Sheryl Shenberger: Good morning, and thank you very much for being here on such a beautiful day. I’m Sheryl Shenberger. I’m the director of the National Declassification Center. And again I welcome you to our public forum. It’s our third forum in the summer. The challenges levied on government agencies by the September 2011 National Action Plan and the Open Government Initiative were greater transparency and accessibility and access to the essential documentation of our government. Executive Order 13526, Classified National Security Information signed by President Obama in December 2009, established the National Declassification Center as a presidential-mandated way ahead for government-wide declassification of that essential documentation of our government. The executive order levied certain requirements on the NDC. To streamline declassification processes, to facilitate quality assurance measures, and to implement standardized equity identification training regarding the declassification of permanent historical records. A presidential memorandum accompanying the EO provided a simultaneous challenge for the NDC. By 31 Decemeber, 2013, the NDC must process for release or properly exempt the approximately 400 million pages of previously reviewed archival federal records accessioned to NARA. What we fondly call the backlog, currently assessed at 371 million pages as of the first of January 2010. The agenda of today’s public forum has been crafted to address how our center has addressed the executive order requirements. Our speakers – some on the stage, some down front – will be addressing past weakness in the system that prompted the EO and the presidential memorandum. They’re going to address the present processes that we have implemented to deal with those weaknesses. And they’re also going to talk about methodologies that could address future government-wide declassification. But first, where are we with the backlog? There’s a fine thermometer chart in the lobby for you. And here’s the same fine thermometer chart. This is a pictoral rendering of the progress the NDC has made since standup. And I’m just going to kind of simplify it for you on a monthly basis. What this chart tells you about our progress is that we have been dealing with 5 million pages a month in 2012. Which is way higher than the just over 1 million pages a month in 2011. And it’s significantly more successful than the less than 600,000 pages a month since 2010 and earlier. Of that 5 million pages a month, we are currently releasing at least 3.5 million pages a month. And our release rate, because I’m giving you approximates now, has wavered from a low of 56% back in June, to 95% during the first two weeks of August. So if you look at the chart, you can see that at the top we started at 371.5 million pages on 1 January, 2010. But we have assessed – which means evaluated or assigned to, remanded to original agencies – 335.8 million pages. Or 90% of the backlog. We fully expect to finish assessing the backlog over the next couple of months. We have determined through all of our work that as of the 21st of August, 131 million pages need page-level review required in the Kyl-Lott Amendment associated with sensitive DOE equity. Or certification for highly unlikely to contain RD/FRD. Or documentation that the review was done in the past. We’re hopeful that we’re going to at least get to some more highly unlikely that will clear out some of that 131 million pages much faster through our sampling process. We’ve currently completed the second little red line. Quality assurance sampling on 191 million pages. Right now it’s 194 million pages. Which is 52% of the backlog. Those are the records that are off the re- and re-review merry-go-round. They don’t get seen again, they’re done. We’ve completed all declassification processes and segregation for national security information and RD/FRD on 63 million pages, or 17% of the backlog. Today’s panels will address the backlog, as well as the other elements of the executive order. The speakers are going to help us answer these questions: What have we learned since our January 1st standup? How are we moving with our mandates? I’d like to now turn the stage over to NDC Deputy Director David Mengel, who will lead our first panel focusing on equity referral, efficiency, and risk in backlog retirement. Thank you. David Mengel: Can everybody hear me okay? Good. First let me introduce the two panelists that will be discussing the quality assurance piece of this with me, the first part. To my immediate right is Robert Warrington. He’s the chief of the CIA declassification team at the National Declassification Center. And immediately to his right is Kenneth Stein, director of the Office of Document Review, Office of Classification, U.S. Department of Energy. First, before we get into the questions, I’m going to pose to these gentlemen, I’d like to give you a little background about the QA process and how we got to where we are today. Prior to the president’s Executive Order 13526, as you all know, there was the automatic declassification provision of the Clinton Order 12958. Under that order there was a lot of review that was done, and a lot of review that was done very quickly. What we saw – and this is part of our lessons learned under the old process – there were several things that didn’t go right in that initial QA review process. There was a lack of standardization of declassification guidance. Agencies protecting different types of information. We saw a lack of coordination between the agencies in equity review. One agency reviewing their records wasn’t necessarily briefed on the types of information that belonged to another agency. So they weren’t always protecting it the way it should have been. And there was a lot of ad hoc decision-making that was being done at the time. So the old QA process led to a series of re-reviews. One agency would review, another agency would review for their equities, and so on. It got into this cycle where nothing was getting through the QA process. Or very little was getting through the QA process. This would not work given the new mandate by the president. As a result, as the Archives started to stand up the NDC, we put together an inter-agency panel that went through a Lean Six Sigma process to look at the entire declassification process and determine what we could do to improve the QA piece. Now any improvement at that point would have been good. But we worked very hard. It was an inter-agency team. There were a lot of people that had their own interests in the room, that they were willing to set aside and work together to come up with a process that we all agreed on. The general agreement – I won’t go into the exact details of the process, but what we’re looking at is identifying those records that are highest risk for containing the sensitive two equities that the president mentioned in his memo: The human intelligence source, the confidential human source. And the key design concepts of the weapons of mass destruction. Those were the top priorities. So records that did not contain – were unlikely to contain these – we could put them through a very quick process that allowed us to assess them very quickly and move them out into the release portions of the process. And allow us to focus most of our efforts on those records that truly contain these sensitive equities that the president mentioned. So the group came up with a process that we all agreed on. We’ve seen big improvements. You can see from the chart that Sheryl showed, that the QA process – we’re more than halfway through. Which is huge leaps from where we were two years ago when we started this. As part of all of this, the agencies, as I said, were very cooperative. We were asked to do this jointly. We did. There were certain pieces that we had to take into account. One of those was risk. The other was working together to develop this QA process. So I’ve got two panelists here, as I mentioned. And we’re going to talk a little bit about their roles in this. As they are the equities that have a large interest in those two specific categories, we thought that they would be the most appropriate to have up here with us today to talk about some of the quality assurance pieces of this. So I’m going to open with the first question here. According to the president’s memo, further referrals of the 400 million page backlog are not required except for those containing information that would clearly and demonstrably reveal the identity of a confidential human source or a human intelligence source, or key design concepts of weapons of mass destruction. So obviously each of your agencies has a key interest in these types of records. Bob, can you tell us a little bit about your approach to identifying information that falls into the human intelligence source / confidential human source category? Robert Warrington: Executive Order 13526 defines the objectives and focuses the efforts of the declassification community. Two basic truths underlie the executive order, and they are expressed in its opening paragraph. That democratic principles require the American people be informed of the activities of their government, and that the protection of democratic institutions requires that certain government information be maintained in confidence. Promoting Open Government and safeguarding sensitive information. Those are complimentary, not competing goals that the declassification community must achieve. Within this framework there are good and just reasons why information revealing the identities of confidential human sources and human intelligence sources must be referred for declassification review, and is exempt from automatic declassification at 50 years. Human intelligence is an essential and irreplaceable source of information for CIA to complete its mission. Note the two adjectives I used: essential and irreplaceable. Advances in technical intelligence collection have been impressive over the years, but there are kinds of information that can only be obtained through human sources. An aerial photograph will tell you how many missiles are on a missile pad, but that overhead photography won’t tell you what are the opponent’s plans for using those missiles? What is the opponent thinking? What are his goals? Uncertainties? What are his perceptions of us? Human sources provide insights into the intangibles of human behavior and intentions. And there are no substitutes for obtaining that information other than through human sources. There are a number of reasons why the identities of confidential human sources and human intelligence sources need to remain protected. Their exposure would end the sources’ value as sources and endanger their lives as well as the lives of their family. They would discourage prospective sources from coming forward to offer their services to us. They would reveal methods the CIA uses to contact and interact with human sources. And they would undermine the value of information the CIA obtains covertly. The advantages of knowing secrets of your opponent are largely diminished when your opponent finds out you know them. And finally – this is very important to us – it violates the pledge that we make to human sources to protect them to the greatest extent possible. And that oath we make to the sources depends on keeping their identities secret. Thanks. David Mengel: So Ken, the second piece of this is the key design concepts of weapons of mass destruction. Can you talk a little bit about your focus on the information and how this review relates to the Kyl-Lott Amendment of the Defense Authorization Act? Kenneth Stein: Everyone in this room, I’m sure, does not want the nuclear weapon design information that we have be proliferated on the internet, through the world. We are concerned about irresponsible nations obtaining nuclear weapons. We are concerned about subnationals obtaining nuclear weapons. Yes, the nuclear weapon program is now approximately 70 years old. But it is just as dangerous as it was 70 years ago. The Department of Energy fully supports Executive Order 12958 and its successor 13526 in having transparency that is holding the government responsible to its public. But it has to be done in a responsible fashion. Unfortunately, if you look back at the way classification has occurred – classified documents have been protected in the past – we have a problem in that, in the past when documents were generated by the federal government, they were not properly marked in many cases for their content. Particularly nuclear weapon design information content. We also have the problem that when the records were archived and stored away they were not separated out according to whether or not they had nuclear weapon content or not. And now we come to the executive order, which in our view was a very good idea. ’95. To make all these documents available to the public. And there was a great deal of pressure put on the agencies to produce – to get these documents out there. And because of that pressure, there wasn’t the proper precautions taken to make sure that all this nuclear weapon design information was identified and taken out of the distribution of these documents. So we had for about 3-5 years a process going on in which certain collections were “vault declassified” without review. In other words, roughly sampled without looking for our information. We also had the problem of individuals doing the reviews not knowing what they were looking for. So what happened was, for the first 200 million pages or so that were processed by the Archives and made available to the public – there was a significant amount of design information in there. And all of this information was eventually identified by the Department of Energy and pulled back. Yes? Someone said something? Alright. It was pulled back. And we identified that information in these reports to Congress. We are required by public law to report to Congress and the National Security Advisor for the president all the instances that did occur where design information was inadvertently released. Congress, in its oversight of the executive order, realized in 1998 that certain precautions or certain safeguard measures had to be put into place to make sure that this information didn’t get out. This is the Call Amendment to the fiscal year ’99 National Defense Authorization Act. That’s reflected in Public Law 105261 Section 3161: Measures to prevent the inadvertent release of RD and FRD. That particular public law requires that a page-by-page review be conducted of the records with individuals trained by the Department of Energy to recognize the RD and FRD content. And that applies to all documents from that day forward that were reviewed. In the subsequent year, ’99, in the 2000 National Defense Authorization Act, Public Law 10565 Section 3149 says that not only do those requirements for page-by-page review apply to records from this point forward, but also applies to those records that were reviewed from ’95 until the enactment of Kyl. That is Kyl-Lott. We are required under public law to review these documents. I understand there’s some objections to that. I respect those objections, but you have to understand that the reason we do it is to make sure we don’t proliferate the information. Now going back to David’s question here. We made significant changes in our processing at the Archives. We realized that even though we had trained 3,000 individuals in the federal government to recognize RD and FRD, and we had these page-by-page reviews being conducted, they weren’t of a proper quality. There were still items being missed. So the Department of Energy still is sampling, on a quality control basis, these documents to make sure we don’t release any RD and FRD. At the NDC over the last two and half years, we have cleared about 130 million pages of the backlog. But to clear those 130 million pages, we identified approximately 3,000 documents containing RD or FRD. Of those 3,000, 20% of them were actually marked with RD/FRD content. But for some reason they were not identified by the agencies when they did the processing. I think that answers your question. David Mengel: Okay, we can go on to the second question then. The memo also directs agencies to promote efficient and effective utilization of finite resources and strike a critical balance between openness and secrecy. Each of these statements implies agencies must complete their work with existing resources and accept a certain level of risk. In other words, you will not be able to review everything for missed equities. As I mentioned in the opening, the new NDC processes are built on accepting a certain level of risk, yet risk does not come easily to your agencies. Can you each discuss your agency’s approach to risk? And I know you both have changed your views on this as the NDC was established. And how you’ve adapted to the president’s direction? And a little bit about the culture shift this may have been for your agency’s declassification programs. Robert Warrington: It was immediately apparent to CIA that eliminating the 400 million page backlog by the end of year 2013 deadline would involve managing risk. We didn’t have sufficient time or personnel to look through every page. Nor was such an approach necessary. What it did mean for us was that we had to expand the scope and accelerate the pace of records review far beyond historical precedents. The task is complicated by the fact that the backlog contains large numbers of documents that have not been accurately reviewed for equity referrals by originating agencies. Something my colleague Ken just mentioned. To achieve the executive order’s goals while contending with sizable amounts of missed equity, CIA invested significant sums to increase its presence at the National Declassification Center. More than doubling the number of reviewers on its declassification team at NDC. We established the Records Equity Assessment Program or REAP to organize and concentrate our efforts. REAP’s function is to review substantial quantities of backlogged records and perform a quality assurance evaluation for missed CIA equity. REAP, moreover, is where the philosophy of managing risk by targeting resources comes together. It is obvious by the nature of the content of certain kinds of records that they are more likely to contain CIA equity than others. Records of the U.S. Army’s Assistant Chief of Staff for Intelligence, for example, are of much greater interest to us than the records of the U.S. Army’s Quartermaster Corps. A correlation exists, not surprisingly, between the volume of CIA equity in records and the frequency of missed equity referrals to CIA for declassification review. Concentrating our efforts on the records with the highest risk of missed CIA equity necessitates less attention being paid to other, still risky records. Managing risk is critical in our strategy, because the probability of missed CIA equity being present in records is more or less, not yes or no. Given this we base REAP’s workload on a comparative assessment of the likelihood of different record collections containing missed CIA equity. We exercise risk management at REAP, therefore, by interpreting and balancing across records the relative risk of inadvertent equity exposure. REAP has proven its worth as a filter to protect CIA equity in the records of other government agencies. Since it began operations in 2011, REAP has discovered thousands of documents containing CIA equity that were not referred to CIA by the originating agencies. In closing, I wouldn’t say that all these changes have produced a cultural shift at CIA. What they have done is allowed us to evolve our culture in ways that would not have been possible prior to Executive Order 13526. Kenneth Stein: The Department of Energy did a two-part process to deal with the NDC backlog. The first thing is to increase the resources. Prior to the president’s December 29th memo, 2009, we were put on notice that the NDC was going to be stood up to handle this backlog. We had a workforce of only seven reviewers down at the NDC. We ramped up to 35 reviewers. A factor of five in increasing our staff. The other thing is we looked at the way we’re reviewing the collections. And we decided to take a very aggressive risk management approach. The sampling is targeting certain boxes within collections which we believe would have our equity in it. And by doing so we have made a significant increase in the amount of throughput in our system. I don’t call this initially a culture change, I call it more of an engineering decision as to how to handle a problem in getting the most identified with the least amount of resources. With 35 individuals, as I have said, we have already cleared 135 million pages worth of materials. And we’ll continue to push hard to try to achieve as much as possible by December 31st of next year. David Mengel: As you both mentioned, you increased your staff at the National Declassification Center. And a lot of this is in order to protect the equities you’re most concerned about. But there’s also a piece to this that is the cooperation that we started with establishing the NDC and the NDC processes. Can you talk a little bit about how you’re working within the agency, within the NDC, and with other agencies to improve that cooperation across the community? Robert Warrington: A lesson the executive order has highlighted for the declassification community is that eliminating the records backlog by the executive order’s deadline and doing so in a responsible manner means that we’re all in this together. No agency can go it alone in performing its role while protecting its interests. This was recognized in the creation of evaluation teams and Kyl evaluation teams – inter-agency bodies formed by the National Declassification Center to reduce the backlog. Representatives from CIA and other agencies work together on Kyl evaluation teams, for example, to complete Kyl-Lott certification of the records. They look primarily for key design concepts for weapons of mass destruction while also watching for information that reveals the identities of confidential human sources and human intelligence sources. Because CIA representatives are present on the team, we can make instantaneous decisions on whether records with CIA equity must be referred for declassification review. A lot of them don’t. This expedites the overall evaluation process and avoids unnecessary referrals. One might wonder why CIA is on a team searching other government agency records for key design concepts of weapons of mass destruction. Isn’t that DOE’s problem? No. It’s the declassification community’s problem. We have a shared duty in reviewing records for release to safeguard still sensitive intelligence. When working on evaluation teams and Kyl evaluation teams, CIA relies on other government agency reviewers to identify and refer its equities. Just as those agencies rely on CIA reviewers to identify and refer their equity. As I said previously, no agency can go it alone. We are all in this together. This reality is evident in the important improvements in inter-agency cooperation that have occurred under the NDC. These are initiatives that CIA supports with its resources and reflects in its operations. To summarize my message today, I want to leave you with these thoughts: CIA is dedicated to eliminating the records backlog by end of year 2013, the deadline set by the executive order. CIA is pledged to do so in a manner that achieves the two goals embodied in the order. That democratic principles require that the American people be informed of the activities of their government. And that the protection of democratic institutions requires that certain government information be maintained in confidence. CIA, as evidence of its commitment, has increased significantly the allocation of resources to its declassification team at the National Declassification Center. Expanding its productivity and output far beyond historical levels. Lastly, CIA will not regard its mission as successful until the backlog is gone and the American people have access to all of the records to which they are entitled. Kenneth Stein: It’s difficult to follow up on that. David Mengel: It is difficult to follow up on that. Kenneth Stein: Okay. We’re in this together. And we realized that at the beginning of 2010 when we sat down with David in multiple meetings trying to figure out how to make this all work. It required compromises in a way – compromise and learning from each other. Just figuring out how we can jointly do things together rather than sequentially do things. And we found that if we just step back for a second and look at what was going on – just by making some very small tweaks in the way we’re handling ourselves – that we could work together as an evaluation team in the stacks to get through a great deal of material without having to bring all of the boxes into the rooms. That was extremely important. That is the most efficient way of handling the problem. And we realized that. And it works. We’re very happy, it works very well. Also we made sure that we recognize each other’s equities. Sort of watching out for each other. I was very impressed with the training that we had last fall at the NDC, where each of the agencies made a presentation of their particular equities. I learned a lot about the CIA’s equities, and I’m sure they learned a lot about ours. In this way we are actually watching out for each other, making sure we get everything covered. And it works. It works extremely well. As far as our commitment to the program – I’ve already said and Bob has already said, we’re here to get the job done. We will get the job done. David Mengel: Thank you. We’re about out of time here for questions, but just to sum up. Both Ken and Bob mentioned this: there is a lot of inter-agency cooperation going into this QA process. Ken and Bob were invited here to speak today, but we have many other agency partners that are participating in the assessment piece and participating in the evaluation piece, which is actually going into the stacks and doing sampling review of records for the two categories that we’ve mentioned. So this is not just a DOE and CIA and NDC initiative. This is a CIA, DOE, State Department, Department of Defense – all the component agencies. All those people that have an interest in these records. And there actually has been a lot of cooperation across the community to get together and reduce this backlog, and get it done by the time the president has asked us to. Sheryl Shenberger: I’m the break between so you know when one ends and the other one starts. Thank you so much. When Bob talks we want to send eagles up every so often. It’s quite patriotic. But both gentlemen really did sum up the cooperation. I am very fond of saying that we are successful together or we fail together. You know? We sink or swim together. And that’s exactly what they said. And I appreciate that. Right now we want to turn the stage over to Don McIlwain. He is the NDC FOIA MDR Division Director, and he’s going to lead a discussion on streamlining processes and avoiding future backlogs. Don McIlwain: Thank you, Sheryl. Good morning. With me this morning to talk a little bit about what we’re doing to streamline processes and to work so that on January 1st, 2014 we don’t open the doors to another monstrous backlog. Doug Richards, on my immediate left, who is the Declassification Chief for the Joint Staff. And reprising the role of Ernie Banks, today, is Ken Stein with the Department of Energy. So what we’re going to do is talk a little bit today about measures that our agencies are taking, in some ways in a laboratory sense, to make sure that declassification processes are streamlined. Not just at the systematic review level, but we’ll look a little bit at what our agencies are doing to prevent backlogs of records so that the records, when they hit the door of the Archives, are reviewed correctly. We’ll also talk a little bit about what happens to those records that may have been exempted as part of the systematic review process. What happens when you want to make a FOIA request for those records? Or a mandatory declassification review request. Because believe me, the work that Bob and Ken and David and all of the other agencies are doing at the systematic review level means you will be able to make a much better informed, much more concise FOIA request down the road. Now I wanted to talk a little bit about why that’s the case. When we had an unprocessed, undescribed, almost unknown backlog of about 400 million pages, and you knew that ten boxes had been transferred to the Archives but no one had any idea what was in those records, and they may or may not have been reviewed. Equities may or may not have been identified, you were somewhat stuck. You could maybe make a FOIA request for all ten boxes in that particular accession or transfer. The work that my colleagues are doing on the systematic side – the assessment process, the evaluation process, the quality assurance process – means that coming out the back end as we reduce the backlog – and again I would look at that thermometer chart – instead of having to make a request for ten unprocessed boxes, you can now go look in those boxes and see where a large part of that information has been declassified. Has all the archival processing been complete? Are they pristine? Probably not yet. But you can go in and look at those, and if you wish to file a FOIA or a mandatory you’re no longer requesting ten boxes that you know nothing about, necessarily, other than a title an agency may have assigned those records. But you’re able to look at specific withdrawn items notices and file a request on those. That means your requests will get processed much more quickly, because my office has no ability to unilaterally declassify classified information. What we need to do with every request is send it out for consultation. We went through out own Lean Six process improvement study about two years ago. We made a number of changes to that process, and as a result we have doubled the amount of FOIA and mandatory cases we’ve completed between 2010 and this year, 2012. Things that we’re doing: Contacting researchers as soon as we receive the request if the request is broad. To use what tools we have to help them narrow their request. Instead of making paper dolls and xeroxing piles of documents to send to three or four different agencies, scanning the document into our system once and sending that electronically to the agency so they can provide us their determinations. That actually does a couple things. It takes a lot less time and it’s a quality assurance measure, because when we scan the documents in, the computer counts the pages, numbers the documents, and eliminates error and human mistakes. So that’s worked. We’re beginning to see the results as agencies return their decisions to us. So from that perspective, on what I would call the small boutique part of declassification – that of FOIA and MDR – I refer to what David and Ken talked about as the CostCo operation because they’re dealing literally with almost warehouses of boxes. And they’re working systematically through them to get them through the process. But on the FOIA side, we’re working too. And what Doug and Ken will talk a little bit about is some innovations their agencies are doing that can be a laboratory for future and continuous improvement. Before I get to them I just briefly want to talk about – and I need to put something up on the screen. If you will bear with me. An innovation that we have recently done with our records management folks – declassification does not occur in a vacuum. And one of the major things we see is making sure that that doesn’t happen again – is working with agency records management folks. And I briefed the bridge in June and in August on ways to get agency declassification managers talking with agency records managers before those records are even transferred to the National Archives. A couple of weeks ago, what you see up on the screen was approved as a National Archives form. It is currently optional for new accessions coming into the Archives, but we are working to update our regulations in the code of federal regulations. And this will be a required form, hopefully beginning in January. It depends on the Federal Register process. But what this does is it opens the door for that conversation to take place before the records even come to the National Archives. And I’m just going to go over it very briefly. We’re asking that when agencies get their records ready to transfer that they answer a couple of questions about those records. First, have the records been reviewed for declassification in accordance with the executive order and its implementing directive? And when we built this form we conveniently included hyperlinks. So that if the records officer scratches their head – maybe they don’t do a lot with classified records – they can go and see what those requirements are. The next one – and I know Ken in particular probably appreciates this – Have these records been reviewed under the rules and the law of Kyl-Watt? There is a special historical records review plan that describes in great detail what needs to be done. We’ve included a link to that as well. Also, the implementing directive for the executive order says referrals need to be tabbed in accordance with 36-CFR. There is a standard form that agencies use to tab records. That works great in paper. We’ve begun exploring how we do that with electronic records, but we’re still getting a lot of paper into the National Archives and probably will for a few more years. So we asked the agencies, “Have you talked to your classifications folks? Are the records coming in in a way that we can put them through the NDC process without having to reinvent the wheel?” So this is something that I’m pretty excited about. I think records management in agencies and the Archivist and the Acting Director of the Office of Management and Budget recently put out a new Records Management Directive that really puts muscle in the records management program for the federal government. I see this as an opportunity for the declassification community to work with an invigorated records management community to get the permanently valuable classified records into the National Archives to reduce the possibility of us going back to the days of the backlog. So with that I’m going to turn it over to Ken, who’s going to talk a little bit about some of the things the Department of Energy is doing to improve declassification. I’ll go sit back down. Kenneth Stein: We handle a large amount of FOIAs and mandatory review as it comes over from the Archives each year. And Don’s initiated a program where he would send them over – instead of paper form, he would send them over in CD or DVD form. And we have set up our system so that we have electronic processing of the records as far as we’re concerned. It used to be we’d get a document in, it would go into a folder, that folder moved from desk to desk. Instead now things come in on the CDs, they get uploaded into our classified LAN system. And the documents are reviewed online by our reviewers. We use off-the-shelf software such as SharePoint for storage of the documents and Adobe Acrobat for the actual review of the documents. And again as I said it just goes from desk to desk. Instead of paper form, it’s all electronic. They get onto the LAN system and do the processing. The output going back to Don, unfortunately, right now is still in paper form in the sense that we print out an excerpt of pages that have to be bracketed. And that’s going to change shortly, and we’ll actually send back CDs to Don. What we would like to do in the near future is to cross-connect our LAN systems so we would not have any discs. We would do it all through the internet. Classified LAN cross-connects. We’d like to do that. Hopefully it will happen soon. There are some protocols that have to be worked out for that to happen. I’m sure some time in the near future it will happen. Now as far as the actual processing of documents, we have come to the realization that it’s extremely costly to do bracketing review manually. And we’re also concerned about the future inundation that will take place with the e-mail that is being generated by the federal government. We are working on IT systems to actually screen documents electronically. I have a little work station set up in my office where we have a little platform. Whatever you want to call it. Where we’re processing Department of State telegrams that were generated in 1978-1979. For instance, in 1978 there were 500,000 documents. We set up a nice electronic filter, loaded it up with a word search of about 1,000 words, and from that we were able to call out about 30,000 out of 500,000 to be looked at by reviewers. We looked at that again and further figured that if we cut it down to maybe 1,000 documents it would be nice. We did that through modifying the word search list. And we’re getting to the point now where we think we can start to use the system for quality control checking of documents that have been reviewed by other agencies as well as the Department of Energy. Eventually we will get to a point, as we learn more and further refine systems, that we can actually start to have a hybrid between man and machine in review of documents. Eventually, possibly, perhaps before I retire in about 8 years or so, we will come to a point where we actually can take man out of the loop. And actually have pure electronic processing of documents. That’s it. Don McIlwain: Thanks, Ken. Doug I know you have some PowerPoint slides. Go ahead. You should already be on the desktop, if not just give me a holler. Douglas Richards: Thank you very much for having me. Good morning. I’m Doug Richards and I’m the Branch Chief of the Joint Staff Declassification Branch. I hope to keep you awake for the next ten minutes. This is not a truly exciting topic, streamlined declassification, but I’ll do my best. We have been fully automated now – digital – for the last 15 years in terms of our record management and our FOIA and MDR processing. We have not digitized all of our auto-declassification projects yet, but I’ll get to that in a minute. But for declassification process for FOIA, MDR, security reviews, Congressional reviews, special reviews, and foreign relations of the United States – all six of those reviews are fully digital. And we’ve been that way for a long time. So my goal today is just to give you an overview of what tools we use and the prediction of the way forward. Before I do that, though – this success that we’ve experienced with our IT tools over the last four years primarily is due to being a centralized declassification center. We’re in the Pentagon. We’re physically co-located. We’re within 20 feet of all of our AOs and legal councils across the hall. We have – we’re linked to Records Management. They’re the other branch in the division. And we have a sort of addiction to transparency and accountability. And change. We’re always willing to change. Those three things are crucial for what we do, and I will show you some metrics that show that. The tools that we use – there are two that we rely on extensively. One is JSAP, Joint Staff Action Processing. This is our primary tool for the entire Joint Staff for tasking, tracking, developing policy. It is the tool that we’ve used since 1996. About four years ago we integrated all of our declassification processes into JSAP, including MDR. So when we receive a task from OSD, we digitize the request document and all the documentation and incorporate that into JSAP. We get rid of the hard copy documents. We’re about 90% digital for secret and below. JSAP is archived, and it will be accessible for future generations. They can do searches, they can find the decision that we made on a FOIA 20 years ago. It’s accessible to everyone on the Joint Staff. I have a couple slides showing the JSAP. This is the main tasker. It shows the subject line, the suspense, the classification. This is what the AO receives as a FOIA or MDR or security review. It requires action, you cannot put JSAP on ignore. If you receive one you must review it. And if you need an extension, it has to be granted. The days of hand-carrying documents to the Pentagon for the Joint Staff and providing them to an AO for a FOIA review are over. Everything is electronic. Next slide. This shows that we can incorporate anything we want into a JSAP. This happens to be a special review we did as an MDR for a retired admiral. We can include justification documents, such as when William Burr provides justification documents on a declassification decision. We will incorporate those into the JSAP. We have notes. We have the executive order. We have FOIA Express guidance on how to make redactions. Everything the AO needs is included in the JSAP. It is a standalone action. They do not need to go to another site. They do not need to even call us. Everything is included. The AO receives an e-mail link to the package. You click there and it opens up, and all the guidance is available. It’s not a phone call. It’s not another e-mail. It’s a system-generated notification. For records management we rely on Documentum on our classified site. This is our electronic records management repository. We do not have any C-drives. We don’t have any local drives anymore in the Joint Staff. We have an enterprise level records management structure. So documents from cradle to grave are managed in this system. And our approved policy is incorporated into the different folders on the left, where declassification 0901 is incorporated in terms of FOIA, MDRs, and SRs. So all the documents on the JSAPs are integrated into Documentum. And for declassification, this provides a benefit in the sense that we can conduct an MDR or FOIA search in five minutes. The corporate archive of the Joint Staff is available for immediate access by anyone on the staff. This was deployed in 2009. We have other variations that go back to he mid-90’s. So we have hard-copy documents ready for review up until I think about 1995, and then we have all the rest of our documents electronically. Either on Documentum or on our JWICS. We also rely on a classified chat room. This is just a screenshot of an example. I created this one just to have one. An AO on the Joint Staff – if she has a question about a FOIA or an MDR, and in the future an auto-declassification, they do not call. They don’t stop by. They send us a chat directly. We can then provide them links to other classified documents, or better instructions on what we’re asking them to do. In terms of transparency, we rely on Intellipedia. We have two sites, one for automatic declassification and co-com support. And one that is available to personnel at OSD. So if an AO in the FOIA office over at OSD has a question about a particular action, they don’t call us. They go to this website. And they know exactly where it is. What’s taking it so long? Or who is the AO and how do we reach that person? Our metrics show the improvement. Using these tools in a more robust SOP that we deployed. Our FOIA processing for example. In 2010 it took us 182 days to close a typical FOIA action. That is an average, so it could be a 5,000 page history, it could be a one page e-mail. On average 182 days to close the FOIA. In 2011 we dropped it to 56. And closed all those actions within the first two months of 2012. And so far for 2012 we’re closing our FOIAs on average in 38 days. We’ve essentially killed our backlog. I’ll show you that in a minute. This mirrors our MDR processing, the magnitude. The volume of MDRs have increased, but we’ve managed to keep it below 75. Our goal is to close all actions within 90 days of receipt, regardless of the complexity or the volume, for both MDRs and FOIAs. The backlog has improved as well. These are columns that reveal the age of our backlog. So for the end of July 2012, we had a total of 32 open FOIAs down from a high of 400 back in 2006. 32 open FOIAs. Of those only 2 were over 90 days old. And I predict that we’ll never have more than a handful of actions over 90 days. And these are very complex. Typically they’re about detaining affairs and targeting. They’re not easy FOIAs. Going back just to April 2011, almost half of our backlog was over a year old. So deploying JSAP, Documentum, the chat room, and having a much more transparent operation, we have, I think, reached the goal of responsiveness and accuracy that we were looking for. So for auto-declassification, how are we going to use these tools and this dedication that we have to incorporate large collections? We have two opportunities. One is to rely on Documentum to produce these records as they matured with certain age. AOs tag documents that go into Documentum for a declassification date. So when you create a document on the Joint Staff, you automatically have to metatag it for the general declassification date. These documents in ten years will come out of Documentum as e-mail notifications to my branch saying that Thursday there are ten documents that have matured to ten years old. We will review them. That will first occur, I think, in six years. Or we could – and we are planning to do so – scan a special collection of Joint Staff records and ingest them into Documentum and into JSAP. We’ve contracted the FBI to scan a collection. It is ready to go. And I expect to do this in 2013. We’re going to take a very high level special collection of significance, put it into Documentum, create probably 10,000 JSAPs, and send it out to the staff with the suspenses that I just mentioned. Those are the tools we use. If you have any questions, concerns, or doubts please let me know. I’m more than willing to get into greater detail. But we have grown this program into 16% of all Joint Staff action processing. The Declassification Branch is now 16% of all JSAPs produced by the Joint Staff. And I think that’s rather impressive. Thank you. Don McIlwain: There will be time for questions and answers afterwards. I just want to say it was interesting to hear both Ken and Doug talk about things that they’re doing within their agency that will lead to a brighter future, I believe, as we move forward. And folks, we want the backlog to go down. But we are aware that the amount of records being created and the formats in which those records are being created are growing and changing. So those are just a couple of examples of some forward thinking that will hopefully mean the future will be bright. And again, Ken and Doug, thank you very much. And Sheryl, on to the next presentation. Sheryl Shenberger: Okay so we’re limited in time, naturally. We could have gone on and on for both of those prongs of the executive order. But since we are limited, I’m going to take you now to the third prong of the executive order, which is talking about equity training. I’m going to turn this over to Matthew Sherman. He is our equity training coordinator extraordinaire. And he’s going to talk to you about how we are creating a curriculum. Mathew Sherman: I’m going to just switch gears here with the slides if I can. Here we go. First hurdle. Good morning. As mentioned, I’d like to give you a quick once around the world of what training does look like at the NDC. And a couple of things that you’ll hear that will be a constant thread will be that of standardization and cooperation. I’d like to first start by saying that training is not new at NARA or the NDC. It’s something that’s always taken place in terms of reviewing for declassification, as a necessity. No one is going to an academic institution and coming out with a degree in how to review and declassify information. You can go and get skills on managing it. Records and what not. But to really become good at this, you need to work in that field. So it’s always been there. And as an instructional designer – you can think of me somewhat as a curriculum developer. My piece here is to come in here – again following the executive order to take that information, the training that was available, and looking at ways to improve it. And not only to improve it, but again the key as we’ll see is standardizing it. Before I get too far ahead of myself… This is just the – you can see the language that’s in the order. And Sheryl mentioned a large part of the mission and the challenge to the NDC is setting up and standardizing that training. Before I go on too much further, I’d like to just mention what’s available right now. We have an equity identification course. There’s courses on managing NATO information, nuclear information, preservation issues, the responsibilities and liabilities when you get a security clearance. We have an NDC orientation. We are about to pilot a processing orientation class, which is kind of a – it’s how we work inside a life cycle. And as was mentioned earlier, we sponsor periodic conferences to which we invite the entire community. The declassification community. And as my colleagues have mentioned, we share information. And hopefully I’ll walk away a little bit smarter on how to identify and refer each other’s equity. The NDC’s training mission, as you can see: Establish, implement and maintain training. Looking forward, the big piece of that is going to be the standardized training. And the standardized training is going to take the form of a certificate program. I like to make the analogy of something of a vocational trade license. Something like a plumber, or electrician even. Where if I’m certified, a future employer has some expectations of what kind of basic skills I bring to the agency. Because turnover and moving around within agencies is not uncommon in this field. Which takes us to the goals. First we need to establish a foundation of common knowledge for all reviewers. I think, as you’re hearing, that’s a pretty consistent thing. We all need to be on the same page. We need guidance to ensure that the information of interest, the equity, is treated consistently by all. The guidance on what types of information not to be referred is also just as important as what needs to be referred if we want to avoid future backlogs. Improving understanding of the exemptions and the categories. Particularly in the spirit of releasing all we can and protecting what we must, which is the NDC’s motto. The certificate program will provide the evidence for a standard level of training in competencies and proficiencies. Number six is creating training that will serve as the foundation for more advanced and agency-specific training. Again, every agency is a little bit different. They’ve done things a little bit differently. And there’s some things that they’re going to continue to do individually a little bit differently. But there is an awful lot of it that is – you can establish a common ground. And that’s what we’re trying to do with the core certificate program. So what does it look like? First of all, when we went to outline this – again my task in here as an instructional designer is how to standardize it, how to improve it. And how to maintain it and continually improve it. So also, as you heard about, the level of cooperation is truly immense. Because if we want to present a standardized way of doing something – and it’s going to go on paper. It’s going to be in print, on slides, in instructor books. There will be permanent students using resources in the future. It has to reflect the community as a whole. So in the higher level development, in outlining processes, the entire declassification community was involved. And we did receive a lot of good input from all of them that helped shape this. So the curriculum as it stands consists of six core modules broken up into four that are web based – the WBT, and two that are instructor-led. The four web based courses will be accessed through NARA’s Learning Management System. And again they’re laying the foundational knowledge. Each of these modules is considered a prerequisite to the instructor-led training, which you would come on-site and take at a later date. The instructor-led piece is going to be probably about eight days of training. Whereas each of the modules in the web based is probably about an hour’s worth of training. So the instructor-led will be very hands on. There will be a lot of reinforcement of the WBT, but more importantly we’re trying to emulate the real world. So learners will get a stack of papers or documents, and they will as much as possible review them just as they would on the job. An example of the web based training, again this is just an example. But I will list a few of them for you. The first one is just an overview. It introduces the purpose, what we’re trying to do. And it gives information on how to navigate. The second one really gets into the content. And the objective here of identifying how classified federal records progress through the records life cycle. That’s a loaded statement. Very complex. So that’s how this initiates. And it moves into declassifying federal records. Learning objectives, identifying the agency’s requirement. You’ve heard about the different types of declassification. So this module is going through the processes of each of those. And then lastly, and perhaps of particular interest here, is public access to those records. Dealing largely with the FOIAs and the MDRs. How those are processed, what the exemptions are like, and what kind of impacts those have. So as a learner I’ve gone through all these modules. It goes into NARA’s Learning Management System. I can verify that I’ve taken it. I’ve passed all the tests for each one of them, which is required. Then I’m able to register for the instructor-led program. And that looks like this. Again just an example. A cover of what the slides in the student guide might look like. The instructor-led again is very application based. It’s going to reinforce the web based. The instruction and discussion will be expert led by folks like you see on our panel. We are fortunate to be able to draw from the entire community, from experts who have been doing this for decades. And it really takes that to get very, very good at it. So what we envision is the instructor pool, if you will. So it’s by no means a NARA or an NDC training event. But it’s truly going to be a collective effort. And there will be speakers from all the key players and agencies. Topics included in the instructor-led include our purpose, that is of the NDC, classification authority, original and derivative classification, you can read through these. A lot of the common topics are the ones you’ve been hearing all along this morning. The last course in this is the equity and exemptions – the EO 13526. This one’s looking more specifically at the nine exemptions themselves and getting into those details. So again, all throughout these two instructor-led pieces will be interwoven the knowledge components from the web based training, but in a very hands on environment. So they’ll be reinforced all along, but we’re physically reviewing documents with experts in the room and then we’re discussing Why did we do this? How did this happen? In trying to establish a common perspective, in a way of viewing how to refer or not to refer. In terms of implementation, the courses I had mentioned – including the equity identification course, which is a very key component in the type of training that we offer, the web based pilot that I’m discussing is scheduled to be piloted this November. And the instructor-led component will probably be early into the following year. In addition, all along we will be assessing and evaluating the effectiveness of the program. If we don’t, we don’t know if it’s working. We don’t know how to fix or tweak something. So the two areas that I’m going to focus on – I just want to talk a little bit of jargon for a second. In the instructional design world we have a formal evaluation system. We call it the Kirkpatrick System. Level two is aimed at measuring learning that takes place in the classroom. So basically you come into the classroom and there’s a test. That test measures what learning took place. Which is great, because that tells us about the instructional content and if the instructors are doing what they’re supposed to be doing. What’s more important is what we call level three, and that is the graduate goes back to the workplace. Spends some time – weeks, months, whatever it may be – and we follow up. So I can do a survey, or reach out and make a telephone call and talk not only to the graduate but their supervisor. And get a better sense of how their performance has changed based on the training. So if we do this – and this is a regular piece that we’ll be continuing, it’s part of the maintenance, that’s how we will tweak and improve and modify our system and our training to continually meet the needs. And once the core curriculum has been established, there will be more opportunities for working individually with agencies. Again, we don’t expect the core curriculum to cover all bases for all agencies. There will be some specific needs that are idiosyncratic to a given agency. And so if they came to us, or collectively we discover that there is a deficiency or gap in analysis, and we can help identify an area that we can help assist, we will certainly be there to do that. Again, as Bob said – I won’t top it either. The level of cooperation truly – from the actual work, the reviewing, the referral, even the training and its organization, the conferences that were mentioned, is truly a collaborative effort. And it will not happen with any one agency. And I don’t know if Don wanted me to wait for questions? Okay. Thank you. Sheryl Shenberger: Our final regular agenda speaker is going to be John Fitzpatrick. He’s the Director of the Information Security Oversight Office. And he’s going to spend a few minutes talking about declassification policies. Following his talk, Don McIlwain will take the stage and handle the question and answer period. John Fitzpatrick: Thank you Sheryl. Good morning everybody. It’s a pleasure to be here. Doug mentioned the challenge of staying awake for the next ten minutes. It’s 25 minutes later then when he set that challenge. But there is not another room, and not another audience in the world that can stand up to that challenge better than you. You came here and we’re grateful for it. So I appreciate that, and your time and attention. I want to talk a little bit – the higher challenge, I don’t have slides of IT systems and statistics. We’re talking about policy here. So a bigger challenge. I do want to put a plug in to invite you back to this room in about a month. There’s a flyer out front, save the date. On the 27th of September, the Public Interest Declassification Board will hold a public meeting and release publicly its report to the president on recommendations for transformations to the security classification system. The release of Executive Order 13526 in December 2009, in the accompanying memorandum, the president tasked the Public Interest Declassification Board with coming up with these recommendations. Over time and in public forums and blogs, and other inter-agency activities, the board has discussed these issues with all kinds of constituencies in and outside of government. And they’re ready to say what they have to say. So I want to draw your attention to that and invite you back to this public session. And there’s a flyer out front that will highlight the details for you. Executive Order 13526 laid out a number of important changes related to both classification and declassification policies. It’s important to consider these as a matched pair, if you will. The classification system is an end-to-end life cycle system. Its parameters are affected by what gets into the system, how long it stays there, and what it takes to get it out. And we’re hearing a lot about what it takes to get it out. And from a policy standpoint there were changes made relative to both ends of that that are important. And I want to highlight a few. And there might be a way for me to share those slides with you, but I don’t know what it is. So let me ask for a little help. I promise there’s only two slides. Alright. A list of some of the policy elements that I wanted to highlight for you today. The first two are particularly important to have declared over the signature of the president. If there is significant doubt about the need to classify information, it shall not be classified. This was perhaps an aspiration of many who consider that information should be declassified in its proper time. And I think a purposeful defense against over-classification by placing this in the order as close to the top as possible. As close to those words that Bob had highlighted with regard to making information about the activities of government available to the American people. And so by no means do we expect that every classification decision is considered every time information is considered for classification. It would be great if it were. But we have it in there, and it’s an important defense. When individuals are considering whether information should be classified, and whether there is significant doubt, this is there as a provision to provide a little push back when needed in ensuring that information that is on the bubble really does need to be classified if it gets there. I’m happy to report I have utilized this as recently as this week in negotiations with another agency about publication of declassified documents. And so this is there, and it’s important to be able to call on. Similarly, with the next one. No information may remain classified indefinitely. Information gets classified and it gets thrown behind that wall, and it stays there for a very long time. And as all manner of constituencies consider getting it back to the other side of that wall, there is a dynamic tension there as to how long it needs to stay there. The provision that it not be there forever is an important one, particularly as we begin to negotiate the release of some significant information that has some age to it. This is an important tool. The Fundamental Classification Guidance Review, referred to cryptically as the FCGR, was a provision put in this executive order. And it was a page pulled out of the Department of Energy’s playbook. The Department of Energy had done a classification guidance review in the mid-90’s base-lining all of its guidance and reasons that its personnel and programs would classify information from that point forward. And so the framing of this document inserted that practice and made it a requirement for all departments and agencies that have classification authority. It required those agencies to conduct that review and report out on it by June of this year. And I’m happy to report that later this week, we will post on the Archives website all of the unclassified reports that were sent in. There was an order to send those in to the Information Security Oversight Office. And provide some specific details about the review at each agency, how it was conducted, some statistical information about it, and what were the results. There’s lots of goodness in that. The discipline of going through an agency’s programs and examining what time has retired the need for classification on certain information. And to give everybody the opportunity to conduct a fresh start on some of that stuff. Certainly there is better classification guidance as a result of this. What is less clear is how much information that would have been classified under the old guidance is no longer classified under the new guidance. We sought to elicit this information from department and agency heads. It’s a very difficult thing to do. It’s not directly measurable as information elements are hard to quantify. But we’re not done trying to get at that aspect of the classification guidance review. One other point on this practice is that this is not a one-time shot. We will revisit this with agencies periodically, and again in about five years. So that’s something that we’ll have the opportunity to get better and better at as time passes on. There’s also a requirement to list source materials for derivatively classified records. This is important because at the time of declassification it will aid declassification reviewers in identifying what material in there caused the document to be classified and to examine that more closely. The more information you have about the reason for classification can lead to a more effective declassification review. No information may be excluded from declassification based solely on the type of document on which it is found. This was a very popular provision in the order for those who want access to historically-significant information that was often in a form that was so special that release of it caused a lot of debate within the community. And having this declaratively stated has been an aid to getting at the information that’s in those documents. Not the form in which it is. And to examine that information as to whether or not it merits declassification on its own basis. When making declassification decisions, agencies shall consider the decisions of the inter-agency Security Classification Appeals Panel. So as the highest level panel for dispute resolution in the declassification process, the ICE CAP effectively sets precedent in the kinds of decisions that are considered by the liaisons and by the members of the board. And so often first-time declassification of types of information happen as a result of the panel’s deliberations. And so once that first time has occurred, subsequent times should be possible. And so it’s important for agencies to recognize that withholding on the basis of a certain type of information that has been declassified as a result of ICE CAP review shouldn’t occur. To aid this, the order also included the last bullet on this form. The ICE CAP is doing the form of public declassification decisions. This is a new function for ICE CAP. And we expect that by the end of September we will have a new section on our website that will host these releases of information. It’s a bit of a follow-through beyond where typical ICE CAP decisions are made. And then the appellant and the agencies involved get notified of that. There is a last 6-8 day period to make sure that all the proper versioning of documents is complete. And any other reasons to withhold information are considered before release of a final document to the appellant. But we want to make it also available to others, so that they can see the kinds of decisions that are made at the ICE CAP. And the second-to-last provision, the reason we’re all here – the executive order also created the National Declassification Center. So the important aspects of what we’re doing. Automatic declassification was also significantly addressed in changes in this order. In the predecessor Order 12958 it was established, but clarification was necessary as to what happenned to records after the 25 year exemption period. So in 12958 the 25 year exemptions and the reasons that those would be available were established and enumerated. But what happened after that was less clear. And so practices came into being in the intervening years, and those were codified in 13526. And that set milestones for types of information to have exemptions at the 25 year mark, at the 50 year mark, and then at the 75 year mark. So let me say a little bit about that. 50 and 75 year milestones. So let me back up. The previous exemption at 25 years, the practice that was established was that would be reviewed again at 50 years. What would happen after 50 years was less clear. And so what we’re talking about here is – 25 years is late 80’s. 50 years is the mid-60’s. And 75 years is the late 30’s pre-war period. And so there are significant historical record sets and events in history that fall into each of these time frames. And they are of interest. So it’s important for us to clarify what would happen. So the order already established, as was discussed earlier, that identities of human intelligence sources and key design concepts of weapons of mass destruction were already exempt for 50 years. But there are exceptional circumstances and types of information where even at the 50 year mark, information requires continued classification. And so the provision was made that the ICE CAP would hear the justifications for such information and render a decision to make sure that the guidance is as finely tuned as possible. And that the declassification community is absolutely clear on what earns this additional protection and what does not. And so the ICE CAP, as you may be aware, consisted of senior representatives from the Department of Justice, Defense, the National Security staff, the Archives, and the Office of the Director of National Intelligence. And they don’t work as a standard inter-agency panel representing their own agency equities, necessarily, but rather as a panel that’s advisory to the president. And so the ultimate declassification decision authority that exists is the president. And this panel hears these cases and deliberates on the president’s behalf. The president himself can consider and overturn and decision of the panel, but that would really be the last bit of resort. The last element here to discuss is that implementation of the declassification exemptions or review at the 50 and 75 year marks will ultimately fall into the protocols established for the National Declassification Center. The completion of significant activities that are going on now in terms of backlog. There will be prioritization of other records awaiting review. And these records would fall into that work load and come for consideration. So with that I think I’ll stop right here. And I think our next move is to Don and the Q&A. Don McIlwain: Thank you. And this is probably the moment many of you in the audience have been waiting for. If I can have my microphone folks on either side. If you have questions for the panel – Sheryl and John have joined the panel. I’m coming over here to do a little MC work. If you have questions, please line up. And what I’ll try to do is take questions fairly, one from each side. I also have a number of questions that were submitted in advance, and I’ll try to work those in as well. If I do not get to any of the advanced questions, I will be responding to those with input from our experts over the next few days on the NDC blog. We’re going to try with the blog to make it more active. For those of you that don’t know, go to Archives.gov/blogs/ndcblog. And it’s right there. So without any further ado we can go ahead and have the first question, please. Please let us know who you are. Steve Aftergood: Hi, I’m Steve Aftergood with the Federation of American Scientists. First, thank you for the work you do. You probably don’t get thanked very often, but we recognize it’s important. And so thank you. Last year at this meeting I was very excited by Mr. Richards’s statement that almost all of the imagery from the KH9 Intelligence Satellite was due to be released within a matter of weeks or months. I wrote down “almost all” in quotation marks. That did not happen. I don’t think – I know it’s not Mr. Richards’s fault, but the fact is that expectations were raised and then not fulfilled. And that’s a shame. I would actually like to ask a different question, though. And Mr. McIlwaine sort of alluded to it, but the answer is not crystal clear to me. Are we ever going to reach a steady state level where the number of classified records coming in is matched by the number of declassified records coming out? The thermometer chart we looked at only looks at the backlog from January 2010. It doesn’t even mention all of the records that must have come into NARA since that time. All the time they must be coming in. Are we ever going to reach a steady state? And maybe that’s a question for Sheryl. Sheryl Shenberger: Well I can give you the official “I feel good about this” response or I can probably tell you what I think. So what I think is no. There will always be a little bit of a gap. I don’t honestly think we’re going to get to this 400 million page backlog again. I don’t think we’re ever going to fall into that again, honestly. I think we have fixed these processes to the point where – and I really mean this – that we will never fall into a hole this large. I can tell you that the stuff that’s coming in – the important historical records that have come in since January 2010, we are tracking that. We’re getting ready to set up a team to go after those, take a look at them, and evaluate them. Put them into the process. We’re not waiting to put them into the process. We’ll get them started very soon so that we don’t create another backlog. It’s not nearly like you might think. Maybe 25 million pages. Because remember, this is what’s accessioned to College Park. That’s when it becomes part of our problem, so to speak, to get through it. I also think the agencies are doing a much better job. Maybe not the first few years worth we’ll see an impact, but certainly as time goes on there will be a whole lot better work done. Using checklists and the form Don showed everyone. Again, things will get much better. The data will be better captured. The documentation will be better. So that’s why I really feel comfortable saying we will not fall into another 400 million page hole. I’m not sure you can ever have a perfectly steady state. I think you’ll always run into complex collections, complex equities that are just going to take longer. And we can sit on them and not do anything with them, or we can go after them. And when you start going after those, that’s what’s going to take your time. So while I don’t think we’re going to ever get in any kind of huge hole – and I do see us at some point really getting close. I don’t think we’ll ever be perfect at the in-go and out-go. Don McIlwain: Over on this side please? Jeff Morley: My name’s Jeff Morley. I’m a journalist here in Washington. I write often about the CIA. Thank you for this public forum and giving people a chance to comment. Thank you also for the online presence with the NDC blog. It’s a very helpful and efficient way for us to keep up with what’s going on. With that said, I want to say that I think the NDC has made a terrible mistake in the decision that we were notified of in June that the CIA will not review 1,100 documents withheld under the JFK Records Act until 2017. We were notified that the CIA will not, and the National Archives will not review these records in time for the 50th anniversary of JFK’s death. The NDC asked for public comment on NDC priorities. And online it was clear that the majority of people who spoke online were in favor of review and release of this material by a clear majority. Bill Kelly, one of the people who responded there did an online petition calling for a review and release of this material. That petition has been signed by 700 people. I spoke with Dr. Larry Sabato of the University of Virginia about this decision. He said that it was deeply disappointing. It deprives everyone of a fuller understanding of the JFK assassination. The 50th anniversary of that terrible event is the perfect opportunity to shed more light on the violent removal of presidency. This adds to the widely-held public suspicion that the government may still be hiding some key facts about President Kennedy’s murder. We don’t want that. The common-sensical position for the NDC, for the CIA, for the National Archives, and for everybody is that if the CIA has time and resources to review sensitive historical material on important issues like the Cuban Missile Crisis, the Katyn Forest Massacre, it also has the time and resources to review records about the JFK assassination. We’re not talking about a large body of material. There’s 1,100 records. Informant information can easily be removed from this. And this material can be made public. At the beginning they said, “Is there significant doubt that there’s any problem with this material being declassified?” There is significant doubt that this material should remain classified. It should be declassified, and it can be easily as part of the NDC process. So my question would be for Sheryl. Would the NDC be willing to reconsider its decision in this matter? And my question for the representative of the CIA would be, would you be willing to have a meeting in which we could identify JFK-related material that could be reviewed and released in time for the 2013 anniversary? Thank you. Don McIlwain: Coming up is our general counsel Mr. Gary Stern, who has some things he can address on that. Gary Stern: Thanks for your question. We had a feeling somebody would ask this question. And that’s why they asked me to be here to respond if necessary. Jeff is referring to a June 12th, 2012 letter that I wrote to Jim Lazar responding to his query earlier this year requesting that we expedite the disclosure of the withheld JFK assassination records. The approximately 1,100 documents that have been postponed for release until 2017. Expedite them for release in time for the 50th anniversary in 2013. And that is something we looked into, took very seriously, and explored. And in an ideal world, wanted to be able to do. There were two separate components of the issue. One was: Does it fall in formally to the NDC backlog project? Which is the what the focus of this discussion has been today. The 400 million page backlog to be completed by the end of 2013. And that was sort of how the question came in. And I tried to explain, hopefully as clearly as I could, that it does not fall into that project. That backlog is for the materials and the records that hadn’t been completed in the review that was required from the 1995 executive order. And just to give you a quick background, remember these JFK assassination records are subject to a whole separate law. The 1992 JFK Assassination Records Collection Act, that mandated accumulation of all these records and declassification review under very narrow standards for what could be continued classified. With a maximum absolute deadline of 25 years. So roughly – the 3-4 million pages that were collected and given to the National Archives from all the agencies. The vast majority was released in the five year period of the board. And then there were standards set up in the act that records could be postponed for up to 25 years. And some records are postponed for some interim period. Those were re-reviewed and released with the help of the CIA back in 2006. And then this proportionately small category of 1,100 documents continued to be postponed in full until 2017. And that review has actually been completed and the decisions made, they need to be postponed for that long. They are not sort of pending declassification review. They’re not part of the project. But nonetheless, because a serious request was made and we know there is serious public interest, we still explored if it could be done logistically as a discretionary disclosure by 2013. And we looked into it, and we consulted closely with the CIA. And all I can do really is read you what the letter says. The June 12th letter says we consulted with the CIA, and even though CIA shares NARA’s interest in wanting to be responsive to this request, they have concluded that there are substantial logistical requirements that must take place prior to the release of these remaining records, and there is simply not sufficient time or resources to complete these tasks prior to 2017 given all the component coordination that would be required to speed up the release of those documents. So as much as we appreciate and recognize the interest, as much as it would be a great thing to match it with an important calendar event of the 50th anniversary, there’s really no way it’s going to be able to be done given the limited resources and the other priorities. Including dealing with the backlog in 2013. So we’re kind of, I’m afraid, at an impasse. Which is unfortunate in the short term. The good news is it’s only another five years. I know it’s frustrating for people who want the records now. But in the long term, and from the perspective of the National Archives, it will happen in five more years. And there’s not a conspiracy going on. It’s just simply the logistics as well as the requirements of the statute. Remember, the statute was passed in 1992 before the 1995 order and the 2010 order. There wasn’t any other way of getting records out and the statute has been a very good thing. And it kept a very narrow standard, just like the executive orders do. Some records need to be withheld for a longer period of time. And that’s the best we can do. So sorry I can’t solve it for you. Jeff Morley: So just to be clear, the answer is the NDC is not willing to reconsider this decision? Gary Stern: Correct. Jeff Morley: Okay, and so the official position of the CIA amidst a government-wide campaign to declassify records, is that it doesn’t have sufficient time and resources to declassify records that are related to the assassination of President Kennedy. That’s the position of the CIA? Gary Stern: These particular remaining records. Remember, the standards of the JFK Act are by 2017 everything has to be released in full. And so these are the documents that have been withheld in full but need to be and will be required to be released in full by 2017. There’s no exceptions after 2017 except the direct appeal to the president, who must personally decide whether anything could still be withheld after 2017. So these are records that have enough complicated information that can’t be redacted, reviewed in a limited way. Because there are other records in the collection that were redacted and withheld in part. And in 2017 those have to be released in full also. These are records that at the time, back in the 90’s, approved by the board, were determined couldn’t be redacted and segregated. Had to be withheld in full. Even in your comments you said, “Oh, they could easily withhold and release the source information.” That review was already done, and the decision made as to whether you can do that. And they were required in the statute to redact as much as possible where they could. Small number of documents withheld in full. And the logistics of doing another review just to determine again that they’ve got to be withheld in full until 2017, it’s not practical and it’s just not able to be done. And I know it’s disappointing, but we’re sorry. Jeff Morley: My second question was directed to Mr. Warrington. If he wanted to comment on this request and situation, that would be in the public interest. Robert Warrington: Gary has articulated the CIA’s position on this matter and I have nothing to add. Don McIlwain: Thank you. Over on this side? Bill Burr: Bill Burr with the National Security Archive. I have a question about finding aids. So far as I know, the collections that the NDC has processed and made available – researchers don’t have finding aids at the phylum box level. So a lot of researchers don’t know exactly what’s been made available. Now I know that it’s not the NDC’s responsibility, but it’s a major problem. Because the measure of transparency at an Archive is the availability of good finding aids. So I wonder if the NDC is going to be teaming up with other offices in NARA like reference to try to get them produced over time. I know it’s a huge problem, but it’s something that’s got to be tackled to get these records fully available to researchers. So we know what’s actually in the files. That’s my question, basically. Don McIlwain: Bill you make a good point. I’ll take a brief stab at that before I turn it over to the panel. As someone who does historical research in archives as well, good finding aids are indispensable. Mr. Burr was very correct that a lot of times, all you get when we put the declassified boxes of records on the shelf is maybe whatever finding aid – a standard Form 135 – perhaps a box list. For joint staff records, Doug makes a great folder list. But a lot of times, those aren’t just there. We do work closely with our processing folks in the Office of Researcher Services as records are coming through the declassification pipeline to identify those so they can put those newly-declassified records into their processing plans to create finding aids. They’re getting stuff from us. They’re getting the unclassified records newly accessioned as well. So they’re getting a lot of stuff thrown at them. But we are acutely aware. And when I’m wearing my hat as the FOIA and MDR chief, I am very aware that good finding aids lead to better access to the records. And it’s partly a matter of resources. It’s partly a decision the Archives made to get the declassified boxes out there, even in an imperfect finding aid world. And work on building the finding aids through cooperation with research services. But that is a dilemma that I certainly share. And I don’t know if anyone else on the panel would like to add to that. But it’s a problem we’re aware of, and we do work on. Thanks Bill. Over here? Jim Lesar: Jim Lesar, President of the Assassination Archives and Research Center. Question for the representative of the CIA. Of the 400 million pages of historically backlogged records, what amount are CIA records? Robert Warrington: That originate with CIA or that contain CIA equity? Jim Lesar: Both. The total figure. Robert Warrington: I’m sorry, I cannot answer that question. I don’t know. Sheryl Shenberger: We’re not trying to keep anything from you. There’s not that much original, yes? Robert Warrington: I doubt there’s very much of the original CIA records in there. And the percentage that have CIA equity, it would be misleading for me to try to guess that for you. Because I simply don’t know the answer. Jim Lesar: Could you put it in very rough terms? Tens of thousands of pages? Hundreds of thousands of pages? A million pages? Robert Warrington: Again, because I don’t have any information that I can base my estimate upon, it would be inaccurate and almost certainly misleading for me to try to give you a figure. Jim Lesar: Could you provide that information? Robert Warrington: We can take it as a question. Jim Lesar: And post it on the blog? Robert Warrington: We’ll take it as a question, sure. Jim Lesar: Alright. Let’s suppose you had 1,000 documents that comprised say, between 30,000-50,000 pages. Given the improvements in your performance that you’ve described here this morning, give me a rough estimate of how long it might take to process those records? David Mengel: Which piece of the process? Are you talking from start to finish in the NDC, or just an individual agency’s review? Jim Lesar: Talking about you reviewing records as part of the NDC project for declassification. I want to know if you’ve got 1,000 documents, and they fall within the range of 30,000-50,000 pages, how fast you can do that. You seem to have speeded things up greatly, so you’re going to be able to review and process records much faster than previously. I’d like to get you pinned down as to roughly how fast you can do that? David Mengel: If I can just clarify, there’s several different review processes we’re talking about. One is the original agency’s page-by-page review that they conduct to do the original declassification work. Is that what you’re referring to? Jim Lesar: Well let’s take each one of them. Start with that. David Mengel: I’ll leave that up to you and Ken to determine how long you take to do a page-by-page review. And maybe Doug can chime in on that. I don’t know specifically for each agency. Robert Warrington: The answer is highly variable. If you have 1,000 documents, it depends on the amount of CIA equity that is within the 1,000 documents. That largely determines how much time it would take us to review them. And it also depends on why we’re reviewing them. Are we reviewing them simply for missed equity? Or are we reviewing them to render declassification decisions upon the documents? There’s a big difference between the two. But again the primary driver is how much CIA equity is in that 1,000 documents to which you refer. Jim Lesar: Okay, well say it’s a hard case. How much time would it take? Are we talking months? Are we talking years? David Mengel: Let’s just focus on initial review at this point. How long would it take you to do an initial agency-level review if they’re your records, to do 30,000 pages? Worst case scenario? Robert Warrington: Worst case scenario? If it’s heavily laden with CIA equity and they are very complex documents in terms of the nature of the equity and how it has to be interpreted for purposes of release in full, release in part, or exempt in full, 30,000 pages could take us a couple months. Jim Lesar: Alright. David Mengel: Did you want to hear from some of the others as well? Jim Lesar: You’re familiar with the JFK Act? Or should I harass Mr. Stern about that? Robert Warrington: I have no comments to offer on that particular topic. Perhaps you’d wish to direct your question to Mr. Stern. Gary Stern: Jim, we understand. You’re trying to say if it only take so long to review 1,000 documents, why can’t they spend that time on those 1,000 documents that are remaining? Jim Lesar: Well let me ask one further question before we get into that. Gary Stern: We’ve got a line of folks here. Jim Lesar: I understand that. You would concede, would you not, that review under the JFK Act should precede far faster because the JFK disclosure standards were far more liberal? Gary Stern: To try to sum up the issue: These 1,100 documents were already reviewed under the JFK Act. Which has a much higher standard of what can be and has to be disclosed during these time frames. Not ever having seen the records, but with my understanding of the law and the standards and all, if these same 1,100 documents were reviewed under the executive order, almost certainly they would meet the exceptions for 25 years, for 50 years, because of human sources. So the point is, under the executive order, even if they were re-reviewed they almost certainly would be exempted and not disclosed. Under the JFK Act, the much higher standard, the mandatory 25 year deadline, these records are going to be opened at 25 years even though under any other standard that currently exists they would otherwise still be closed. So the notion of putting them through a new declassification review under the current executive order, whether it’s a 25 year or 50 standard which still has exemptions for human sources, all sorts of narrower but existing exemptions, they probably would be withheld. And so for those reasons it doesn’t make sense to put them through another review. Jim Lesar: Is there any reason why the National Archives – doesn’t it have the authority under the JFK Act to review them under the JFK Act terms? Gary Stern: We’ve made our decision, we’re stuck, you’ve got to wait five more years. Don McIlwain: Thank you, let’s move on. Thank you Gary. Over here? John Judge: Well not to belabor a point, but I’m John Judge of the Coalition On Political Assassinations. And we were formed in 1994 in response to the JFK Records Act, to help implement it in large part. That act was passed because in 1990 a film, JFK, by Oliver Stone came out. And there was mention that the files were still locked up. 75 years for the Warren Commission Records on presidential order, and 50 years under the House Select Committee Congressional rules. And there was tremendous public clamor and pressure which led to the JFK Act. As far as I know – you archivists may know better than me – it’s still one of the most, if not the most visited collection at your archive. And we worked with the board, which started not until 1994, until ’98 when it closed, to help them identify the records that needed to be released. In ’98 they released the final report. And I recommended it to folks here before, that you should read that report and its recommendations, but also its criticisms. Because they said that there was obstruction by certain agencies. And the nature of the obstruction, and what’s central to this debate I think, is that records that were JFK related records under the definition were never introduced for review. To the board. And were withheld even at that point. And new records continue to be found. A more recent example is that we understood that the Secret Service records of the Tampa visit were destroyed. And now we find that one of the Secret Service agents had the records and requested that the archive get them from him. And apparently that process is going. But continuing examples of records being discovered. The problem is that when they’re discovered and we ask for them to be released under the JFK Records Act, the agencies revert to the Freedom of Information Act standards. Which are much stricter, and which delay – there’s no presumption of release. If there’s redaction there’s rarely substitute language. And the standards of appeal and other things change completely. So many of these records, I think, should still be under those standards. And by 2017 the Archivist is supposed to sign off on a full release, as you’ve mentioned. We have also been working on a Martin Luther King Records Act. And these records are also similarly – in part because they’re congressionally locked up – outside the standards of this review. About the life and death of Dr. King. And we’re informed that certain records, even in 2017, may not be released. And the certain records that have been released – Don McIlwain: Is there a question, sir? John Judge: Well the question I think is, I think the process – the JFK Act was passed because the FOIA wasn’t working. And I’m not sure that this process continues to be working in any good way. You mentioned that you want to take people out of the loop with these machines. Most of us people have been out of the loop with or without machines. And we’re the public, and we’re trying to get access to our own history. It belongs to us, and the national security state has basically stolen it from us. Don McIlwain: So is there a question? John Judge: Well I’d hope you would think about this. You’re facing not only 25 million pages of printed material. You’re already at a petabyte every 18 months of digitized material. Which is beyond the ability of humans to recognize. Don McIlwain: That is what we were talking about a little bit earlier. Using smart ways to do review. And so we are working at that. In particular, probably the preeminent collection of electronic records that are accessions, the State Department cables, we continue to work with our partners at the Department of State. With CIA and energy to make sure that when those records come in, they are properly reviewed. And how do we continue to make those available. So yes, that is working. John Judge: I hope the process that you’re going after does work. I’m afraid that you’re plugging holes in a bucket to empty a spring-fed lake. Don McIlwain: Well thank you. Over here. Jim David: Jim David from the Air and Space Museum. How many classified records have been accessioned since 1 January 2010? Sheryl Shenberger: As far as under our purview, it’s only about 25 million pages. That’s why I said it’s not coming in at such a huge amount that we’re not going to be able to get our hands around it. The rest of the stuff that’s restricted and protected is protected under different restrictions of national security information. So that’s not something that the NDC itself has to address. Don McIlwain: And to follow up with Sheryl just briefly, we deal with national security classified records. Records that come in that are protected for reasons of privacy. Records that may be law enforcement investigative records that are protected under the B7 parts of the FOIA. We do not deal with those unless they are also national security classified. We do however work collaboratively with the folks in the archives and the agencies on issues of personal privacy, on law enforcement, and even as some of the records are not as old that are coming in, we have to bring in concerns of business confidentiality and stuff like that. So we do work with that, but there are situations where it may not be available but it’s not a national security classification issue. Jim David: Just to follow up on that, the authority of the NDC to declassify is for NSI and code word material. Obviously not RD and FRD. Is that correct? Don McIlwain: That’s correct. The RD and FRD is the realm of Ken Stein, and I’ll let Ken follow up on that a little bit. Jim David: I’m just trying to – Sheryl Shenberger: NARA does not have the authority to classify, NARA has the authority to release. It’s up to the agencies that own the equities who do the declassifying. They’re the ones with the authority. That’s why the NDC itself has to be so collaborative. Because the agencies themselves make the decision on their information. Jim David: Maybe I’ve had a mistaken impression of the NDC’s charter. But I’ve understood it that all the agencies working together, and obviously RD and FRD is exclusively or in part a DoE decision. But that the NDC only has the authority to declassify NSI and code word material. And RD and FRD is not declassified as part of the process. Sheryl Shenberger: Right, they’re excluded. Jim David: Okay, thank you. Don McIlwain: Thank you Jim. One more question from over here? Sharon Bradford Franklin: Thank you, hi. Sharon Bradford Franklin with the Constitution Project. My question is primarily, I guess, for John Fitzpatrick. You outlined in your remarks a number of policy changes that were commendable, that were made with the executive order. And among those you listed the fundamental classification guidance review and noted that that process seems to be resulting in improved guidance and better clarity. But that it’s unclear whether it’s helping with the problem of over-classification in the first place. And you also mentioned a change in the new rule that if there’s significant doubt, don’t classify it. So I’m wondering what steps if any ISOO or any other entity is taking to assess that. And trying to figure out whether we are, with the fundamental classification guidance review, helping to reign in the problem of over-classification. And whether we are in fact implementing this, “If there’s significant doubt, don’t classify it.” John Fitzpatrick: So with regard to the fundamental classification guidance review, we are following up with the reporting entities in those agencies to get better answers to the question I asked in my January memorandum, where I asked agencies to provide an estimate of this amount of information. How much information, as a result of the decisions you made, would no longer be classified that previously was? So that question hangs in the air, if you will. And our intent is to follow up with agencies in regard to that. The broader question of over-classification and the implementation of the one provision about significant doubt. I think that would be, in a traditional assessment process, which is what ISOO conducts with regard to classification and declassification activities and the data collection that we perform on an annual basis, it would be very difficult to measure that one. How often were you able to invoke this to not classify information? My guess is your interest would be in how much information wasn’t classified as a result of that? And that’s sort of a dog that didn’t bark kind of question. We primarily focus on systemic performance measures about what gets classified and why. And you can do analysis and perhaps consider other measures in the policy realm that would over time take a shot at over-classification. I think the insertion of those provisions in this executive order is a direct result of that kind of analysis over time. Over-classification is a very fresh topic with us. We have as recently as two weeks ago been up on the hill addressing staff members and concerned committees about it. I think there are other ongoing discussions about it within the administration. We’ll see over time how they bear out, but we at ISOO bring into bear all the information that we have. In fact, we’re the one source for this kind of thing. And in all of those venues, making sure that we help to shape – if something is done about over-classification, we’ll be there in the room. That’s as much as I can say for you now, Sharon. Don McIlwain: Thank you folks. There are a couple of questions. Unfortunately we are past our time. There are a couple of questions relating to RD-FRD reform. Relating to what’s on crest. Relating to prioritization of high-level records in the Defense Department for declassification processing. I will be posting answers to those on our blogs. If any of you think of other questions, please e-mail them to me. [email protected] I would like to thank our panel, I would like to thank some of our guest speakers who came in and chimed in with some help, and most importantly I’d like to thank all of you for your interest and your helping us work on our goal to meet declassification challenges. Thank you all, have a great day.