Transcript

FOIA Advisory Committee Meeting (Virtual Event)

Thursday, June 8, 2023

10:00 a.m. (ET)

Michelle [producer]: Ladies and gentlemen, welcome and thank you for joining today's FOIA Advisory Committee meeting. Before we begin, please ensure that you have opened the Webex participant and chat panels by using the associated icons located at the bottom of your screen. Please note all audio connections are muted, and this conference is being recorded. To present a comment via Webex audio, please click the raise hand icon on your Webex screen. The raise hand icon is located in the lower toolbar. You will hear a beep tone when you are unmuted. At that time, please state your name and question. If you are connected to today's webinar via phone audio, please dial pound two on your telephone keypad to enter the comment queue. If you require technical assistance, please send a chat to the event producer. With that, I will turn the meeting over to Dr. Colleen Shogan, Archivist of the United States. Ma'am, please go ahead.  

Dr. Colleen Shogan: Thank you. And good morning. My name is Dr. Colleen Shogan. I am the Archivist of the United States and I welcome you all to the fifth meeting of the fifth term of the Freedom of Information Act Advisory Committee. I join you less than one month into my tenure as the 11th Archivist of the United States. Since I took the oath to the Constitution, the same oath taken by every public servant, I have embarked on a tour of National Archives facilities across our great nation. Recently, I visited the National Personnel Records Center in St. Louis and the National Personnel Records Center annex in Valmeyer, Illinois, that serve as the central repository of personnel related records for the military and civil services of the United States government, respectively. Reducing the pandemic related backlog of veterans records requests is one of my top priorities as Archivist of the United States. 

As I have told the over 2,600 dedicated public servants who work at the National Archives, our mission is straightforward and complex, strengthening our nation's democracy through access to the public records we hold in trust. If there are no records, there is no accountability, and accountability is crucial to American democracy. Here at the Archives, we hold a great deal of records in trust. Our permanent archival holdings include 13.5 billion pieces of paper and 835 terabytes of electronic records. I can best describe 835 terabytes by sharing with you that that prefixe tera is derived from the Greek word terra, meaning monster. So certainly every records agency manager and FOIA professional and the 121 departments and agencies across the federal government can relate to a monster number of records. Mammoth challenges such as these come with enormous opportunities, which I encourage the Committee to continue identifying. 

Since arriving at the National Archives, I was pleased to learn that one of my many roles will be to receive from the FOIA Advisory Committee sound and reasoned advice on improving the administration of FOIA across the government. Although you are still in your first year of a two-year term, I understand that today you'll be discussing draft recommendations, a draft recommendation regarding the legal privileges that agencies can use to withhold information in records responsive to FOIA under Exemption 5. I look forward to reviewing any recommendations that you pass or consider whether today or throughout the year. 

Next month, we observe the 56th anniversary of FOIA’s enactment. Upon signing the bill, president Lyndon Johnson noted in his signing statement that “a democracy works best when the people have all the information that the security of the nation permits.” No one, he said, should be able to pull the curtains of secrecy around decisions which can be revealed without injury to the public interest. Thank you to the FOIA Advisory Committee members for all the work you do as requesters and public servants to strive towards a more perfect FOIA. I am confident that together we can work together to help ensure that the United States remains, in President Lyndon Johnson's words “an open society in which the people's right to know is both cherished and guarded.” Now, I'm going to turn this over to you, Alina, to begin the work of the meeting.  

Alina M. Semo: Thank you so much Dr. Shogan. Hopefully you can stay and for a little bit and join us in our meeting today. But if you have to go, we completely understand that. I want to welcome everyone as the Director of the Office of Government Information Services, OGIS, and this committee's chair. It is my pleasure to welcome all of you to the fifth meeting of the fifth term of the FOIA Advisory Committee. It's hard to believe but here we are. And I especially want to welcome our wonderful new archivist, Dr. Shogan, who is less than one month into her tenure. We are very excited to be working with her. Welcome to Lauren Harper who hopefully has rejoined us. She was having some slight technical difficulties. Oh, Lauren, there you are. Hello. She was recently appointed by the Acting Archivist of the United States Debra Steidel Wall in March following our last meeting in March, actually, to fill a vacancy that had developed on the Committee. 

Committee member Eira Tansey, [I] know you're there, she has left the University of Cincinnati in late April to launch an archival consulting business. We're very excited for Eira. Because her seat is a non-designated seat, Eira will remain on the Committee, and she is going to be fully engaged with the Resources Subcommittee. So we're excited to have her stay on. 

And last but not least a change is ahead for Dave Cuillier. Professor Cuillier is leaving the University of Arizona, or have you already left, leaving in transit and headed to the University of Florida where he will direct the Brechner Center for Freedom of Information. So that's very exciting. David, we're excited about all the work you have ahead and all the great work your students are going to be helping the committee with. 

Okay. I want to welcome all of our colleagues and friends from the FOIA community and elsewhere who are watching us today, either via Webex or with a slight delay on our NARA YouTube channel. All of our committee members' names and biographies are posted on our website. I have a few housekeeping items to go through, and then we will launch into our full agenda today. I am advised that committee members Paul Chalmers, Stefanie Jewett, Adam Marshall, and Ben Tingo are unable to join us today. Kirsten, I'm going to turn to you as our Designated Federal Officer. I believe you have taken a visual roll call. I just want to make sure that we have a quorum. Can you please confirm that?  

Kirsten Mitchell: We do indeed have a quorum, which is, 13 and we have 16 today. So, welcome everyone.  

Alina M. Semo: Thanks Kirsten. I appreciate that meeting materials are available on the committee's webpage. Click on the link for the 2022 to 2024 FOIA Advisory Committee on the OGIS website. We will upload a transcript of minutes of this meeting as soon as they are ready in accordance with the Federal Advisory Committee Act (FACA). Kirsten and I have certified the minutes from our March 2nd meeting, and those along with the transcript are posted on the OGIS website in accordance with the FACA (Federal Advisory Committee Act). During today's meeting, I will do my best to keep an eye out for any committee member who raises their hand if they have a question or comment. I encourage committee members to also use the all panelist option from the dropdown menu in the chat function when you want to speak or ask a question. Or you could chat me or Kirsten directly. We'll try to keep an eye out on all those things. 

In order to comply with the spirit and intent of FACA I just wanna remind committee members please do not put any substantive comments in the chat, keep it only to housekeeping or procedural matters as they will not be recorded in the transcript of our meeting. 

As I mentioned earlier, we have a full agenda today.  So we are not planning on taking a break. However, if we're running a little bit longer than anticipated, I may call for a short five or 10 minute comfort break. If a committee member needs to take a break at any time, please do not disconnect from the web event. Instead, I encourage you to mute your microphone and turn off your camera. Please send me and Kirsten a quick chat to let us know if you'll be gone for more than just a few minutes, and join us again as soon as you are able.  

And a reminder to all committee members, please identify yourself by name and affiliation each time you speak. That helps us tremendously down the road with our transcript and minutes, both of which are required by the Federal Advisory Committee Act. 

Members of the public who wish to submit written public comments to the committee may do so using our public comments form that we rolled out earlier this year. We review all public comments and post them as soon as we are able if they comply with our public comments posting policy. In addition to the written public policy, sorry to the written public comments we have already posted, we will have the opportunity for oral public comments at the end of today's meeting as we do at every meeting. As we noted in our May 15th, 2023, Federal Register Notice announcing this meeting, public comments will be limited to three minutes per individual.  

Okay. Any questions? Let me just pause for a second. I'm just going to toggle back and forth, make sure our committee members are all good. I don't see anyone, I'm seeing some shakes of head, nodding yes. Great. Okay. Terrific. 

So on to our busy agenda for today.  First we're going to hear a briefing on the FOIA reference model from David Bloom and Dr. Eliot Wilczek of MITRE Corporation. We're very excited to welcome them today. We will next hear from the Modernization Subcommittee, which has at least two items, if not three, to bring to the full committee's attention today. I will let Jason and Gorka drive that present part of the presentation.  After that, we're going to hear from the Resources and Implementation Subcommittees, both of which have been quite busy. They will provide us updates on their work. And as I mentioned earlier, we will close the meeting with a brief comment period from the public. Okay. I think we're running pretty close to on time, so that's good news. 

I am very happy to welcome from MITRE Corporation, David Bloom and Dr. Eliot Wilczek. David is a senior data management and business process analyst at the Mitre Corporation. With over 30 years in information technology consulting, David began his career focusing on business intelligence, data warehousing, and data management. Since working at MITRE, David also led business process redesign efforts across civilian and defense agencies. Prior to MITRE, David worked at American Management Systems, Wipro Technologies, David, sorry if I'm mispronouncing that, and Decision Path Consulting. David graduated from Carnegie Mellon University with a Bachelor of Science in Information Systems and Industrial Management. Welcome David. And Eliot is a records and knowledge management engineer at MITRE, where his work focuses on knowledge management, records management, and FOIA. He previously worked as a records manager and archivist at higher education institutions, and he has a Ph.D. in Library and Information Science from Simmons University. So David and Eliot, welcome. We are so very happy to have you here, and I hope you will entertain questions from the committee as we go along in your presentation. But we are going to turn it over to you now and let you drive. Thank you.  

Eliot Wilczek: Okay. Thank you very much. Alina. Can, can you hear me okay? All right. Very good. So, as Alina mentioned we are going to talk about a business reference model for the Freedom of Information Act and focusing on a presentation of the FOIA reference model, which the MITRE Corporation developed through its Independent Research & Development Program. As Alina mentioned, I've worked in records and records management a long time. Currently I've been supporting FOIA projects at MITRE. And I've found the work very interesting. And of course, it's you know a process and a law that's vitally important to the health of our liberal democracy. So if we go to the next slide the questions that we're going to address are who's MITRE and why do we focus on FOIA? What can defining FOIA business standards, how can defining FOIA business standards, excuse, excuse me, address agency FOIA challenges? And what are the components of the FOIA reference model, and how will it be and how can it be used to establish government-wide FOIA business standards?  

So MITRE is an organization that operates six federally funded research and development centers, as well as supports MITRE Labs and has an independent research and development program. Together with government and public public-private partnerships MITRE works to improve the safety, stability, and wellbeing of our nation. Next slide.  Federally funded research and development centers are organizations that promote objective collaboration to solve large scale problems. They serve as long-term strategic partners to government, providing objective guidance in an environment free of conflict of interest. Today, MITRE operates six federal funded research and development centers working across government in a partnership with industry to tackle challenges to the safety, stability, and wellbeing of our nation. Next slide. You know, MITRE works in a wide range of challenges such as air traffic, collision avoidance, and electronic health records. To give just two examples. Next slide. 

So, what did MITRE focus on? FOIA? So, we've supported agencies' FOIA processes in various ways. This includes defining requirements for FOIA system acquisition, assessing agencies FOIA programs, enhancing programs processing, and preparing interim solutions for agencies and MITRE’s also developed a natural language processing and artificial intelligence enabled FOIA assistant tool that streamlines exemption identification work. Our sort of general observations from these engagements that we've had is that FOIA processing can be both very labor and time intensive, yet inefficiently supported, and that implementations tend to be agency specific and vary in functional and analytics support. Look at the next slide. We've sort of thought about these challenges that we've observed, and I'm sure many people at this meeting have observed it in three areas. One is around the difficulty and challenges of meeting demand. The second area is around inconsistencies in release. And the third is around challenges with monitoring and measuring FOIA processes and activities in a way that can identify and enable opportunities for improvement. 

So, on the next slide, um, this sort of leads to the question of how can government-wide FOIA business standards address those challenges? So, I think most importantly and centrally, they provide a common language that can empower agencies to be more informed and precise in their conversation. Amongst FOIA staff, FOIA solution and service providers and other mission support functions having agreed upon government-wide FOIA business standards can help enable agencies to: identify inefficiencies and gaps in agency FOIA business processes and workflows; help them ensure that FOIA staff training is consistent and comprehensive; enable evaluations of FOIA solutions and service offerings, and make sure that those are thorough and systematic; identify and integrate process touchpoints, data exchanges, and related technologies with other agency mission support functions, for example, FOIA systems talking to financial systems about FOIA fees; and identify resources needed based on activity and performance measures, systematic and measurable, and bringing those measures against concrete targets for FOIA business operations. Next slide. 

So, the history of the FOIA reference model. In 2021, MITRE initiated its work on the FOIA reference model through its Independent Research and Development Program. As that work at MITRE proceeded in 2022, we engaged with the Chief FOIA Officers Council's Technology Committee to stand up the Reference Model Working Group whose members are on the slide. That working group provided invaluable feedback to us to help validate and improve the reference model as we were developing it. That working group also published on their own a white paper about the reference model, and that white paper is posted on the CFO Council website and we published the FOIA Reference Model in January of 2023. So just a few months ago. Overlapping that work as we were completing the FOIA Reference Model in November 2022, the Department of Justice Office of Information Policy (OIP) was designated as the lead agency for developing the government-wide Federal Integrated Business Framework (FIBF) to develop the FOIA business standards. And this is something that was mentioned at this committee's last meeting, and is highlighted in the Fifth Open Government National Action Plan. In January a few months ago, MITRE delivered to OIP an initial version of the FIBF FOIA business standards that were derived from the MITRE FOIA Reference Model. Next slide.  

So this is looking forward a little bit to what happens next with the FOIA reference model. As mentioned the OIP and the FIBF FOIA working group will continue to draw on the MITRE FOIA Reference Model to incrementally develop the FIBF FOIA business standards.  That process involves review and feedback.  This includes the business standards being posted on regulations.gov to get feedback from other agencies, other FIBF functional areas such as financial management, which I touched on earlier, industry and shared services and of course the public. The FIBF FOIA business standards will be reviewed and approved by the Office of Management and Budget. And the business standards will be incorporated into federal solutions and service requirements such as GSA’s (General Services Administration) multiple award schedule. The FIBF on your right, the wheel and the upper right hand corner, outlines the five functional areas of the FIBF. And then the table in the lower right hand corner outlines some of the connections and how the reference model is informing the business standards. So, on the next slide, I'm going to turn things over to my colleague, David Bloom, who's going to talk about the reference model itself and our approach to developing it and some of its components. So, over to you, David.  

David Bloom: Thank you, Eliot. Can you all hear me and see me? Okay, I'll take that as a yes. Okay. Thank you. So before…we're going to spend this time on this section to talk about what this model actually looks like, we'll open it up and look at some of the pieces. It's a fairly dense document, and we're going to help you navigate that document so you have a better understanding of how you might use it, how you could, you know, reference it.  Just real quick, how did I get into FOIA in the first place? Well on a MITRE project, I was brought into an agency part of which one of the divisions was using our dear friend and tool Excel to manage their tracking of FOIA requests. Not surprising necessarily, but, well, I do love Excel, and I've done a lot of work with Excel [it] is by far no means the proper tool to manage the process. So I kind of got a glimpse from the sort of maybe one of the worst ways to manage the requests and the process to having an opportunity to look at, you know, a more comprehensive approach and, and thinking about all the piece parts that that could really go into a model for executing the FOIA process.  

Okay. So before we dive in, let's just be clear about when we say reference model, what do we mean? Lots of definitions out there. I've just checked, plucked these two, one from Carnegie Mellon, my alma mater. But like these two, because they kind of hit at two key points.  So one is a division of functionality.  We're talking about decomposition of a business capability into its smaller pieces and understanding how those pieces interrelate. It's really a focus on the what and not the how. And the second part is DoDAF (Department of Defense Architecture Framework). Some of you may be familiar with that if you're on the defense side. DoDAF is simply an architectural framework for, again, modeling business functionality. But one of the key aspects of DoDAF is, again, to model it in such a way that it's not specific to a particular agency. It's really about the function, which is also what the FIBF is going to do. You know, GSA’s FIBF work that Eliot just mentioned. 

So these two aspects of essentially being the what and not the how and being applicable to multiple agencies is really the purpose. It's a business architecture. It's this description of business. It is not a description of technology or how technology would necessarily be applied, although there are hints in some of the reference model as to where certain types of technology may be advantageous. I keep wanting to hit my arrow button to move the next line. I got it covered here. 

So there are a few principles that we wanted to adhere to model this FOIA work. One is to use sound systems engineering practices. We at MITRE, we do follow that every day. But also to make sure that it's a description that can stand up to some scrutiny from a systems engineering practice, we also want to make sure that it was relatively easy to understand.  While it is dense the diagrammatics that we used, use UML, unified modeling language, which is notably easier to consume than some other techniques. We used use cases which are really the activities that get performed. And in the model, you'll see there's literally hundreds of activities that get performed as the way we've decomposed it. And then user stories are another way of saying a business requirement.  We used to talk about system shall statements, you know, the system shall do this or shall do that. A user story is another way of doing that in a little bit more artful way where it talks about who's doing, who needs that requirement and why? And so that's the user story. 

So we wanted to have useful relationships and descriptions of the information. We did use some sequence modeling. These are workflow diagrams, but we very quickly recognize there is no one workflow. You know, every agency is going to perform things a little bit differently. So in our workflow diagrams, our focus was on key dependencies between the activities and not so much about a specific workflow. And in fact, you'll see in the model itself, we model workflow as a capability that can be extended and should be able to be extended if possible, especially for the larger, more complex agencies that should be able to define their own workflows especially in carrying out a particular request. We call those tasks. You'll see more about that later. We used user stories, as I mentioned, to model business requirements, and we wanted to make sure that there was traceability between each user story, that at each activity and a requirement or one or more requirements in the user stories and the use case or the activities. So each user story is a requirement. Each use case is an activity. We wanted to make sure that as we described a requirement, something that the system should do, it was aligned to one of the activities we had described in the model. 

You'll see in the, in the description, and we'll talk a little bit more about the roles, which are, we called the hats that people can wear, we wanted to make sure that they were fairly specific and granular, and that's to allow both complex and simple shops, FOIA shops to operate. We'll touch that on again a little bit. And then we also aligned the processes and specs to business functions, or processes, key processes. There's about 26 processes that you'll see in a moment.  

We also wanted to make sure we follow some good data driven approaches. There's some data models that describe not only the kinds of data you would expect to see in a case, but also the kind of data that you would expect to manage the process of the FOIA request itself. What information do you need to keep about a case? And what information do you need to keep about the process that's going on around that case? And you'll see this in the terms of configurable workflows, communications, time, accounting, and other areas. We also wanted to identify where we could find them, controlled vocabularies. These are, you could think of them almost like the pick lists that you would want to see if you were doing a dropdown. A good example here would be you know the exemption types, right? So, you know, whether you're talking about national security or the deliberative process, you know, these should be standardized. They already are. So those are specific to policy. And we also wanted to make sure that there was additional vocabularies that would be useful for process perspectives. And so that level of standardization, which can be completely customized by any agency, would allow for more rigorous analytics and consistent analytics by having certain data that is tied to, to sets of values and not just random text. And then the same thing similar with the, the processing [of] configurable rules and the processing of policies. Those are also user stories, but we sort of call those out separately so that they could be viewed as something that agencies would want to be able to configure in the data itself.  

We've talked about in the model that there's a need to integrate various FOIA data. It's the source data and the record keeping systems with the FOIA repository, and ultimately to the public records that get pushed into the FOIA library. Easy to do on paper, not so easy to do in real life, but we wanted to make sure we acknowledge that. And we've also highlighted some of the data exchanges that we think would be important, many of which, or some of which, you know, you already are doing and you're already managing between FOIA.gov and others and pay.gov. But we wanted to sort of call out others that we could identify that would be important. Those exchanges would likely be addressed in part or fully by the FIBF as well.  

And then we also wanted to focus on analytics, both advanced analytics, and what we would call traditional business intelligence or performance management types of analytics. From an AI artificial intelligence, machine learning, human language technology perspective, these are examples of where we think advanced analytics would play out and we've called that out in the document. Being able to search for records in source systems. And the for repository e-discovery can be certainly aided by advanced analytics, learning, you know, advanced learning systems, duping source system records. I know that's particularly challenging when you have email threads and the like…identifying likely exemptions and redactions. Now our sister project here, another MITRE effort that Eliot mentioned is the MITRE FOIA assistant tool actually supports that. It is actually delving into that with artificial intelligence.  

And while our specifications don't specify the models or how to do that exactly, we identify where in the model those things would play out, finding similar requests; analyzing the responses could help to have consistency in response; and then recommending processes for and queues. So, for example when a request comes in we need to route that to the right people, to the right process. Using AI could help with that as we start to see what kinds of text is coming through on the request. We could at least recommend using artificial intelligence to the intake function as to where that should go. And also tagging the request to aid in various ways. One way we'll talk about later is in, proactive disclosure as a matter of fact. So that's some examples of AI. Business intelligence, it's a little more traditional. It goes back to my old days as, back in the eighties and nineties, when we were doing this kind of work, but it's still valid today, is how do we measure the process and the performance of those to continually improve. Again, Eliot hinted at that as well, we've indicated in the reference model a sort of starter kit or draft set of that information as well. And we'll take a look at that in a few minutes.  

This is a schema, I'm not going to walk through the details, but essentially describes the kinds of information that is in the model. And these four major sections here, you'll see there's a functional model, a data model, requirements model, and an interaction model. And so what we're describing are business processes. We've got about 26 processes. They comprise over 300 use cases, also, we call them activities. And over 50 distinct roles, also called actors, depending upon the modeling technique, you're using. There's over 400 user story requirements. So these kinds of things can be fed into acquisition efforts, or they can be used to jumpstart a gap assessment, perhaps to look at, see where we are, what kinds of things we might want to consider. And then we've got, within those 400 to 40 configurable processing rules, 30 data classes. So this is, these are sort of, what do we want to keep information about? And over 60 controlled vocabularies. I mentioned that earlier. Those are things like the pick list, for example, that you might want to capture. And then the model is available as a PDF, but we also provide it through Excel and Visio. Those contain a lot of the models themselves. And those are actually downloadable.

Just to point out the structure of the document itself. So there's eight sections with a series of appendices [I] put [on] the left side as a kind of reading material, because I guess I feel like those are the high level information that's most consumable, but it is a reference model. And by that we mean you're not necessarily going to read through, you know, the details as a book, okay. You can refer to it where it helps, but it wasn't meant to be done that way. It is over 300 pages. I would argue that the first four or five chapters are more readable and this is the content that we have. There's an overview. We've got a description of the top level. We talk about the standardized exchanges, some relationships to other functional areas, we'll talk about that in a moment, a focus on the analytical capabilities and reporting. Again, that's sort of business intelligence here…all the detailed process descriptions, that's a fairly dense section. We've identified a bibliography, and then these are the appendices. So we'll leave it at that. You're welcome to leaf through it. 

Some of the key features I'm going to talk about in the next few slides are the types of cases, what we look at when we say case management in terms of the content of cases, when we say configurable workflow, to kind of give you a sense of that. The different records environments, a little word on proactive disclosure since it's such an important part of where we are. And then a little bit on fee management, just to clarify a few things.  

Okay. So there are different types of cases. This is sort of a piece of work that we start with an open and we need to get it closed. It typically would go through some sort of a workflow to get it done. And these are categories. They blew our categories of cases. So we have the overall case, and then these are different types of cases. And we usually think of the FOIA request as a case. We also treat FOIA request item as a case. And what that means is if a requester was asking for multiple things and they were really going to require, you know, two different types of search, or two different, maybe even multiple responses, we might divide up the request into request items. And so that's all that means. We're calling it a subcases as well. But also appeals and tracking litigations. Now, while litigations, we aren't trying to model the process of litigating in this document, that was never our intent. It's really about monitoring the activity that's going on in the peripheral, because we need to know in FOIA what's going on and how that will affect the outcome for requests. So, but we do track it, sort of, kind of a case, minimal case, I guess you could say. And then we have interagency actions like referrals, coordinations, consultations, and different types of inquiries, request or inquiries and mediation types. So it kind of represents the overall taxonomy of case types that we are referring to in the, I'll call it the firm federal, sorry, the FOIA reference model.  

If you want to take a look at the types of information there's a set of attributes, data elements that we have in the model that describe the case itself. We also know that those can relate to sets of notes and documents, and we want to track the communications that relate to the case. I mentioned [the] process queue before, you know, traditionally, or at a minimum, maybe we say that we need to track those that are in a simple, complex, or expedited queue. And that's a way to track the case, to manage its workflows, to prioritize perhaps the expedited. But in our model, we see no reason why an agency couldn't define, you know, more cues or subcues. Maybe there's, you know, a first person, you know, simple request. Maybe there's a first person complex request. You know, there this idea that we could allow for extensibility is a theme throughout the model. We also talk about being able to configure tasks and actions that relate to the case. And then different types of people and groups of people may be involved in processing the actions and receiving actions and doing tasks as well as capturing time sheet information. So this kind of a summary view  gives you a sense of the types of information that we think would be relevant, but it's not all the data that you would see about FOIA. This is just focused on the case portion.  

The configurable workflow concept is sort of in two parts. The right hand side is what we would call the template, which is the idea that an organization could set up and predefine its queues, which is at the top here, a processing queue and may have various actions, tasks, and actions. An action is something that is done to someone. A task is something that you need to do. So an action might be a notification. A task might be, you know, need to go get a certain mid-level approval before you can proceed, or you need someone else to look at it. Some subject matter expert, you could define a task for that, but these, this would be sort of templatized so that for certain types of process cues that are defined, we already have that all configured. And then the left side is essentially drawing on that template, applying that template, and maybe even further customizing it as needed for a given request. And so that's why you see the FOIA case on the left side, because this is kind of the real deal. This is the actual, and the right side is kind of the proforma that an organization might set up for as many cues as they want to define.  

In terms of records, environments, you know, we…talk about the three major types. The record keeping systems themselves, so that which we're drawing from the FOIA repository, which is the internal set of records, some of which may or may not get disclosed or even released, but it is, they were considered responsive. And then those that, a subset of those, not all of them, but a subset of those that may go from the repository into a public reading room, a.k.a. the library. And so this isn't meant that all go here, but just that's kind of a typical flow. So just for terminology, these are the terms that we're using in the reference model.  

We hit proactive disclosure in a couple of different ways. You know, it's a little, it's hard to get into this when you don't see the whole model, but I'll just kind of call these out just briefly. There is a process called provide pre-request support, which is really before a requester would even submit or necessarily submit anything. And the idea is to provide functionality that may perhaps preempt the need to even submit a request, right? So for them to know that maybe there's records published elsewhere to provide that kind of information, some content, perhaps through a portal being able to search effectively search the FOIA library. I'm sure that's not unusual. That's pretty typical.

And then also this would be wonderful if we could get to this point where we could identify, a requester could identify, where that information may be that other agencies may have. So, in other words, I know you can kind of get that sense from us from I think FOIA.gov. When you go onto an agency website, it would be great if there's some way to connect it so that someone isn't submitting a request for records that then have to get referred. We do that, but, or, or not just referred, but maybe it was the wrong agency. So you know, having that information available could certainly help scoping and science and the request. So at the beginning when we're sort of getting our hands around the request, you know, we could determine whether some of that is publicly available. And so that's why we have request items. Maybe some of the pieces of that request could be furnished through publicly available and some of it may not. So we can divide it up that way. And also to use topics or tagging, this is where the AI and machine learning could come in to help recommend in later phases when we were reviewing responsive records that perhaps maybe given agency policies and rules and that there's to proactively disclose that content. And that would happen in the review of responsive records. Also the three times rule is done there. And then finally, managing records, applying any agency criteria. So being able to capture what the agency is viewing as proactive disclosure for themselves.

So I'm going to….fee management. I'm not going to get into details of this. This was an interesting topic during our development. But just in terms of terminology, we have sort of routine payments. Those are the ones that you, you know, you pay when you get, you know, you pay after you get your result. We talk about pre-payments, which is prior to conducting the search advanced payment, which is prior to providing a response, we had to come up with a terminology. That's the terminology we're using just to keep them separate. And we do recognize that fee management does vary quite a bit across the agencies. Some don't even use fees. We're not trying to prescribe that you have to use this, but it is a framework from which you can draw.  

All right, the top level of the model, I know you can't read this, that wasn't the intent of this slide. Just to point out that and we'll look at this in a little more detail. We've got different agency actors, external actors, other agencies that have to interrelate through, you know, collaborations and coordinations and referrals. Other agency components perhaps for misdirected FOIA requests. The portal, government wide portal, the NARA OGIS reports, the AR (Accounts Receivable) systems where we deal with, pay information that's up in the upper right corner. And then of course, the requesters, even though the requesters are shown in the upper left hand corner, they're really inside the system, right. We want them to be viewed as internal because they're partners, they're players, they interact, ideally interact with the functionality more directly than some of the other actors. So they're kind of viewed as internal.

In terms of the processes, there's 26. We've got these core processes. The top level is really your traditional life cycle, right. We conduct an intake, we confirm that it's perfected. We scope and size it. We may estimate fees, we may not, depends on where that happens. We've got a search, we're going to review, we may have to coordinate or consult, we need to respond. And, then there's payment, processing and potentially an appeal. Some of the enabling processes, maintaining policies, managing HR, the HR as it relates to FOIA, managing the FOIA records environments, responding to referrals and, and, coordinations and consultations, modifying the scope of a request, which could happen from the requester or other internal, monitoring litigation. Again, not capturing and managing litigations, but monitoring them. And then managing user accounts. Supporting processes we have is providing free request support, managing the records for proactive disclosure, maintaining requests or dashboards and providing customer service. And then finally, overarching monitoring, analyzing, and reporting on FOIA performance. 

So without getting too hung up about the, how we separated these, because that's not that material. This is intended to reflect the full scope of FOIA activities. We'll see how that holds.  

This diagram is in the firm itself. I'm not going to walk you through all the details. Essentially what we're showing here is a set of dependencies that would be between those processes I showed you on the prior page. So how do these processes kind of interrelate with each other? This is not meant to be a sequence diagram. You're not necessarily following exact paths, but if you put yourself in any box, you could see that there's alternatives to where you might go from there. And that's a reflection of the different, you know, end points for any of these processes.  

We also identify common services. These are shared essentially like shared services that would provide additional capability to the activities that are performed. And so we've mapped these services to specific activities where we felt they would be most useful.  Things like email integration, using AI, ML (machine learning) and human language technology to enable decision search [and] review and redaction. We even have the FOIA processing clock being able to start and stop this. The tolling is a unique service depending upon where you are in the model. Sometimes we only can start and stop that clock, based on rules, right. So that's, so that's also in there.  

In terms of roles. There's over 50. It may seem daunting, but the roles are really based on the responsibilities because they're tied to the activities in the model. They can be used if you were defining a system from this model, they can be used to define system privileges. You know, who can do what and when we say who, we don't mean person, we mean that a person can wear more than one hat. They can serve more than one role. So in a large organization, you might find that it's more spread out and people can do less in their role. Perhaps in a, maybe in a smaller organization, a person might wear multiple hats. Maybe they're not only doing the intake function, they're also doing some of the confirmation, the scoping and other pieces of the process. And that's fine, that doesn't hurt anything. But as I put down here, you can wear more than one half, but you certainly can't tear a hat into pieces. That was the intent to have it to be fairly granular. It makes it easier to apply from a system standpoint. 

These are defined roles, you can take a look. I'm not going to walk through the details here, but, this is this set of roles that we can have in the model. You can see everything from, you know, the requests are certainly there. You've got the appeals roles, different appeals roles, different authorized official roles, intake specialists, intake supervisor, et cetera.  

In terms of the data components, I mentioned that there's a kind of a core data model. We've got core data attributes that describe the case and some controlled vocabularies. We talked about business intelligence. I'll show you some examples of that. And then opportunities for standardizing some data exchanges that go outside of FOIA, connect FOIA to other functions. 

This is a conceptual data model. These are the major data categories or entities about which FOIA would need to keep information. So these are not low level elements. They're groups of data. Each one represents a collection of information. And each of those collections is modeled on a diagram that kind of shows how those collections would kind of relate to each other. From a BI (business information) perspective we followed some basic principles that we can sort of decompose a process into the things that we'd want to do better or improve. We call them bios, but essentially it's, you know, we're trying to achieve something. We're trying to improve our performance. Performance measures are the ways, are the actual indicators of how we, how well we do against those. I'll show you an example here in the next slide. But this kind of framework was used in the model. I think this might be a little hard to read. But given one example under scoping and sizing the request, improving the accuracy of the estimated effort is an improvement. So if we want to do a better job of our estimation, we need data to be able to do that. So we have performance measures that align to that improvement. Again, because we're running [inaudible] here, I'm not going to go through all the details, but you can find all this content in the, in the model.  

And again, another picture of the measures. So we have the business process here in the blue, we have the improvement in the italics, and then the specific measures underneath that. And then the dimensions are the ways that you'd want to be able to slice and dice that measure. So I want to see the number of appeals by year, by component, by appeal, this position, et cetera. We've also indicated which ones are required for [the] DOJ (Department of Justice), which ones are not yet, perhaps are not and any other information about the DOJ, descriptions and the citations. So it represents a fairly comprehensive view, but it goes, is it intended to go beyond the DOJ reporting. 

We outlined some of the data exchanges I mentioned before. Some of these, you know, we already think that we're doing right. So we already coordinate with pay.gov. There's already feeds that happen there, but as you could see, some of these empty sets, I think, are not there for interagency, mostly interagency communications. So that might get addressed in the FIBF and so that could be helpful.  

And then we have in the document a little dissertation on each of these other functional areas that FOIA essentially needs to kind of communicate with or interact with. So records management, AR, human resources, litigation, and of course user account management. This is really an IT function or partially an IT function. 

The process flow diagrams. So this is a description pulled out of the document describing this process. This is the request intake process. Again, there's 26 of these processes and each of these little boxes are the activities being performed with, again, we indicate which roles are performing those activities. We indicate whether some of them are automated or some of them are manual. Some of them might be partially automated. Meaning requires a human to do it and some of them are fully automated. But that certainly would be up to an agency to determine. 

Again, these are not meant to be prescriptive. We're not saying you have to have this kind of interaction between these organizational entities, but these reflect largely the dependencies that would need to be taken into account. And so again, we've given you the Visio diagrams that could certainly be amended or extended for your own purposes. 

And then another view of that is the use case diagrams. These are the same set of activities, really highlighting who are the little figures on the left and right, who's touching these boxes. Not so much about the sequence, that's not really the intent here, but really who can do what. And then are there any dependencies between some of the activities. Similar to the prior diagram, in this diagram, we have an entry point and we have exit points. So these are possible exit points depending upon the scenario that comes out of the conduct request intake process. So you'll see that in all the diagrams, it's in arrows and out arrows.  

And then this is an example of, this is the list of activities we just saw on the prior diagram. So these are the little boxes on that page. And I highlighted the create and log a new case, because the next page is essentially a description of that. So again, every single, let me just go back for one second. Every single activity of which there are several hundred, include a description, which is on the left side of this page, from description all the way down to the supporting services, for each of those activities. And then one or more user stories. These are sort of like proforma requirements. In this case, it's just the system doing this, but typically would be like, you know, as the intake specialist, I need to do this or as the reviewer, I need to do that.  

And so that's how we've modeled this information. So I know we're just at the hour here at 11:01. That is what I wanted to share with you.  Again, it is a lot of content, but I think if you kind of focus on the first few sections it would probably help to introduce the content and then you can decide how much you want to go into the details. Thank you for your time. I appreciate it. And if there's any questions, I guess we will entertain them.  

Alina M. Semo: David, thank you so much. Eliot, thank you so much. I suspect our committee members may have some questions. So I'm going to turn it over to see who wants to go first. I want to invite Alex or Jason.  

Alex Howard: Jason, do you want to go or do you want me to hop on this?  

David Bloom: You can go first, Alex.  

Alex Howard: Okay. Thank you for that. Thank you for this presentation. I think I might need some time to delve deeper into some of these slides as they are indeed very information rich. Um, I'm curious as looking, you know, into these diagrams, the processes that you're digging into here, it, um, do these derive from research looking at how this currently works? Or is this an idealized schematic that describes how processes could or should work at agencies? Um, if so, what would be the next steps?  

David Bloom: That's a great question. So I'll take that one and Eliot and I are going to, we'll each take different questions. I'll take this one if you don't mind, Eliot, if you want to add to it. So yeah, we do, we do actually speak of that in the document we talk about idealized yet realistic. You know, while we've looked through, you know, all the guidance and everything that is in the various documents available that you all use, you know, on a daily basis, the intent was to make it as all encompassing as possible. So without, without regard to incrementally what would be better to implement or automate versus others, the intent was to be a superset of all possible capability. So that an agency when they're looking at that can decide, well, no, we, you know, we don't need to automate this or that piece, or, you know, we're simple enough, we do this well at hand manually.  That's fine. So idealized in the sense that it is a super set of as much functionality as we could muster. But also that it is an expression without waste. You know, the processes are designed to show no recycling, for example, sort of idealized in that sense. You know, if you were modeling a current process, you might see, some organizations might have handoffs that perhaps could be streamlined, right? So we've kind of taken all that out, and that's partly why we call it idealized. We also talk about it being realistic. And by that we mean that, you know, that there are technical solutions available to automate the things that we think could be automated. And so, and also that we wouldn't expect any changes to the law to be necessary to make that happen.  

Alex Howard: So to be clear, this is an idealized set of schematics that, in theory, agencies with complex processes could adopt and adapt and apply to their systems in the future.  

David Bloom: Yes, that's exactly right.  

Alex Howard: Okay. Um, so in that vein, I have a follow up question. On your slide of proactive disclosure, you talked about the relationship between FOIA.gov and data.gov, and the situation where, in theory, requesters would be able to search for information that's relevant to what they're looking for. I didn't see any reference to the Open Government Data Act or enterprise data inventories maintained by agencies. Are you aware of that statute of what it requires of agencies?  

David Bloom: Eliot, do you wanna take that?  

Eliot Wilczek: So I think they didn't specifically address those particular requirements in the reference model. This model is articulating and modeling exchanges between FOIA systems and other systems. And, you know, there may be additional systems, additional system communication that, you know, is necessary.  

Alex Howard: Okay. If you'd like to follow up and discuss that, I'd be certainly interested in doing that. I was involved in working on that reform years ago with the idea being that agencies, using the Chief Data Officers that now exist, through many of them would systematically catalog assets in data warehouses, and then would proactively disclose them and make them searchable through their websites in machine readable formats. So that requesters would be able to find them using search engines, using data.gov, which you did reference and that would all be tied together. 

As we've discussed earlier in other previous committees and recommendations, implementation of that act has not fully moved forward because of the lack of guidance from OMB (Office of Management and Budget). It certainly would be useful to hear from you all how much impact that lack of guidance has had and how and where agencies could develop a more holistic approach where requests move from, where they come in from FOIA.gov through case management systems, and then are disclosed as structured data through FOIA libraries, which are in theory are indexed and findable through data.gov, because that's not currently the case.  

So thank you for your presentation. I definitely learned a lot. I'm going to have to go through the white paper next.  

Eliot Wilczek: Thank you.

Alina M. Semo: Okay. Thanks, Alex. Jason?  

Jason R. Baron: Can you hear me?  

Alina M. Semo: Yes. Loud and clear.  

Jason R. Baron: This is an outstanding effort on the part of MITRE. And I appreciate that on page 18 of your 300 plus page report, you said that you were informed by the FOIA Advisory Committee's work in past terms. So I have two things to ask. One is, how do you envision MITRE interacting with federal agencies to implement the recommendations on all these processes? Is MITRE just, I understand that you're working with a small group on business requirements, but if an agency approached MITRE, would you have the time, resources, effort to assist them in figuring out how they could improve their processes using this model? That's my first question. And the second is, are you open to this committee's observations, recommendations, recommendations with respect to an improved model in a 2.0 version? And picking up on Alex's point? There's an information governance perspective to the FOIA efforts model where there are many other statutes, including the Federal Records Act, but many others that come into play when thinking about FOIA. And so it seems to be, that kind of perspective might improve what is already an outstanding product.  

David Bloom: Thank you for those questions.   

Eliot Wilczek: Thank you. Do you want me, so I think, you know, in terms of MITRE working with individual agencies, that is work that MITRE engages in through its six FFRDCs (Federally Funded Research and Development Centers) that it operates. So certainly when you know, MITRE works with individual agencies about its FOIA programs, you know, it would use the reference model to help inform that work and help agencies address the problems that they're addressing. I think, you know in terms of the work going forward and opportunities for improving upon the reference model, I think that's, you know, where there's, that's a conversation to have to, you know, work closely with the OIPs Federal Integrated Business Framework work, to, you know, figure out what's the appropriate vehicle for continuing this line of work around articulating FOIA business rules and business requirements that take in as you, as you mentioned, and are able to triangulate the full range of laws and regulations that touch on federal records and their access and dissemination.  

Alina M. Semo:  Jason, any other questions? 

Jason R. Baron: No, thank you. 

Alina M. Semo: Okay. Any other committee members have any questions for Eliot and David?  

Bobby Talebian: Hi, this is Bobby from OIP/DOJ. I just wanted to thank both David and Eliot and MITRE for all the work that was put into this in helping us really accelerate, really tremendously, our effort with the FIBF. And just tag onto what you just mentioned, we're finalizing right now that work with the CFO Council. And then our next step would be this summer, as you mentioned, publishing those for public comments. So that would be a great opportunity for not just everybody in the public, but the FACA to also provide that type of feedback.  

David Bloom: Thank you for the opportunity to, , to share this with you.  

Eliot Wilczek: Yes, thank you very much.  

Alina M. Semo: Okay. Anyone else seeing any nods or hands being raised or anyone waving frantically at me? Does anyone else have their hand raised, that I missed? Speak up now. Okay. All right. Eliot and David, thank you again so much for coming today. You're welcome to go off camera and hang out with us for a while longer or go back to work, whatever you would prefer. And I know that we will definitely take a closer look at your paper and folks will look at the powerpoint presentation and we'll follow up with any questions we may have. So thank you again.  

Eliot Wilczek: Thank you.  

David Bloom: Thank you.  

Alina M. Semo: Okay, terrific. Okay, so with that, I wanna turn to the next part of our agenda, which are subcommittee reports. I can attest to the fact that the subcommittees have all been extremely busy. Everyone is working very hard. I just want to thank everyone for that. It's very meaningful that everyone is so engaged and it's definitely helping the work of the committee overall, and it will make all the recommendations that much more meaningful. The Modernization Subcommittee I believe is up first according to what Kirsten told me. So I follow whatever she tells me. And I want to turn it over to co-chairs Jason R. Baron and Gorka Garcia-Malene. So, over to you too, gentlemen.  

Gorka Garcia-Malene: Thanks, Alina. Can you hear me?  

Alina M. Semo: Loud and I'm clear.

Gorka Garcia-Malene: Wonderful. Good morning, as Alina just alluded to my name is Gorka Garcia-Malene, I'm the FOIA Officer at the National Institutes of Health. Together with Jason Baron, I co-chair the Advisory Committee's Modernization Subcommittee. And today our subcommittee brings two items for the Advisory Committee’s consideration. 

And the first one is a proposal for a more precise application of Exemption 5 in government redactions. After discussing if the Advisory Committee deems appropriate we might even be in a position to vote on this today, who knows, but at least we'll have a conversation. And the second item is a draft model letter on which we wish to solicit feedback, and not just from our fellow Advisory Committee members, but also from agencies, the FOIA requester community, and the public at large. The Modernization Subcommittee would then take these recommendations and ideas into account to finalize the language of the model letter and then bring it back to this Committee for further consideration and potentially a vote at a later date. I should mention, by the way, that anyone wishing to follow along with our proposals can see the proposal and documentation on the NARA’s Advisory Committee website under today's meeting materials. And the recommendation is labeled Modernization Subcommittee Transmittal May 30th, 2023. And these three documents are attachments A and B on NARA’s website. 

So regarding the first item, Adam Marshall, who is our fellow Advisory Committee member, he couldn't be here today, he's from Reporters Committee for Freedom of the Press, and he's spearheaded an effort to improve how agencies assert Exemption 5. Okay, so I'm going to present this on his behalf. But we've spent, and he's been particularly involved, we've spent many months of delicate crafting to bring a proposal on how agencies assert Exemption 5. And that's for when Exemption 5 is being asserted in a document in part or in full. And so, as we know, the three most common privileges cited in connection with Exemption 5 are deliberate process, attorney-client privilege, and the attorney work product privilege. Unfortunately, requesters rarely know which privilege is being cited to when Exemption 5 is applied. And this is in contrast to other exemptions, right? So, for example, Exemption 7 similarly captures a multitude of basis for withholding information, right? So it has an assumption that has six subparts, A through F, however in those cases, agencies actually cite the specific subpart that's being asserted. So what our subcommittee is proposing here, what you seek to remedy is this disparity, and it's not a simple matter of form or function here. The reality is that without further understanding the basis for a given existing client redaction, a requester has very limited information to evaluate whether it's a challenging redaction. And in many cases, in our shared experience, they do, right? Whereas if they had more information, right, if maybe the exemption the redaction said, you know, this is a connection with the attorney-client privilege, maybe they wouldn't. 

So to resolve this, what we're proposing is that agencies specify particular b5 privilege being invoked in any given redaction, just as we do for Exemption 7. And if the record is being withheld in full, maybe the specific claim can be cited in the determination letter right. Now, before we move forward with our discussion, you know, the Modernization Subcommittee obviously knows that agencies leverage a wide variety of technologies when applying redactions, and that may impact what the, you know, redaction looks like. And, it may limit what agencies can do.

So just generally our proposal seeks agencies to find some way of identifying the precise spaces for Exemption 5 redactions. And for those working, again, to follow along what we're proposing and it's just an idea, right, this can take any number of forms. What we're proposing is on NARA’s website and it might look something like a redaction. The reduction for instance citing deliberative process might read, you know, in the text, in the reduction box, it might read b5-DPP for the deliberative process privilege, right? And one that's based on the attorney client privilege might read b5-ACP, or attorney-client privilege. I hope that provides a sufficient context and supportive discussion now by the Advisory Committee. And of course, Jason and I look forward to any questions this proposal may prompt. Jason, do you wish to add anything?  

Jason R. Baron: No, that was very well said. I do want to thank Michael Heise for being the prime mover of this and Adam Marshall for doing the draft of this recommendation.  

Gorka Garcia-Malene: Excellent point.  Hi, Bobby. I think you have a question.

Bobby Talebian: Yeah, no, I was just going to say, this is, thank you, Bobby from DOJ. I think this is great. I, we, do [think] agencies should be marking the documents so that you know which privileges are applied. I would even go a little bit further, I think you have the three main privileges as an “other” as a marking. I would suggest. And I would think that even if it's, you know, an “other” that you should specifically mark it with that privilege and not so that there's any guessing as to what the other privilege is. So presidential communication privilege is another one that should just be marked as PCP. But I think this is an, this should be an important, it's an important practice that agencies should be able to convey to their requester under Exemption 5 what privileges they're applying.  

Gorka Garcia-Malene: Thank you, Bobby.  

Alina M. Semo: I'm, we've got two hands up. I don't know who was first Michael or Gbemende. So you guys fight it out amongst yourselves. Michael, go first. Okay, go first.  

Michael Heise: I'm sorry. Was that me to go first? 

Gbemende Johnson: Yes. 

Michael Heise: Oh, okay. Thanks. Well, actually I think Bobby covered it. I was going to say that I remember it the first, our first meeting as the FACA that this issue was raised. And, you know, Bobby raised a good point then that, you know, of course b5 encompasses a wide array of different privileges that would exist in civil, you know, the civil discovery sort of litigation context, but there are these big three. And so I appreciate the Modernization Subcommittee having as part of their proposal this other bucket. But I do take Bobby's point that I just heard now, which is, you know, to the extent agencies can give additional color on the other bucket, they should. I think that makes a lot of sense. So anyway I appreciate you, Bobby, for commenting on that part. That's it. Thanks.  

Alina M. Semo: Okay. Gbemende, go ahead.  

Gbemende Johnson: Thank you, yeah, I definitely agree with the previous comments. I think this is excellent. I just had a question in terms of the research for this recommendation. Were there any examples, and you don't necessarily have to name them, of agencies that are already doing this, that could serve as a model for other agencies and bureaus?  

Gorka Garcia-Malene: Well, I can address that. So at NIH we try to provide as much specificity as we can when we're labeling redactions, but it's not always a simple practice. And so I would say that this provides a sort of a very simple approach to providing this sort of clarity, because I think sometimes at NIH we try to provide a little, you know, even more detail, and that just really slows things down. So at a certain point, we just go back to just b5, b5, b5 just to make things a little bit smoother. This is a nice balance, at least from the perspective of NIH, where we're providing enough information so that a requester can weigh whether it's a challenge, you know, appeal, sue, whatever the case may be. Because sometimes the context is not sufficient to understand whether, you know, something is being withheld because it's attorney client privilege, which is rarely challenged, you know, or maybe the more traditional b5, you know, pre-decisional deliberations.  

Alina M. Semo: Thanks, Gbemde. To your point we have a comment from one of our attendees who processes FOIA requests at the Nuclear Regulatory Commission. That person says that they add a note on the document that is seen by the requester citing, which b5 privilege is being used, but agrees that changing it globally would cause more efficiencies. I wanted to share that too. Lauren, please go ahead.  

Lauren Harper: Thank you to the Modernization [Sub]committee for this. I think this is a terrific idea. We get a lot of b5 denials and we rarely see which privilege is being identified. The one thing I would love to see, particularly when it comes to the deliberative process privilege, is with that kind of ping that this is what's being cited, also saying, dear requester, there's a 25-year sunset on this, just so you know… Any kind of additional information like that to help inform the requester that there are kind of different sunsets for the, well, there is a sunset for the deliberative process privilege, I think would be really valuable.  

Gorka Garcia-Malene: And that's, that's a really valuable conference for, for the next part of our recommendation, right? Because we have a proposal for a determination letter, and maybe that's something that could be incorporated there. That's a very good point. Thanks Lauren.  

Alina M. Semo: I saw Catrina’s hand up. Catrina, go ahead please. Catrina, can you unmute or did you? 

Catrina Pavlik-Keenan: No, I finally learned how to unmute. Okay, so the question or the comment that I had was, and I agree with the person who's who mentioned about in the Nuclear Regulatory Commission, we do cite in our letters when we're using b5, what portion of b5 we're using. And what I think you're saying is where we mark b5 on the document, you want us to put something in also by that b5, I don't know, the guide that you had or whatever, how you all had had planned it out. And the question I have is if this is, if this is something that, because some of the, sometimes when we do these things or we talk about these things, you know, and make a recommendation, again, there's no, there's no way that we can, as a Committee try to force anybody to do anything. My question would be, would the Department of Justice then require this as something that would be sort of a requirement under b5 to actually have this done? Or would this just be a best practice kind of thing? Because then I'm not so sure that you would get, you know, sort of get the bang for your buck about what you're trying to achieve. And that's just the comment I had.  

Alina M. Semo: Bobby, any thoughts on that?  

Bobby Talebian: Yeah, so thanks Catrina. So the recommendation is for us to consider issuing guidance. So that's definitely what we'll do. But I do think it's important that the requester know what privileges are being applied, especially if there's multiple privileges being applied so they know what redactions those privileges are. I could see a situation where maybe if only deliberate process is being applied, well then it's clear that all the Exemption 5 in the documents are DPP. But I think it's important. I think it's something that we should be doing. So yeah.  

Catrina Pavlik-Keenan: Thanks Bobby.  

Bobby Talebian: Absolutely.  

Alina M. Semo: Catrina, did you have anything else? Your hand is still raised. I just want, okay, thank you. All right. I'm looking around. Anyone else have any other questions on this part of the Modernization Subcommittee's presentation. Going once. Okay, Jason, or Gorka I'll [send it] back over to you for the next issue.  

Jason R. Baron: Well, Alina, is it timely to take a vote?  

Alina M. Semo: It is entirely up to you, the subcommittee co-chairs in terms of how you want it to proceed. Would you like for us to go over the voting procedures if you believe that it is time to take a vote? I'm fine either way. We could do it today or we can do it at our next meeting.  

Jason R. Baron: I would say we strike while the iron is hot.  

Alina M. Semo: Okay. Kirsten, would you like to go over the voting procedures to remind everyone as the DFO or do you want me to do it?  

Kirsten Mitchell: Sure, I can do it. Happy to do it. So voting procedures, any member of the Committee can move to vote on the recommendation. Although the motion does not need to be seconded let's continue our tradition of seconding the motion as per Alina. There are three ways a motion can pass. It can pass by unanimous decision, which is when every voting member, except abstentions, is in, in favor of or opposed to a particular motion; there’s general consensus, which is when at least two thirds of the total votes cast are in favor of or opposed to a particular motion - I've done all the math ahead of time, so if it comes to that, I will not have to spend too much time doing math. And finally, general majority, which is when a majority of the total votes cast are in favor of or are opposed to a particular motion.  

Alina M. Semo: Okay. Terrific. Anyone have any questions? Tom, you have your hand up, please go ahead.  

Tom Susman: My hand is up to move the motion be adopted by the committee.  

Alina M. Semo: Okay. Thank you, Tom. Do I have a second? 

Allyson Deitrick: I second. 

Alina M. Semo: Okay, thank you Alison. Okay, let's please take a vote. All those in favor of this recommendation please say aye.  

Multiple: Aye. Aye. Aye. Aye. Aye.  

Alina M. Semo: Okay, anyone opposed? Please say, nay. Not hearing any nays. Anyone wishes to abstain?  

Bobby Talebian: This is Bobby from OIP, consistent with how I've voted before and particularly since this is an OIP recommendation. I'll abstain.  

Alina M. Semo: Okay. So noted. Kirsten, are you good on the vote count?  

Kirsten Mitchell: I am good on the vote count. It sounds like it is 15 to zero, with Bobby abstaining. Alina, you're voting…abstaining?  

Alina M. Semo: I'm a yay.  

Kirsten Mitchell: A yay. Okay. You're a yay. Okay. 15 to zero, then the motion carries.  

Alina M. Semo: Okay, great. Great work everyone. Our first recommendation to pass– everyone should be very proud. Thank you for that. Okay, let's move on to the next item. Jason, is that you or is that Gorka?  

Jason R. Baron: I'll take it. And we actually have two items. And the proposed model agency determination letter, which Gorka mentioned is at attachment B in the materials available publicly on the website is a proposal that was drafted by Adam Marshall for us on the subcommittee. And I want to thank Adam particularly in absentia. And the proposed determination letter is not being proposed for a vote today. As Gorka said, we have put up a proposed letter as a model for comment. We can talk about it today among the committee but we really would wish to have the FOIA community, public at large, [and] agencies to have time to review it and so that we can as a subcommittee and as a committee receive feedback and then come back with a tweaked model letter for a vote at the next public meeting.  

So the proposed letter, I don't intend to go through it in detail, there is a current variation in determination letters that agencies send. MITRE talked about inconsistency in release as part of their presentation this morning, and that is certainly the case with determination letters. Many agencies do a very good job of setting out what are the elements in this model determination letter. Other agencies supply what I have found as a FOIA requester, you know, less than complete and transparent responses on all the points that are raised in this letter. So this is really a best practices proposed model. So, what does it cover? Well, um, the issues that are foremost in the letter, how searches are conducted, what to say when there's a complete denial, a full withholding of documents, and, other matters that do an explanatory job of walking through what a requester needs to know, including on their rights for appeal and whether any referrals have been made by an agency, fee, waivers and other matters.  

But on the particular matter of search and explanation, one of the keys to this model letter is that there would be a better explanation of what kind of search was conducted to locate responsive records, including a search methodology, what keywords were used if there was an electronic search where the agency searched in terms of locations or repositories that were searched. And as a result of these searches, what was the volume of responsive records that were located? And after completing review, what is the volume or other descriptions as to how many of those records were responsive? In terms of explanations the model letter goes through a framework for explaining partial grants or partial denials and full denials. And in each case, we give examples of what a response should consist of.  

So, for example, we have an example here of where an agency might say, pursuant to Exception 5, and the deliberative process privilege noted as b5 DPP, this agency is partially withheld 91 pages consisting of the content portions of 36 internal emails with attachments regarding recommendations on a new agency policy. And then it would also include in the determination letter that the agency reasonably foresees that disclosure would harm an interest protected by the deliberate process privilege because with a concrete explanation of that. Those kinds of descriptions are not in my experience, and I leave it to others to judge in your experience, the kind of information that is received in the first determination letter that a requester gets. And so with that, I'm going to stop. You can all read the model letter. Our proposal is that everyone in the public and agencies have 30 days to give us comments on this letter. I defer to Alina whether 30 days means July 8th, which is calendar days, or 30 days means business days, which would be July 20th. But that's the proposal of the Subcommittee for a timeframe for comment. And then we would look at the comments and come back with a revised letter. So now I open it up, Gorka, if you wanna say something, and then committee members.  

Gorka Garcia-Malene: You've captured it perfectly. Thanks, Jason.  

Alina M. Semo: Business days seems consistent with FOIA. So how about July 20th? Does that sound fair to everyone? For comments. Okay. Anyone have any questions or comments on, oh, Michael's hand is up. Michael, go ahead please.  

Michael Heise: Thank you. So I think the only thing I would say with this letter, I think it's clear, but just in case, I just wanted to note, and I think this is fine with everybody, well, who knows - is that in your rights section - I think it's clear that should apply whether the agency is doing a full grant or a full denial or a partial grant. But, I just, you know, want, you know, respectfully to all the various people in government on this call or whatever, that you know it to the extent that it may not be clear, right, because some people on the agency side might say, hey, we're granting you in full. I mean, you know, what's your deal, you know, we're granting it in full - it's up to the requester, right, to determine whether or not that, you know, they're full grant. So I think that appeal rights, well, I don't distinguish this, I think it's the case that appeal rights should attach and the language should be present in the letter even when there's a full grant. And to the extent that this letter makes that clear and to the extent it matters if it's clear or not, that would be the only thing I would add on to this. Because I, I do personally think it's important that the requester community have notice that they have the right to appeal even a full grant because, and I think it's be good for the agencies too, because if we don't do that, I think technically, you know, they don't have an ability to administrative remedies so they can go to court and that wouldn't be the greatest thing for the agencies, also. So I would just, you know, I think it's clear, but just in case I'm putting it on the record now, I think, and if anyone has an objection to that, it might be interesting to hear.  

Alina M. Semo: Okay, thanks, Michael. I certainly don't have an objection, and I think that's consistent with the guidance that OIP has been giving out for years. But Bobby, I'll let you speak to that.  

Bobby Talebian: Yeah, no, that's fine. I would just say more generally my reaction to the template, and I appreciate, I think it's a great idea to open it up and just have more detailed feedback that you can get that we can't do in this forum and expanding it beyond just everyone on the committee, I think for sure consider our guidance to provide as much context to the requester that's practically feasible in the initial request letter. And that's more practical in some cases than others. For example, if it's like a, you know, one single type of document or as opposed to many redactions over different types of many, many different types of records, I think there's definitely probably room for improvement in describing things. I would say probably more generally, so that requesters know, you know, generally how an agency approaches, how did they process, conduct a search. I like the suggestion Lauren had made about including the information that there is a sunset privilege for deliberate process and those types of things. But I think there needs to be a practical balance because the paper makes it seem that this would cause a lot of efficiency with less appeals. But I would argue that a level of detail, which this is kind of approaching almost like a mini declaration or a Vaughn [Index], that would slow down the process and would not be maybe practical. So I think there's a balance there where we want to make sure we're giving as much information as we can without slowing the process. And I think that's something maybe to consider just more generally.  

Alina M. Semo: Thanks, Bobby. Luke, I think your hand up was first and then Lauren and then Kirsten. Hopefully I got the order correct. Luke, go ahead please.   

Luke Nichter: I’ll go very quickly. I was just going to second Michael's suggestion to provide the appeal information for a full grant for a different reason. As a requester, oftentimes, you know, it's happened a number of times where I'll get a full grant, but it's because the agency interpreted the scope of my request a certain way, perhaps more narrow or just somehow different than what my intention was. And so I can see for another reason, you know, I have had to appeal, you know, sort of a full request, even though it might not seem like that happens very often. So just wanted to second that suggestion. Thank you.  

Alina M. Semo: Thank you. Lauren, please go ahead.  

Lauren Harper: Thank you all for working on this letter. I went through it with great interest. I have a comment that I'm also happy to submit in writing when we open it to public comment. The section where it specifically talks about the following locations or repositories, researched as they were determined to be most likely to have responsive records, I think as a requester, something like that is most useful when the requester is aware kind of the universe of repositories or records databases that are searchable. So I think a parallel effort that might be outside the bounds of this particular letter would be examining if we could recommend that agencies on their FOIA websites list things like their organizational charts or their major record systems or repositories and things like that to really help requesters determine like, gosh, do I think that that was also the best place to search for those records? So that's my major comment, but I thought this was a terrific effort and I look forward to submitting some comments in writing as well.  

Alina M. Semo: Okay. Perfect segue to Kirsten, who was going to just inform everyone again about the way to submit public comments.  

Kirsten Mitchell: Great. Yes, thank you, Alina. I put in the chat to everyone a link to our public comments portal. So you can just go to that site, you can see the dropdown to FOIA Advisory Committee, and then you can submit your comments. And as Jason said, we'll be accepting those and sharing them both with the committee and publicly. And OGIS will also be tweeting and blogging about it.  

Alina M. Semo: So… 

Kirsten Mitchell: And for those, if you are a  member of the public, if you have some comments, please submit them.  

Alina M. Semo: And for those of you who can't see the Webex chat, it's www.archives.gov/ogis/public-comments. So thank you for all of that. I'm looking around to see if anyone else has any questions on this item. Don't see any hands, I don't see anyone leaning forward eagerly. Jason, back to you. You said you had a third item to present?  

Jason R. Baron: Yes. And I'll do it quickly. Another proposal from our Subcommittee is that coming from the FOIA Advisory Committee and whoever wishes to sign it would be a letter about FOIA.gov interoperability [and] the deadline coming up in the sunset of FOIAonline. And what we have proposed is that a letter be sent to agencies that highlights best practices in the implementation of the FOIA Improvement Act of 2016, which mandated that all agencies accept FOIA requests through FOIA.gov and configure their case management systems to be interoperable with it by August 2023, as outlined in an OMB memo in 2019, and particularly to apprise Chief FOIA Officers of a recent initiative taken by the Department of Homeland Security (DHS) to proactively engage stakeholders and the public regarding its transition from FOIAonline to a new case management system and modernize its system for first party requests as part of a renewed commitment to open government. So the letter points to DHS as a model for doing a FAQ, and we would hope that Alina and others would consider this proposal and determine what is the best form for sending a letter out rather expeditiously given the August deadline with respect to FOIA.gov and the sunset of FOIAonline.  

Alina M. Semo: Okay. Thank you so much. Any other questions or comments on that third item?  

Alex Howard: I would just share what Jason's concern or about the expeditious nature of this. I think as I've expressed publicly and privately there are ongoing concerns and considerations associated with the sunset of FOIAonline. I've heard directly from OIP that the interoperability question of FOIA.gov appears to be fairly well addressed in terms of agencies moving forward and making sure that their case management systems are in fact interoperable with it as mandated by OMB guidance in 2019. As we talked about in the past, I think it would be very helpful if OMB would come and talk to the committee about how they're helping to administer the FOIA and the steps they're taking to modernize or support that effort as mandated in the FOIA Improvement Act 2016.  

We had a terrific discussion because of the efforts of OGIS to get EPA (Environmental Protection Agency) staff to come and talk with us about the decisions behind sunsetting FOIAonline. But I have not seen a shift from them or other agencies with respect to public engagement, requester engagement, regarding that shift. And I think it would be extremely meaningful if the FOIA Council could make an ongoing effort to vastly increase public engagement around this shift to make sure that there are no requests that are lost, and that requesters who are using that system are not surprised by it. As people who visit FOIAonline know, and if they see it right now, there is a guidance that they need to download the records. But it is my, I think, ongoing concern that we might see some loss in upholding their right to know and their right to get responsive records. And the extent to which the whole US government could be using its considerable communications resources, I have not seen it here in terms of a change. So I hope that you and Bobby will consider publicly and privately messaging agencies to step up requester engagement so that everybody knows what's happening and there are no surprises which could further diminish public trust in government at a time when I don't think anyone can afford that.  

Alina M. Semo: Okay. Thank you, Alex. Appreciate that. Okay, Jason and Gorka, that completes your presentation for today, correct?  

Jason R. Baron: It does. I wish to…it will come as no surprise that I wish to acknowledge Alex as someone who really pushed our Subcommittee to make this proposal.  

Alina M. Semo: Okay. Gorka, I'm sorry, did we talk over you? Sorry about that. 

Gorka Garcia-Malene: That's what we have for y'all today. 

Alina M. Semo: Okay, terrific. So I wanna move on to our Resources Subcommittee. Gbemende is presenting today. Paul is not able to be with us today. So over to you, Professor Johnson.  

Gbemende Johnson: Thank you, Alina. So we have a number of items to provide updates on. So at the last meeting, Paul discussed the interviews that the Resource Subcommittee was undertaking with high level FOIA officials. And these interviews focus on questions about technology, training, financial and staff resources. We've conducted, to date, perhaps close to two dozen interviews. And to reiterate from Paul's previous comments, we're trying to focus on a mix of agencies in terms of size jurisdiction, for example, agencies that may be more likely to handle classified documents and also variation in regards to the manner in which the agency FOIA offices are organized. We expect to wrap up these interviews prior to the September FOIA Advisory [Committee] meeting. Because the interviews are still ongoing, we don't want to provide a lot of detail yet in terms of the responses as to, we don't want to contaminate the interviewee pool, but we will say something that has come up multiple times, which should not be a surprise as the importance of retaining a qualified FOIA staff.  

So in relation to the interviews, we are also trying to get the perspective of FOIA professionals through the administration of a survey. This survey will be administered at the American Society of Access Professionals, the ASAP National Training meeting in New Orleans, June 27th through the 29th. The target of this survey are FOIA professionals on the ground who are working and processing FOIA day-to-day. So this includes processing requests, responding to public inquiries, and also dealing with appeals. This survey contains 25 questions and takes approximately 10 to 15 minutes to take based on the test run that we've conducted. Most of the questions are adapted from the questions that we developed in the interviews, but we also asked questions about how much time individuals are spending on litigation, how do they feel about the training they've received and whether they've considered, you know, leaving or staying in their position. So this is to get at these concerns about staff retention. And, luckily I haven't mentioned this to you, yet Alina, UGA (University of Georgia) actually approved the survey. It went through the IRB (Internal Review Board) so it's good to go in terms of fielding it at the ASAP meeting. And of course, the goal here is, while the Resource Subcommittee is working on a number of issues that we think are important, we want to make sure to get feedback from those who are, you know, in the trenches doing the work, regarding what their key concerns are. 

So in addition to the interviews and the survey, another issue that the committee has discussed in multiple meetings is the issue of a FOIA repository and this idea of a repository was put forth in a previous recommendation, I believe, the 2018-2020 term. But we felt that this issue of a repository became salient once more with the sunsetting of FOIAonline, which of course allowed users to search for documents that have been released from various agencies subsequent to previous requests. So there are, as we know, FOIA e-reading rooms, but the repository could make the search process, could potentially make it easier for requesters and allow them to have a wide variety of documents in one portal. So in our examination of the FOIA repository, we spoke to multiple officials from EPA, OIP, the Consumer Finance Protection Bureau, and also other members of the data working group of the Chief FOIA Officers Council’s Technology Committee, who were also exploring this FOIA repository conversation. And these conversations have been very informative. We've discussed the benefits, the downsides of the potential repository, the challenges such as who would own it, the development and maintenance of technology and what's necessary to allow for an efficient search if such a repository were to exist.  

And through these conversations, we actually had a detailed discussion about the FOIA wizard tool, which is in development right now by OIP and DOJ, which isn't a repository, but of course has the goal of allowing users to search for information that's in the public domain. So this discussion is ongoing, but it's definitely been complex and layered and very informative and there's more. But I guess I'll touch on one other thing, which is, I think, kind of a big chunk, and that is the issue of budgeting. It's hard to discuss FOIA without discussing funding. And something that comes up a lot in discussions of FOIA is the fact that for most agencies FOIA does not have line item status funding. And the concern here that some have raised is, you know, FOIA has to compete with other agency priorities for adequate funding. But the question we are attempting to explore, and this is early on, is to what degree would a line item status for FOIA improve the allocation of FOIA administration implementation? Would FOIA officials find this beneficial? Would there be some challenges? How would the implementation benefit? To paraphrase a comment, a comment raised by Lauren Harper a while back, you know, would line item status change the culture around the way in which FOIA is perceived and processed as a priority? 

So these are questions that, you know, we don't necessarily have clear answers to right now, but we are working on understanding more about the ways in which various agencies structure their FOIA funding. So we reached out to various agency officials who are involved with the budgeting specifically to get their perspective on this issue. And we hope to have in the near future a discussion with OMB about, a general budgeting discussion, kind of taking a step back and, you know, talk to us about the budgeting process and various ways in which to approach it. So I think I will stop there,there's more, but I think that was enough. So thank you.  

Alina M. Semo: Okay. Anyone else on the Resources Subcommittee, wish to add anything to the great presentation we just heard,  

Lauren Harper: Gbemende did a great job. I don't have anything to add.  

Alina M. Semo: Thanks, Lauren. Any questions for the Resources Subcommittee or comments? Not seeing any hands up. Okay. Well, thank you so much for that presentation. So last but not least, we have our Implementation Subcommittee. We have co-chairs, Dave Cuillier and Catrina Pavlik-Keenan. David, I'm not sure who's presenting it's you or Catrina, but I'll turn it over to both of you and take it from here.  

David Cuillier: Thanks, Alina. I'm Dave Cuillier soon to be University of Florida. I have a one page, we have a one-page summary posted on the committee website. So I don't really need to go into it for time sake, but basically we're gleaning through those past 51 recommendations and identifying which ones still need more work. We have a lot more information gathering to do this summer. Likely a survey, interviews with agencies, gleaning data through, Chief FOIA Officer reports and annual FOIA reports from agencies. And by September, I think we'll have a pretty good idea, we'll be continuing the research in the fall, but we should be able to provide for the committee, at least an initial list of where the recommendations stand and where we're headed. So that's kind of where we are. And thank you, the past three months to OGIS staff, Alina and Kirsten, as well as, Bobby [at] DOJ, very helpful in helping us take that first pass to figure out where things stand on these recommendations, past recommendations. So thanks to everyone and great work by the Subcommittee. Catrina, did you want to add anything else? Or any other subcommittee members?  

Catrina Pavlik-Keenan: No, I just think everybody's done a great job and working really hard on this, so I wanna thank everybody on the Committee.  

David Cuillier: Great. Well, thank you all.  

Alina M. Semo: Okay, any other questions from the rest of the Committee for the Implementation Subcommittee? I'm not hearing anything or seeing any hands up. Just looking one more time. Okay, David, thank you. And by September you'll be in Florida? Yes.  

David Cuillier: Yeah, by this time next week I'll be there enjoying the balmy weather.  

Alina M. Semo: Yeah  

Lauren Harper: Tucson will miss you.  

Alina M. Semo: Tucson will definitely miss you. Okay. So we're done with the subcommittee reports. I just want to pause here to ask if anyone has any other follow up questions or comments or any discussion issues they wanted to bring up, generally, among the committee members. We're actually ahead of schedule today, but I'm happy to give time back to everyone. Jason has his hand up. Jason Baron please go ahead.  

Jason R. Baron: Thank you, Alina. I'm trying to lower my hand, there go. So we heard from MITRE today, and there are two points I want to raise. One is that we saw that there is a 300-page document as a framework that they have developed. I would be very interested in hearing from the government representatives on this Committee, whether they are able to take that document and talk to people at their agencies about how agencies could evaluate it and implement it, aside from any kind of business requirement process that may take months or years for MITRE to work through. And related to that, there was a reference made by David and Eliot to a separate sister project...I've had the privilege of working with a different team at MITRE headed by Karl Branting, on research on Exemption 5 in particular and to some extent Exemptions 6 and 4, to develop a FOIA assistant tool that they have a prototype for that they have been discussing with various agencies in terms of use. I would also be interested in whether the government representatives on this committee would be willing to engage with MITRE to explore the use of this tool in their own FOIA workflows.  

Catrina Pavlik-Keenan: So just in general, just speaking this Catrina, we work closely actually with MITRE on a few things. And we have in the past that some of the components at DHS worked with MITRE on, modernizing the FOIA, modernizing and updating the workflows in FOIA. So, you know, we are, we do utilize FFRDCs for that type of work within DHS in multiple different components. So we are looking into things like that. So this is something that we're familiar with. I think that's about all I can say.  

Alina M. Semo: Michael can, I'm sorry, I didn't want to put Michael on the spot, but, maybe Michael could speak to any progress on trying out that AI tool. But I'm sorry, I heard Gorka speak first. So, Gorka,  

Gorka Garcia-Malene: I was going to add, I actually reached out to MITRE a couple years ago when they started their b5 effort. And my intention was to collaborate with them, but the pandemic has brought such an onslaught of requests that my shop is entirely too busy to contemplate this right now. I'd love to work with them. I just have zero bandwidth right now, but I look forward to doing that in the future.  

Patricia Weth: Hi, this is Patricia Weth with EPA. Joan Moumbleaux in my office, is trying to get the approval to work with MITRE to test out the FOIA assistance tool that Jason mentioned. We're in the process, I've gotten clearance through ethics, but now, we just have to  work out on an agreement. And unfortunately, right now my office is in the throes of implementing our new FOIA case management system. So, I don't think we'll be able to tackle this until probably in the new fiscal year, but we would very much like to help and we're really, really interested. This FOIA assistance tool, I think can be a game changer in the FOIA world. I think it could save a lot of time and energy and you know, I'm on the Resources Subcommittee and one of the things we always have to look at is, you know, how can we work smarter? And this, this is an opportunity to do so.  

Bobby Talebian: This is, this is … 

Alina M. Semo: Bobby. Go ahead, please.  

Bobby Talebian: I was just going to say, and you may have been going to say the same thing, but this has been socialized with the CFO Council, and there are a number of agencies, including actually our, my office that are trying to get to the point where we can work with the tool. But we certainly are very interested in it and want to make sure that MITRE’s good work is put to good use.  

Michael Heise: This is Michael, EEOC (Equal Employment Opportunity Commission). So, to Alina’s point, we've had, I can say that we've had a couple meetings with MITRE where they've introduced the product. And it's moving through internally here. And so,  I don't know exactly the timeline or anything like that, but we're [a] relatively small agency, so, you know, I think that, I'm looking forward to it moving through the process and reaching out to MITRE in the not too distant future.  

Alina M. Semo: Okay. Thanks Michael. And thanks for letting me put you on the spot. I appreciate that. Okay, any other comments from anyone else from the government side in response to Jason's question? Or anyone else who wants to comment? Okay, I don't hear or see anyone else. Jason, do you have any other issues that you wanna bring up?  

Jason R. Baron: I appreciate all the comments and I think everyone knows that I am an evangelist on the subject of machine learning and AI. And so, we will have more to report on in the Modernization Subcommittee as time goes forward on those subjects.  

Alina M. Semo: Okay. We look forward to hearing all of that. Thank you. Okay. Anyone else have other topics that they wanted to bring up today? Alex? His hand is up. Alex, please go ahead.  

Alex Howard: I would love to propose that by the time we come back in September, which is our next public meeting, when we'll have seen what's happened with FOIA.gov agency case management and FOIAonline, that we do something of an after action report. And I would like to propose that, OGIS, go back and look at agency websites and reading rooms, in the interim, as I will be doing as well, to see how many of them have acted in response to the excellent work that OGIS did in seeing how many of them have active reading rooms, how many of them list FOIA.gov on those pages, and how many of them are complying with the mandate of Section 508 with respects to their disclosures and are structuring and publishing data on an open government webpage. I'd also like to propose that we come back and look at how many of them have revived their slash open pages for those agencies as well.  

As people know if they attended our last meeting, the United States has a new National Action Plan for Open Government, of which FOIA plays a role. And Bobby has talked about the great work that OIP is doing to work on its commitments. There's going to be a dialogue over the next three months in which US civil society members are going to encourage the US government to add additional commitments to that plan in advance of the Open Government Partnership Summit in Estonia. And given that the US government is now in the steering committee of the Open Government Partnership again, it's the hope of many people in the civil society that that will result in marked improvements to the administration of the FOIA. It's a particular interest as we've discussed in our subcommittee that a component of DHS might be investing in modernization which is something that I think the Committee has put forth great recommendations before. 

Secretary [Alejandro] Mayorkas asked his advisory committee to put forth recommendations in open government believing that it's central to improving the mission of DHS. And one of the recommendations was to modernize the USCIS system, which many people may know, people use to get records, first party records under the FOIA. And my hope is that perhaps we could have a discussion about that then and to see what prospects are for a similar approach in terms of agencies reissuing open government plans that make a holistic combination of transparency, accountability, participation and collaboration to improve the relationship between the requester community and agencies. Which I think it's fair to say is not as good as that [it] could be. And I say that as a representative of the requester community. I'm very grateful to all the members of the Committee who continue to participate in good faith in these discussions and hope that when we come back we can look and see how things are going and perhaps encourage other agencies to take similar steps to re-engage in open government and to form advisory committees that include civil society like this one so that we can make progress in some of the ways that perhaps I was hindered over the last decade.  

So that's what I'd offer in that. Thank you for that opportunity to share.  

Alina M. Semo: Okay. Thanks Alex. Kirsten, do you want to respond to Alex's question or gauntlet that he has thrown down to OGIS? In your compliance team lead hat?  

Kirsten Mitchell: Sure. So, sure. Thank you, Alex, for that thought. And I can say that we do plan to follow up at some point on our website's report. It's something I think we were discussing just yesterday, so stay tuned. I don't know, but I can’t, we can’t promise it'll be done by the next meeting, but it is something that we will look at again. So thanks.  

Alex Howard: Thank you. If that's something that the requester community can help with, that's the sort of crowdsourced engagement that might be a useful place to start. If OGIS would be willing to share open data of the agencies and the places that it looked at, that might provide a useful starting place or a map of the work that you've already done helping us all to build on it, because it was terrific and really useful work. And I don't want to misrepresent whether agencies have actually acted since you did that work, but it would be really helpful if we could arrive together in a collective understanding of which agencies have linked to FOIA.gov, which ones are accepting requests as mandated under law seven years later, which have active FOIA reading rooms, and how and where their engagement with the requester community is healthy or not as evidenced by that change. And again, thank you for your collaboration on this. I think the work of the ombuds office is invaluable in this area.  

Kirsten Mitchell: Thank you. Two just real quick things with that assessment, which we published, I believe in November of 2022. I believe we published a spreadsheet of our findings and where we went. So you can, it's out there, you can see what sites we visited. There's nothing secretive about it. And there, you know, there's nothing sort of behind the scenes on it. 

The second thing, and I don't want to put Bobby on the spot, but I know that OIP is looking at this whole FOIA.gov interoperability issue. And that's something that we can expect more from from OIP.  

Alex Howard: Got it. I just found that spreadsheet in the post. Thank you.  

Alina M. Semo: I see Lauren's hand up. Lauren, go ahead, please.  

Lauren Harper: I was just going to follow up on Alex's comment and say the archive, National Security Archive, would be especially happy to help any effort if civil society input was required. The FOIA.gov issue, and even agencies just mentioning [it] on the website was the subject of our latest FOIA audit for Sunshine Week. So if help is required or would be appreciated, please, please do let us know.  

Alina M. Semo: Okay, thank you. Bobby, you don't have to respond, but if you have anything you want…  

Bobby Talebian: I've mentioned before, we asked about interoperability and the CFO reports. Our summary and assessment, which will be issued in the upcoming months, will score agencies that are compliant with the guidance and note which ones are on track or have an exception that will be expiring soon. So we'll have that out soon.  

Alex Howard: Thank you, Bobby. Have you evaluated which agencies don't accept requests from FOIA.gov? I know that members of civil society have pointed out to me that the CIA (Central Intelligence Agency) does not.  

Bobby Talebian: Yeah. So that is interoperability. So in order to be interoperable within the guidance, just there's still some time they would be, the agency would be accepting requests going through FOIA.gov to their system. Intel agencies early on, that was the more challenging ones for us to implement interoperability. But we've made a lot of progress and advancements and are helping those agencies get on board. I don't have the numbers, but I think it was 110 agencies right now are interoperable. And then we have a small handful, some just decentralized agencies that some of their components have not yet become interoperable or have the exception like the CIA. But we're working directly with all of them to be able to help them meet the interoperability standards.  

Alex Howard: For the purposes of public understanding. What are the consequences of agencies not complying with that law?  

Bobby Talebian: Well, I don't have a gun that I can point to agencies. So we'll continue to work with them to be able to make it. Obviously, agencies have responded. Over 110 agencies are already compliant within the timeframe that they were supposed to be compliant. We at DOJ and at OMB are working to make sure agencies are meeting the standard. I don't anticipate that, you know we have significant issues especially with the progress that we've made.  

Alex Howard: Okay, thank you. Perhaps we can discuss whether DOJ needs more statutory authority for enforcement or of compliance in the future.  

Bobby Talebian: I'm always happy to discuss anything related to FOIA in my office. And like I said, we have a really great working relationship with agencies, and I think that's indicative of the response we've had with getting agencies interoperable. It was not an easy task. It never is with a big change like that and implementing new technology, but I, I think, you know we've not had problems working with agencies.  

Alex Howard: Thank you.  

Alina M. Semo: Okay. We're right up on our 12:15 mark, actually 12:16. If no one else has any other comments that we want to discuss as a committee, I propose that we move forward to our public comments section of our meeting. Does that sound good to everyone? Yes, some nods. Thank you for nodding. That means you're still with us. That's great. 

Okay, so we have now reached the public comments part of our committee meeting. We look forward to hearing from any non committee participants who have ideas or comments to share particularly about the topics that we've discussed today. All oral comments are captured in the transcript of the meeting, which we will post as soon as it is available. Oral comments are also captured in the NARA YouTube recording and are available on the NARA YouTube channel. Just a reminder, public comments are limited to three minutes per person. So with that, I'm going to turn over to Kirsten first and check in with her to see if we have any questions or brief comments that have come in via chat throughout our meeting.  

Kirsten Mitchell: Thanks Alina. This is Kirsten. We had a couple of comments from committee members, Alex Howard and Lauren Harper, but both, all of those have been read into the record. I think this was noted earlier in the meeting, there was a FOIA professional from the Nuclear Regulatory Commission who talked about the b5 issue. And they note, he or she says, I added a note on the document that is seen by the requester citing which privilege is being used, but changing it globally would cause some inefficiencies. That is all I have. I don't know if Dan has anything that I may have missed.  

Alina M. Semo: Can I just correct that comment? I think the, the commenter said it would cause greater efficiencies.

Kirsten Mitchell: Cause more, yes, it would cause greater efficiencies. I just can't see that. Thank you, Alia. 

Alina M. Semo: Yeah, I just wanted to clarify that. 

Kirsten Mitchell: Yeah. Thank you.  

Alina M. Semo: Sure. Dan, anything on your end?  

Dan Levenson: I have nothing to add. Okay. This covered it all.  

Alina M. Semo: Thank you so much. So Michelle, can I turn it over to you? If you want to provide instructions to any of our listeners today for how to make a comment via telephone, could you please do that right now? That would be great.  

[Michelle] Producer: All right. Absolutely. So ladies and gentlemen, as we enter the public comment session, please limit your comments to three minutes. Once your three minutes expires we will mute your line and move on to the next commenter. If you are joined via, phone/audio pressing pound two, we'll enter you into the comment queue. Once again, each individual will be limited to three minutes.  

Alina M. Semo: So, Michelle, do we have anyone, everyone wants to comment?   

[Michelle] Producer: Yes. It looks like we do have someone. I'm going to unmute Gretchen, You are unmuted, Gretchen, you may go ahead.  

Gretchen Gehrke: Oh, thank you all so much for the work you are doing and for the opportunity to speak. And my apologies that I'm speaking through laryngitis. And so I apologize if it's hard to hear. My name is Gretchen Gehrke and I'm a co-founder of the Environmental Data and Governance Initiative or EDGI. And my comment is regarding the decommissioning of FOIAonline. 

I'd like to first thank Alex Howard for his comments earlier in this meeting about the lack of public communication about this decision, because that's exactly why I am here today because there have not been opportunities for public comment or input in this far-reaching decision and what will come from it. My organization is deeply concerned about what this implies about the federal government's commitment to transparency. Although I will say, obviously you all are doing amazing work as well. So I'm here to urge you as the official voice of the requester community and an advocate for FOIA innovation to ensure that agencies do not retreat in their transparency, but actually utilize this as a moment when they're investing in new FOIA portals. And as this whole FOIA business model is being developed to utilize this as an opportunity to invest in more efficient and more effective dissemination of public records. And EDGI would be so glad to work with you on drafting any guidances for agencies. 

We have three central and honestly, very simple recommendations. First, to make all completed requests public. This includes all requests that have been made public through FOIAonline over the last decade, and all requests to come but truly make good on the Obama-era ideal that a release to one is a release to all because the efficiencies that can be gleaned by both requesters and agencies are obviously significant. 

Second, make the records more accessible. Requiring people to download enormous and obscurely named PDFs to have a sense of what is held within them not only imposes an accessibility hurdle that could disproportionately impact lower income individuals and small organizations like mine, where people have older voices or mobile devices. It also makes extensive research of previously disseminated FOIAs impractical. And that seriously reduces the potential positive impact the public availability of records will have if they're not more accessible. We suggest creating ZIP files of smaller PDFs, adding a naming convention and a legend, but a very simple legend about what is held within the different documents. We also recommend developing a browser-based viewing option so that people don't have to download large files in order to have a sense of what is within them. 

Third, enhance the search functionality. Several of the search features of the search function in FOIAonline don't appear to work. And so many more could be added. I'm really excited about the concept of the FOIA wizard that I've heard about, but I'll say this, it doesn't need to be done by AI. It doesn't have to wait for that. Provide opportunities for requesters to supply the needed metadata for searches. Ask them for the key words, the statutes, the offices. Improving the search functionality could transform the utility of public posting of records.  

[Michelle] Producer: Gretchen, thank you very much for your comment. Your time has expired. 

Gretchen Gehrke: Thank you. 

[Michelle] Producer: Thank you very much.  

Alina M. Semo: I just,  

[Michelle] Producer: We are going to…  

Alina M. Semo: Thank you, Michelle, are... 

[Michelle] Producer: Are we ready for the next commenter?  

Alina M. Semo: Yes. I was just going to add that Gretchen has also submitted written public comments, which we have posted on our website and Kirsten has provided a link. Alex, you have your hand up. Did you want to address anything Gretchen just said?  

Alex Howard: I do. Thank you very much Gretchen for your comments on this count. I had hoped that we would see more change from the EPA on this count after they came and spoke to the committee in terms of their public engagement, the use of social media, or indeed hosting forums that are public with requesters, evidence that they had engaged the requester community directly. I would say that they did come here to the committee in a public meeting and were accountable for the decision and shared quite a bit of their thinking about this with the committee in terms of the cost of this. I would also say that your comments with respect to a release, to one release to all policy for FOIA is something that was advanced and discussed by the Committee and has been recommended in the past. And that the Department of Justice repeatedly pushed back on for years when I came to the Committee. And that I hope that we all will make a recommendation that they in fact do adopt such a policy now. I strongly agree with you that records that are disclosed to one person should be disclosed to everybody through agency websites and with FOIAonline going away, that this is an opportunity for every agency to invest in not just a FOIA library, but an ongoing structured disclosure of information as open government data, as is mandated under the law. And that the holdup for that is clear leadership from the Office of Management and Budget working with OIP to make sure that that is an ongoing concern. So thank you very much for raising your voice in this. I hope that your voice improves as you beat laryngitis in the future.  

Alina M. Semo: Thanks Alex. Michelle tells me we have another caller on the line, but Kirsten is alerting me to the fact that Michael's hand is up. Michael, did you want to comment in response to what Alex just said?  

Michael Heise: Yeah, I feel like I need to, just on behalf of the Commission here, Equal Employment Opportunity Commission. You know the vast majority of our 13,000, 14,000 requests, FOIA requests, are for charge files. And it just will not happen that a release to one will be a release to all, I'm sorry, it won't end. Why won't that be? Because of Title VII confidentiality. It's just literally against the law for the Commission to do that. So I just say that, to say that, you know, the federal government is not necessarily a monolith. There's a lot of different agencies with a lot of different workflows, a lot of different equities, and a lot of different kind of documents. And I would hope that, that resonates with both the requester community and with, you know, within the federal agencies. Because like I said, at least as far as the Commission is concerned, there's just absolutely no way we could do that with charge files.  

Alina M. Semo: Okay. Thanks Michael. Appreciate that comment. Michelle, back to you. I believe we have another caller on the line.  

[Michelle] Producer: We certainly do. Caller, your line is unmuted. You may go ahead. You have three minutes for your comment. Caller, you'll need to unmute your line, we are unable to hear you. Caller, we're unable to hear you. I don’t know if you have your device muted, but we are not able to hear you at this time. All right. We're unable to hear that caller. I'm not sure why, looks like they hung up, but they might dial back in, but we were unable to hear them.  

Alina M. Semo: Okay. If they dial back in, please let me know. I also want to encourage the caller to submit comments via our online public comments portal. They're always welcome to do that. Alex just commented, I feel obligated to read it out loud because we're not supposed to be putting substantive comments in the chat. The fact that EEOC cannot do this with charge files does not mean that dozens of agencies across the federal government cannot do so with their responsive records. That's exactly what's been happening with FOIAonline. Okay, thank you Alex. Okay. I don't believe we have any other public comments, so we need to read into the record. Michelle, was that caller able to call back in?  

[Michelle] Producer: It looks like that's trying to join. One moment. All right, let's try again. Caller, go ahead. Your line is unmuted. You have three minutes. Yeah, we're still unable to hear that caller. I'm not sure why that is. Caller, we're unable to hear you. So go ahead and enter your comments in the instruction that Alina left with.  

Alina M. Semo: Okay. Thank you, Michelle.   

[Michelle] Producer: We can move on. We can't hear the caller.  

Alina M. Semo: Any other callers on the line? I just want to make sure before we start moving to wrap up our meeting.  

[Michelle] Producer: Yeah, let me double check. I do not see anyone else with a hand raised so no additional callers on the line.  

Alina M. Semo: Okay, thanks Michelle. I appreciate it. Okay. 

[Michelle] Producer: You're welcome.

Alina M. Semo: So I think we're ending pretty much on time, which is great and a little surprising because I know we had a lot of content today in our meeting. I thought it was a fantastic meeting by the way. I think there was a lot of great information that was presented to us today and that we all shared amongst ourselves. I want to thank all of the committee members for the hard work that's been going on so far and for your continued work in the next year. So hang in there. Don't give up. We're just in the midway stretch. I just have a quick announcement about our Annual Open Meeting that OGIS will be hosting virtually in this space, next Tuesday, June 13th. It's going to be a very brief one, 10:00 AM to 11:00 AM. We're just going to review our annual report which we anticipate publishing in the near future.  

And we will also have time for any interested persons to present oral or written statements at the meeting. Those who wish to present oral statements will need to register via Eventbrite to participate via our Webex platform. And again, oral statements will be limited to three minutes per individual. Okay. I want to wrap up this meeting today to thank everyone again for joining us. I hope everyone and their families remain safe, healthy, and resilient. We will see each other again virtually in this space at our next meeting, Thursday, September 7th, 2023, beginning at 10:00 AM Eastern Time. So with that, we stand adjourned. Thanks everyone. Have a great day.  

[Michelle] Producer: Everyone. That concludes, that concludes our conference. Thank you for using event services. You may now disconnect.