From declanm@netcom.com Wed Jul 19 00:27:08 1995 Return-Path: Received: from po8.andrew.cmu.edu by mail4.netcom.com (8.6.12/Netcom) id AAA17606; Wed, 19 Jul 1995 00:23:40 -0700 Received: (from postman@localhost) by po8.andrew.cmu.edu (8.6.12/8.6.12) id DAA08620; Wed, 19 Jul 1995 03:23:52 -0400 Received: via switchmail for fight-censorship+@andrew.cmu.edu; Wed, 19 Jul 1995 03:23:48 -0400 (EDT) Received: from po2.andrew.cmu.edu via qmail ID ; Wed, 19 Jul 1995 03:22:09 -0400 (EDT) Received: from netcom18.netcom.com (netcom18.netcom.com [192.100.81.131]) by po2.andrew.cmu.edu (8.6.12/8.6.12) with ESMTP id DAA06020 for ; Wed, 19 Jul 1995 03:22:02 -0400 Received: by netcom18.netcom.com (8.6.12/Netcom) id AAA16159; Wed, 19 Jul 1995 00:20:45 -0700 Date: Wed, 19 Jul 1995 00:20:44 -0700 (PDT) From: D B McCullagh Sender: D B McCullagh Reply-To: D B McCullagh Subject: The CMU Sting Operation To: fight-censorship@andrew.cmu.edu Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Status: RO X-Status: Following are some cogent observations by Jim Thomas (jthomas@sun.soci.niu.edu), who gave permision for his posts to be redistributed and quoted from, with attribution. His early ethics critique of Rimm's study is at: http://www2000.ogsm.vanderbilt.edu/ In the enclosed messages, Thomas says: "That Rimm and Sirbu then submitted a grant to the DoJ that could be used to bust the very people who were his subjects goes beyond any breach that I can recall, ever, in the social sciences." He also notes how Rimm's principal advisor was embroiled in this maze of ethical problems. One would think that Marvin Sirbu -- a big name at CMU as the director of the Information and Networking Institute -- would know better. But then again, he's working on a couple projects, like a Mellon Bank-sponsored digital cash venture, that could stand a little publicity. -Declan --- Topic 1043 [media]: TIME, Martin Rimm and the CYBERPORN scandal #495 of 498: jim thomas (jthomas) Sun Jul 16 '95 (22:58) 41 lines I find that grant application to DoJ more shocking than any previous revelation in this whole affair. The grant was an attempt especialy by Rimm and Sirbu (Banks recently withdrew) to peddle the linguistic parsing software to an agency so (according to Banks) Rimm could make some money. According to Banks, Rimm was aware of the Thomas/AABBS bust, and thought that his "skills" might be useful. The grant had four basic research goals in examining Adult BBS data: 1) It would summarize the stats of porn traffic on individual BBSes on the theory that it would help LE target high volume boards. BBSes with high traffic would presumably be worth prosecuting. The summary stats would calculate the percent of TRAFFIC (as opposed to number of pictures) 2) It would examine trends over time. If porn/pedophile traffic increases, then it indicates that the BBS is trying to cultivate that market 3) It would examine info on individual downloads, using covariance analysis to ascertain users preferences among categories of file types. 4) It would calculate the "worst" offenders (BBSes and perhaps persons) of pedophilia and other images. The professed intent was to use existing data from previous federal investigations to allow LE to utilize their resources more effectively and efficiently. There is nothing inherently wrong with this type of project. The problem in this case is that funding is being sought by researchers who studied a group of human subjects and now intend to use the study to assist in identifying and prosecuting those, ch as Robert Thomas, who were previously research subjects. The problem is compounded by the fact that Rimm deceived those subjects to gain their confidence and access to their files. To my mind, this is a scandalous violation of research standards. Perhaps the New York Times, Time, and other media do not find such a breach as heinous as might researchers, but I would hope they would not, in the future, trivialize the issue by implying that it's little more than an arcane academic debate. --- #477 of 478: jim thomas (jthomas) Sun Jul 16 '95 (21:54) 30 lines Back to Mike's early posting of yesterday: Rimm, and by proxy, Sirbu, stepped well over established lines with their human subjects procedures in dealing with BBS sysops (especialy Robert Thomas). That Rimm and Sirbu then submitted a grant to the DoJ that could be used to bust the very people who were his subjects goes beyond any breach that I can recall, ever, in the social sciences. That Rimm renewed his subscription to AABBS this spring *after* writing the grant that would assist law enforcement in identifying and prosecuting a Thomas and others who had been his subjects is not a minor lapse of ethics or an error in judgment. It is a fundamental violation of the most basic principles of research ethics. I am saddened that Peter Lewis, in his NYT story, glossed over these ethical issues. That Lewis's concluding quote suggested that the problem was simply a matter of provocative or sensationalist style ignores the ethical issues raised. I am also saddened that, despite what has been written here, Peter Lewis reduced the primary ethical issue to the Usenet readership statistics. This was not the primary ethical issue, and has virtually nothing to do with the human subjects issues I have raised. This is not a minor factual error. It is a serious inaccuracy, because it transfers attention from the serious problem of the BBS sysops as human subjects to the very different issue of CMU system user monitoring. Despite the value of Lewis piece as an antidote to some of the hysteria, I found the discussions of ethics disappointing. --- media.1043.360: Jim Thomas (jthomas) Sat 15 Jul 95 13:06 > He not only *is* a jerk, he plays one on TV! I'm wondering about the advice, if any, that Sirbu gave Rimm. It's clear that Sirbu had full knowledge of the project throughout. Initially, Sirbu, Rimm, and Banks authored the grant proposal to DoJ that would apply the linguistic parsing software to data provided by DoJ. The proposal was rewritten to exclude Rimm, because he wouldn't be at CMU, which could complicate the grant administratively, although he would be written in as a consultant. I'll leave it to others to answer the many questions this raises. In answer to a private question: No, I'm hardly a rigid ethical purist in matters of research. But there are some lines one simply never crosses. Rimm, et. al, have crossed these lines, moving well beyond anything that any serious scholar would consider even remotely acceptable. That Sirbu doesn't recognize the ethical issues raised by using his BBS analysis to create problems for the subjects of that analysis is simply mind-boggling. --- media.1043.362: jim thomas (jthomas) Sat 15 Jul 95 13:23 > I should point out that I asked Rimm, point blank, if he had ever been > involved with *any* law enforcement officials, in *any* capacity, regarding > the AABBS case. > "No." was his reply. This could be technically true, especially if he framed the question as meaning an investigatory capacity. My understanding is that Sirbu did the front work on the grant contacts, although I'm unclear on the actual logistics/division of labor. It's also my understanding that the idea for the grant came after the AABBS bust. My understanding is that in early Nov, Rimm met with two faculty, one of whom was Sirbu, and Rimm indicated that he had sunk much of his own money in the project. A grant possibility was discussed, and the two most likely were NSF and DoJ. Says the source: "Also, Rimm was aware that DoJ had busted AABBS and seized their log files. And that info should contain names, logs files, and it seemed reasonable to suspect that DoJ would have that set analyzed and that they would pay money for it." Somebody else can likely confirm this, but the thinking was that because DoJ had the Thomases BBS info that contained images of every category, the researchers could do a covariance analysis on BBS users, the types of files they download, match categories to users, and run correlations beteen users and categories. It would also be a means of assessing which users were accessing the files of interest to law enforcement, which would help LE optimize resources by going after those boards (and users) that seemed most serious. THe grant attempted to link the researchers' interests with DoJ prosecutorial interests. The grant itself sounds intellectually interesting, and from what I can tell, it's a sound project. The problem, of course, is the history and the context of the proposal. --- Topic 1048 [media]: Media, Academia and Ethics - The Search for the "Smoking Burnoose ". #74 of 75: jim thomas (jthomas) Tue Jul 18 '95 (14:25) 86 lines I'm agreement with cooljazz that the Rimm study generates a number of questions. I also emphasize that I'm not, and in fact oppose, ethical absolutistism. I was drawn into examining the quality and ethics of the Rimm study largely because of queries (such as cooljazz's) that required examining questions in detail and elaborating responses. I am part of no "i-team" or orchestrated attack. However, given the seriousness of some of the ethical flaws of the Rimm study, I do believe that an inquiry by CMU is required, and that some action be taken by them. I also strongly believe that action should be remedial rather than punitive. > jthomas is a big player in articulating the ethical side. But I do > think his position is open to question and debate. My specific concern > and I'll say academic concern is that ethical errors, lack of informed > consent and so forth are translated into an open ended call for action. I don't recall anybody saying that ethical precepts are not open to question or debate, especially in human subjects research. Institutional, Federal, and professional guidelines, in fact, are not always useful in guiding our actions. This requires that we continually hold our actions up to public scrutiny and address the problems we confront. That said, there are some boundaries that one just ought not cross, both as a matter of honor/ethics and for the more pragmatic reason that crossing the line violates professional and other codes. I've identified some of these lapses (see http://www.soci.niu.edu~jthomas). Since then, new information has emerged indicating that the CMU computer user policy, by even the most charitable reading, would prohibit accessing system users' individual private files, including Usenet newsgroup configuration files. The new information about the grant proposal strikes me as the most troubling of all the issues raised by the study. > It is true that if an M.D. gave me some toxic medicine or performed > a deforming procedure - that was experimental- without obtaining > my informed consent, I and every person in America would feel comfortable > with "srutinizing that M.D." until they were safely locked in the > Lompoc prison. Now can yo seriously tell me that, erroneous as > simm and sirbu were, that people in society, say the Christian Coalition > will give a single iota to the fact that Rimm exposed "criminal > activities" of "criminals". That is exactly what exon's messaged > (posted in another topic) said (although not in those exact words). It's not clear how the second sentence follows from the first, unless the intent is to suggest that attention to ethical lapses should be rated on an ideological scale of "public concern." If that's your point, then my answer: It's precisely because people might not be concerned about such lapses that makes it an important social issue. And, it's precisely because the study violates so many fundamental professional and other ethical precepts that makes it worth addressing publicly, lest such research be a watershed in how we treat especially socially marginal subjects. > So all in all jthomas, I find your informed consent argument compelling, > but it is open ended in terms of what further action should take > place, in what forum, in what place, the appropriate people to > be involved, and what rights, constitutional or otherwise, those > colleagues of rimm have while being investigated. Remember that the issue isn't simply "informed consent," which is a murky one at best. Some of the issues here include a pattern of deceptive, and possibly fraudulent, data gathering procedures that was used to put the subjects (BBS sysops) at risk. That, coupled with the grant proposal that would provide law enforement the means to identify and prosecute the very subjects that Rimm studied, is unacceptable. Remember: 1) Rimm's BBS population included the largest and most active in the country ("exhaustive" in Rimm's terms). 2) The proposal, according to David Banks, was framed as a means to allow the feds to target the largest and most active BBSes in the country. I can find no convincing rationale that would justify studying subjects by deceiving them and then colluding in the means of their potential harm. Had Rimm entered the study not knowing of the nature of the material he'd find, and then shocked by discovery of PORN IN CYBERSPACE acted in good conscience to report it, the issue would be murkier. This isn't what happened. If David Banks is correct, the proposal was written for monetary gain. To my mind, the only proper official forum for inquiry into the study is that set forth by CMU's own "misconduct" procedures. However, this does not preclude discussing the study and the issues it raises through discussions here, at professional conferences, or elsewhere. As more researchers "discover" cyberspace as a venue for data, the Rimm study has the ironic consequence of perhaps reducing our future serious transgressions. ###