 OK, welcome back. This panel has sort of a mysterious title. I say mysterious because it may be a question that is unanswerable, but the title of this panel is, what is the, quote, privacy interest in a public record? We all recognize the inherent tension there. And this is the point of the conference, for those of you who have attended before, where we try to bring before you some people who will, we hope, sort of raise the level of your thought a little bit. This is what we have sometimes referred to as our academic panel. And actually, this time we do have three professors. And so these are not necessarily people who are always involved in the sort of front line issues that you are mainly working on in the public access and privacy policy development area. But these are people who have done writing and teaching and thinking about these issues. And I think they have some very interesting insights for us. I'm just going to mainly function as the umpire here. I'm going to make a very brief introduction and then turn it over to each one of them for roughly 15 minutes or so. We'll pause between each of them and see if we get some dialogue going with you. And then hopefully we'll leave quite a bit of time in the second half of the program to have some dialogue among the panelists and with all of you. So I'll introduce in the order that they will be appearing. First, we have Professor Helen Nissenbaum. And just one moment here. Professor Nissenbaum is a professor of media, culture, and communication as well as computer science at New York University. She studies the, quote, social, ethical, and political implications of computing and information technologies. And so she will be giving us some insights in those areas. Then we'll hear from Professor Peter Martin, Professor and former dean at the Cornell Law School. He teaches and writes about, and I'm quoting here, the impact of technology on the functioning of law and legal institutions. It's also relevant here, I think, that Professor Martin was the co-founder of the Cornell Legal Information Institute, which for those of you who have had occasion to do research or did 15 years ago have occasion to do internet research on primary sources of law, you'll know that it was basically one of the first sources of primary legal sources on the internet. And it's still a great resource. I understand he's not as involved as he used to be, but I think it's a great resource for all of us. Peter Nguyen will be our third speaker. Professor Peter Nguyen, he is also an assistant US attorney, has been since 1997 in Washington, the state of Washington now. He teaches privacy law and cyber crime law at the University of Washington Law School. And he's written some articles on privacy law, and in particular on access to court records in the federal arena. So I'm going to turn it over to Professor Helen Isenbaum first. I'm going to just talk from down here so I can see what I'm doing. Again, I have a day job at the Department of Justice, and the views you're hearing are not the views of the Department of Justice. I have to say that before I open my mouth in public. I'm going to be repeating a lot of the things that Peter Martin has just said, as well as Helen, Isenbaum. Just a little different perspective. The perspective I want to start with, everyone's been talking about architecture. It's helpful for me, when I think about online access to court records and the problems, is to think about the days before paper, really the days when parchment was about the only way you could record things. And litigation took place in public in a very literal way. It took place in the marketplace. This is Westminster Hall. And everybody basically, the three courts were there, Chancellery, King's Bench, Common Pleas. And then the merchants are there too. Everything's taking place in public because there's no other place to handle litigation because the community is necessary to make sure that the rules are being followed and the litigants are following up on the conclusion of the litigation. It's impossible to have a trial in secret because everything is taking place in the open. Now, the focus of the medieval trial was to get the dispute resolved. And to a very large extent, that is still the focus of what courts do. They resolve disputes. Now, in the Renaissance, when they invented printing, suddenly there was a higher literacy rate. And it was possible for the first time to read legal decisions, publish legal decisions, have published statutes, think about the law in terms of general rules, and to keep records of what courts did. And it was also possible for the first time to have a secret legal proceeding. Now, these were very controversial at the time. And the Star Chamber has inherited perhaps wrongfully a bad reputation because when it was used by Charles I to go after his political enemies, I think the political enemies went after Charles I. The result of that conflict in the early modern period really led to the modern conception of law which was the idea of law representing an agreement of private individuals which came together in the context of legislatures to formulate the law on the behalf of the public. The idea was the public itself or the public sphere is the source of the law. And that government institutions, whether they're the legislature, the executive of the courts, they represent the public. The public is the principal. And the government institutions like the courts work for the public. They're the servants, the agents of the public. And from that principle, it's really critical that the public have access to government information or information about the courts to allow the public to participate in the essence of what they do and to hold their principal, hold the courts, hold the legislature, hold the executive accountable. That's reflected in the political theory and of the time. The idea is that the public has an inherent right to monitor and be involved in the judicial process. And you can see that in the architecture of the time. This is a picture of the old Bailey. And you see all around the judges and the barristers and the solicitors and the parties, you see balconies where the public can watch the legal proceedings. And of course, in the modern US architecture, we see the same ideology reflected. The pews are there to allow the public to watch and keep track of what's going on with their government. Now, the reality has always been, however, that the courts have had to balance this important value of access against the need to manage information. A lot of disputes involve private information, and Peter's gone through quite a lot of it, family law, torts, business disputes, bankruptcy, administrative, all of these civil cases involving sensitive personal information, sensitive business information. In criminal cases, it involves not only sensitive personal information and potentially business information, but also information that affects third parties, witness retaliation concerns, concerns about fair trials, concerns about protecting juries, concerns about making sure that juries aren't googling defendant's criminal records. Those kinds of concerns have always been reasons why the information has been managed within the judicial system. And again, I think I wanna emphasize that it's not simply a privacy versus publicity balance, there's also a concern at the point of getting to a fair trial. I mean, if a juror has access to a defendant's prior criminal record, if a juror can go online and retrieve a motion to suppress information that was granted, suddenly the jury is gonna know that there was a ton of coke that was seized by the, you know, seized from the person's home when a search warrant was executed without probable cause. And I don't know if the jury is necessarily going to be particularly concerned about the fairness of that proceeding to the defendant. Likewise, many of the rules of evidence that we have involves suppression of information and limitation of information to the jury. And much of that information that's removed from the province of the jury through the application of the rules of evidence becomes implicated when all court information simply goes up online. You know, of course, we were protected from much of the problem here by the system of practical obscurity. And the ability to have a system of practical obscurity permitted so much to happen while the document could still be public, okay? So that we could, you know, as I say later, we can sort of have our cake and we can eat it too. We can have our publicity and eat it too. And the problem, however, with what we face with practical obscurity is that practical obscurity also shields much of the information that needs to be public so that the public can hold decision makers and courts accountable. And Peter Martin went through, I think, very, very helpfully, pointing out that the legal standard for sealing is when you actually go through and do statistical reviews of the court files, more often than not the courts are not applying the legal standard for sealing a document which harms this public interest. At the same time, we also know that the private interests that are protected by rules that we've established saying that you're not supposed to file social security numbers in the court pleadings, those two are not being followed. And so the practical obscurity of the legal system also shields much of that information from public scrutiny as well. Also, there's a lot of detritus for one of the better word, Judge Smith, who's a magistrate, judged down in Houston, went through the clerk's office and noticed that the Department of Justice had never bothered to file motions to unseal applications for electronic surveillance. And they just, as he said, accumulated in the clerk's office like kudzu. Well, that's sort of the world we were in in the days of practical obscurity. Now, my focus has been primarily on PACER because that's the system I'm most familiar with, but I think many of the things I'll say will be hopefully applicable to the State Courts as well. You know, when we go to a system like PACER, we get a ton of benefits. I mean, it just makes the lives of the judges, the lives of the lawyers, and the lives of the clerks enormously convenient. I mean, it's hard for me to even imagine going back to the days when you had to line up at the clerk's office, you know, somebody had the file, you couldn't look at it. The ability to just go on, file my pleadings at 11 p.m. at night, you know, from home, is just an enormous benefit, and I wouldn't want to give that up. At the same time, we have the sorts of concerns that Peter Martin expressed, which is how do you deal with the secondary uses of the information by data aggregators? There's an increasing reluctance on the part of businesses to use the court system, because they're concerned about the disclosure of confidential and sensitive business information on the internet. Individuals are always exposed potentially to identity theft, public shame, embarrassment, and then we also have this additional problem about the administration of justice, the jurors are now in a potential position of circumventing many of the rules of due process that we're comfortable with and familiar with from the days of practical obscurity, that practical obscurity in many ways permitted to take place. And then of course we have, again, the sort of who's a rat.com problem where the plea agreements are being put up online which are used effectively to identify snitches, mostly in prison, although in any gang type of prosecution, it creates an enormous problem for safety, the safety of cooperating defendants and cooperating witnesses. If you have a snitch jacket in prison, it's basically a contract on your life. So we have in the case of the who's a rat.com problem, you know, we're really in a completely different world. And the world is different because the values of public access and oversight apply to a plea agreement. And yet that same information places in jeopardy the lives of cooperating defendants. This is a really difficult problem. I don't want to say that this is easy. There are three general solutions that I've often thought of. You can obviously try to get the lawyers to not file pleadings with social security numbers in them to get the lawyers to redact what they file, to get ceiling orders appropriately, follow the standard, try to avoid file, don't put the stuff in the court record if it's going to harm somebody. As we know, however, from studies, the lawyers and judges and the studies that were done on the public resource.org indicated both opinions as well as judicial filings by attorneys failed to follow the social security prohibition rule. We have a lot of bad habits that have remained from the days of practical obscurity. And we also know that agreed ceiling orders are still routinely granted in violation of the standard because if the parties agree, there's no dispute and judges are used to resolving disputes. We can come up with better legal rules and in the federal judicial conference, they certainly have tried very hard to come up with and be creative in terms of offering options. My favorite, of course, the rule of not filing social security numbers is there. They also have an additional rule in the new rule that was promulgated in December of last year of an intermediate access rule, where parties for good cause, which is a lower standard than the compelling interest standard you need to seal a record, but just for good cause or a good reason, allow the parties to file a document offline. That is, it's not private, it's not public, but they can opt into a world of practical obscurity. And that's a, you know, it hasn't really been explored very much, but I think that's a very interesting possibility. Again, we have the problem of education and the problem of the failure to follow the rules. We can come up with much better technology. The rules have, I mean, this is three ways, you know, the rules reference the technology, the technology references the rules, but the technology could be improved quite a lot. And there are different ways in which the technology can have delayed access. We talked about the 10 day delay rule here. So there's opportunities in technology to have automatic redactions. There's also things technology can do. But we also have the problem that the, I'm sorry, I'm jumping ahead, but we have the problem that the Congress has not appropriated any funds so that the courts can buy better technology. The state courts are always struggling to get funds to buy better technology. And we still have the problem of downstream secondary use of the data not being regulated. There's switching costs that people have once they get a system up and running. It's very difficult to come up with a new system. So we have these three ideas. We have better technology, better rules, and better training for lawyers and judges. But at some level, all three of those seem to be unable to really solve the problem. And what I wanted to sort of toss out for the crowd is the idea that there's a fourth component that needs to be part of the analysis. And that's what we call information economics. And the book that I read that gave me this idea is written by Shapiro and Varian. And they point out something that's obvious when you think about it, is that information is a means to communicate, but when you have it collected in large computer systems, it becomes a commodity. It becomes data. And it operates with very different economic rules from the way information works when it's used by individuals to communicate with each other. First of all, you start to have positive and negative networking effects. In other words, the benefits are magnified, but also the detriment of information disclosure and magnified. So you're working with an exponential factor instead of a sort of a normal factor when you communicate to an individual. The copy and dissemination of electronic information takes place at zero marginal cost. In other words, once you've acquired that information, it costs nothing to sell it. So no matter how much you pay for it up front, if you can get a market for somebody to sell, you can always make money. You can also recombine it with other data to create something that's even more valuable. And because of those economics, it's actually in the interest of somebody who's in the commercial data selling business to have increased acquisition costs. The more expensive it is to obtain the information in the first interest, the better it is for the downstream sellers of the information because it protects their market. It's better for them to have the switching costs. And the technology becomes extremely important to understand the technology and understand how it relates to the type of uses and misuses that are available is also important. So what you lead to is the idea of responsible information management. And this is an idea I think that Helen Nissenbaum really focuses in on with her idea of a principle of transmission, context-relative information. The idea is that you want to use the information and set up controls on the information, controls meaning better training, better rules, better technology, that you want to have in mind as a goal here the idea of increasing the positive externalities and reducing the negative externalities. You want to increase public access to core judicial records. You want to facilitate the review of aggregate data to make sure that we're in a position to, for one of a better word, call bullshit on people who aren't applying the rules. You want to facilitate academic and empirical statistical review of how judges and lawyers are performing. And you want to use that information in order to feed back into the loop of making better rules, better technology and better training. At the same time, you want to reduce negative externalities. You want to protect business, private information and law enforcement, sensitive information to the extent appropriate. You want to oversee the commercial secondary use of aggregate information. You want to make sure that the downstream use of information is appropriate with the public interest. And you want to restrict uses of information which might threaten the administration of justice, potentially organizations like who's a rat.com. To do that, you can't really do that in the context of a case-by-case decision-making process. You can't do that in the context of dispute resolutions because you're having to really look at the forest and not the trees. And judges are basically in the business of resolving disputes at the tree level, not at the forest level. You need to have an administrative approach in order to do the empirical data review to come up with more efficient enforcement structures and to come up with better rules, technology and training. Various things are possible at the administrative level that are not possible at the individual case level. At the administrative level, you can modify the Pacer site license to basically come up with sensible downstream restrictions on what appropriate uses of information are and not. You can contractually regulate commercial data aggregators downstream, data dissemination and others. You can analyze data from a systemic point of view to see across the entire universe of your data how effective are these rules to limit the filing of social security numbers or how effective are the ceiling rules being followed? And you can informally send warning letters to counsel and court to let them know that they're non-compliant. And you can come up with technological standards to facilitate better citation formats, redaction rules. You can come up with things at an administrative level that simply are not possible to approach from a case-by-case level. And I would leave you with this analogy. In the 19th century, the way pollution was dealt with was through nuisance lawsuit. An individual would file a nuisance lawsuit against another individual that was polluting. And that worked OK in the 19th century. But in the 20th century, the common law nuisance doctrine broke apart. And you needed some form of agency approach in order to address the situation where there simply were too many different kinds of both positive and negative externalities for those interests to be reflected in a single one-off lawsuit. So I want to leave you with those ideas and hopefully get some feedback from you because the reality is I've been learning way, way, way more than I could ever hope to share with you from the wealth of experience in this room. Thank you. Thanks, Peter. Well, we do have a little.