of saying "thank you" to the participants in the Hermes
study, Gupta makes a payment to three charities.
Participants choose their favorites and at the end of the
study period, Gupta takes the top three vote-getters (two
domestic and one international) and writes a $500 check to
each.  Nice touch.


So where's freelancer Clemente in all this?  To his credit,
he readily cops to his mistakes:  "I neglected to call the
principals [Gupta and Pitkow] of the [university] studies
and I should have."  Clemente said he was thrown off by the
home page of each study that states the research is "free"
for anyone to use, as long as proper credit is given, which
Clemente does do in the body of Simba report.  "I feel
terrible about this," he said, adding that he had spoken
with the researchers and apologized.


Brigish did make good on his promise to "give a statement."
But he wouldn't take calls on it;  he sent it via email.  In
it he insists that "reports of copyright infringement" in
conjunction with the Simba study "are incorrect."  He says
the company "takes copyright issues very seriously."
Brigish then throws Clemente to the lions, saying that it
was his responsibility alone to gain the proper permissions.
But then he tosses in this caveat:  "We should have insisted
that [Clemente] get written permission to use the data."


I checked with a few other market research firms on their
own internal procedures in matters like this.  Each one said
they have freelancers sign statements asserting that all
copyright permissions had been negotiated, but each also
double-checks those permissions, a deliberate effort to
dodge a Simba-like copyright snafu.


Brigish also acknowledges that "Simba clearly erred" by not
citing the sources of its data in the marketing materials
for the study, "and we regret this omission."  Brigish's
statement says he will "seek ways in which Simba and both
universities can work together in the future to support
Internet demographics research and make that data widely
available."


Brigish also promised to issue his statement to reporters.
In fact, he sent the statement only to Kline and myself.  In
addition, the statement was written up on the Cowles/Simba
Daily news service.  However, the initial press release went
out over the Business Wire and PR Newswire, reaching
potentially thousands of news organizations.  The Simba mea
culpa statement was never put on those wires.


In addition, Simbas said it would donate $5,000 to charities as
identified by Gupta when his next research study was complete.


That should have settled things, but Gupta claims otherwise.
The professor insists that during discussions with Brigish,
he agreed to provide "significant research funding" to the
universities."  But that point never made into the official
statement.  Gupta says Simba is now going back on the
promise.  He's also ticked off about the limited
distribution of the statement.


Brigish says he doesn't recall the offer to supply funding
to the universities, saying only that Simba's Niehaus would
discuss with Gupta "appropriate collaboration."  Further, he
says, via Email to Gupta, that the limited distribution of
his statement is enough:  "I see no reason to broadcast it
widely as a press release."


The whole affair has caused Gupta to wash his hands of any
dealings with Simba.  On October 8th, he informed Brigish
that although he had "hoped to find some positive way of
emerging from this episode," none is forthcoming.  "Your
response to [my concerns] convinces me that you have not
fully understood ... and leave me doubtful regarding just
how seriously you do take such breaches of ethical
behavior."


In a pure "well to hell with you" gesture, Gupta also says
he is refusing the $5,000 because it's "best not to have"
Simba's name "associated" with the research efforts of the
two universities "in any manner."  Nor will he entertain
"furher opportunities of associating with your company."


Righteous indignation is great, but hey, professor, you
could have at least made them cut the check for five grand
before you bailed on the deal.


Kline, however, does let his indignation hang out.  In this
week's Market Forces column for HotWired, Kline opens
saying:  "There is probably nothing more disturbing to a
journalist than discovering that he's been unwittingly used
to deceive the public."   But that's just what happened
during "Simba-gate" as he calls it.


"Frankly, it makes my job a hell of a lot tougher," Kline
told me.  "Do I now have to start checking for copyright
violations, too?  All I know is, I was misled, and as a
result, I misled my readers.  That seriously pisses me off."


This episode of "Simba-gate" also hurts the "admirable goal
of providing publicly-available, non-proprietary information
to the Internet community," Hoffman says.  Once the data-
for-profit link is made in people's minds to the GVU/Hermes
studies, "we run the risk" that people will "cynically
approach such surveys the next time," she says.  "This hurts
the ability to obtain cooperation from the Internet
community on subsequent studies, and undermines the
credibility of objective data collection efforts."


Yep, and it doesn't do a whole hell of a lot for those that
forked over the 995 clams, either.  Now ... where did I
put that Maalox anyway?


Meeks out...