Daniel Schuman leads Demand Progress and Demand Progress Education Fund’s efforts on issues that concern governmental transparency/accountability/reform, civil liberties/national security, and promoting an open internet. He co-founded the Congressional Data Coalition, which brings together organizations from across the political spectrum to advocate for a tech-savvy Congress. Daniel directs the Advisory Committee on Transparency, which supports the work of the Congressional Transparency Caucus, and is a fellow at CodeX, the Stanford Center for Legal Informatics. His weekly newsletter, the First Branch Forecast, is just about as nerdy on Congress as you can get.
Hudson Hollister
Daniel Schuman, thank you so much for joining us today.
Daniel Schuman
It’s my pleasure to be here. Thank you
Hudson Hollister
Daniel, tell us a little bit about your current role at Demand Progress and about the portion of your work that focuses on congressional modernization.
Daniel Schuman
I’m the policy director at Demand Progress, a progressive grassroots organization. We do outreach, we have 1.5M+ members, but we also do a lot of advocacy and educational work in DC. I wear a number of different hats. On the 501(c)(3) side, I write a weekly newsletter, called the First Branch Forecast, focused on legislation and government transparency that features a lot of discussion on legislative operations and focuses on how to strengthen Congress and make it work better. On the 501(c)4 side, I focused a lot on lobbying and advocating for reforms to the legislative process, largely around technology, to make it easier for congressional staff and members of Congress to do their jobs, and easier for the public to interface with what they are doing so that we can have a better information ecosystem to make better policies and conduct better oversight.
Hudson Hollister
Can you tell us a little bit about your leadership of the Congressional Data Coalition?
Daniel Schuman
Historically, the United States Congress has not been particularly focused on making information available to the public as data. There have been some efforts to make information available at websites, but historically they have often been difficult to use. This has been frustrating to both internal and external stakeholders.
External stakeholders would have to build tools to try to scrape legislative information, to repurpose it and repackage it to understand what’s happening in the legislative process. Internal stakeholders were similarly disadvantaged. They often were trying to build tools or mesh together different datasets to help keep their stakeholders apprised of what’s going on and they were simply unable to do so. They found themselves relying on external stakeholders like Demand Progress, or places like GovTrack, to make use of the information that their own institution was creating.
The consequences of this dysfunction came to a head a decade ago when there was a real effort to publish legislative [data] online. The folks who were uncomfortable with this process fought an effective rearguard action to stop the effort from succeeding in of itself. They formed a committee…called the “Bulk Data Task Force,” [that] brought together all the people who wanted to modernize these processes. It’s a terrible name, but it became the institutionalization of the political players in the House and in the Senate, the support offices, and agencies who care about this. There are quarterly public meetings with stakeholders, which allows for interchange and exchange of information, as well as private meetings that also take place. It brings in some of the less enthusiastic actors, as well, to either sort of cajole them along or to have them continue to fight the rigid revanchist fights.
The initial result was the directive to publish legislative information as data. That is, the bill summary and status information behind what was then thomas.gov, which was Congress’s legislative information system.
Hudson Hollister
When did that initial publication of status information occur?
Daniel Schuman
2012.
GovTrack, the civil society entity run by Joshua Tauberer, has been publishing it since 2003 — and everyone has been using their data. Finally, Congress started publishing some of that information in 2012. One of the big problems with congress.gov, which thomas.gov became — it’s the Library of Congress’s website site — was that historically it was very slow and it missed a lot of important information about the legislative process.
So the House established its own legislative information portal called docs.house.gov, which is much more comprehensive for a certain set of documents than congress.gov. Sometimes it had transcripts, it had documents that people would submit in advance of hearings — it was the official place where you would find notice about what’s going on. Slowly but surely, different pieces of the institution that otherwise didn’t collaborate with each other have been collaborating more.
For example, the House will soon be able to track how an amendment would change a bill in real time or how a bill would change the law in real time. While this is easy to do at the state level, it’s very difficult to do at the federal level.
Hudson Hollister
Why is displaying the impact of a proposed bill on existing law harder on the federal level than on the state level?
Daniel Schuman
At the state level, all law is codified. The state codes are the law. The code is literally what the law is, reorganized and published. The U.S. Congress has not codified all of its laws. Some of the codification is actually the law and half of the codification is not the law. It looks like the law. It smells like the law. But the actual law is the organic bill as passed, then amended, amended again, and again.
What that means is that if you want to amend the organic law, you need to go back through all those layers of amendment, upon amendment, upon amendment, upon amendment, because codification of positive law, the actual law, is very hard for Congress — for a number of historical reasons that have to do with a vendor botching things, I think in the 1880s, and Congress being gun shy ever since.
It’s very hard to show how a bill would change the underlying law because it’s really hard to know what the underlying law is. A computer needs to be able to read and understand legislative language to go and make the additions and excisions.
It’s actually a very hard project. But the bottom line is: it’s working … It will save staffers tons and tons of time.
Hudson Hollister
So this is an instance where the beautiful is not the enemy of the good?
Daniel Schuman
No, in this case, there is truth in beauty.
Hudson Hollister
So you have a progression from the formation of the Bulk Data Task Force to the House’s decision to establish docs.house.gov to current projects, including impact of bills and amendments. At what point during this long march toward the beauty did you co-found the Congressional Data Coalition?
Daniel Schuman
That was actually done as a consequence of when I left the Sunlight Foundation. When I left, I missed all my friends. That was in 2013. I founded it along with Zach Graves, who was then at the R street Institute, which is a right leaning organization. We founded the Congressional Data Coalition to basically keep everybody together.
It was an organizing and communications mechanism. It’s eight or nine years old at this point. One of the lessons that we learned is, it’s not easy for politicians or for senior, non-political staff to deal with an amorphous crowd of people. But as you create a coalition, then it’s a point of contact. They can be like, “Hey Daniel, would you check with the other folks that participate about when’s the best time to have a meeting?” It facilitates some of the conversations that need to take place. It still exists, it’s very informal. Basically, you just have to show up to be a member, that’s what it comes down to.
But it is a known quantity for congressional staff. They know that they have someone who will be the point of contact. And, you know, if I go and win the lottery and retire to Tahiti, a point of contact will still exist, even if I’m not there. It’s designed to persist and the blog posts that we put up tell the story of what’s happened. It tells you what happened at this quarterly meeting, or what happened at the annual legislative data and transparency conference, or, “Hey, someone’s developed a new tool that’s really cool.” It’s a way of publicly institutionalizing a lot of what’s happened on the civil society side so that we have a memory of what’s happened before.
It also provides a positive feedback loop for the congressionals to see what happened previously. We’ve actually strengthened that mechanism. We’ve built out some of our congressional focus aspects. One of the things that we have is a newsletter called the First Branch Forecast, which every week delves into a wide range of issues, including the legislative technology stuff, and it’s very heavily focused and aimed at legislative staff. It has wide penetration both for the appropriators, the authorizers for leadership, and then the reporters who cover it. So we’re creating this feedback cycle, this loop that allows for better sharing and spreading of innovation, applauding the good things that have happened, and drawing attention to issues that need more attention.
I should say, it was through this work, in large part, that Congress was able to function during the COVID pandemic because a lot of the groundwork had been laid to allow some of the remote work possibilities that were necessary to keep our federal government open and running. It has had a tremendous impact there.
Hudson Hollister
That’s a great segue to discuss the current state of Congress, in terms of the use of data and technology to manage legislative information. Before we get to that, I want to ask an assessment question. Is it too much of an oversimplification, to view the Bulk Data Task Force as the primary in-house rallying point for congressional stakeholders on digital modernization and the Congressional Data Coalition as the primary outside rallying point, balancing each other? Is that going too far?
Daniel Schuman
I think it’s too far. On the outside, I think it’s right, mostly. Only because we have a very big tent. We include journalists, civil society, people for profit and nonprofit, information vendors. It’s the whole range. Anybody who wants to participate can participate.
On the inside, the Bulk Data Task Force is one of the major points of contact and validation. A lot of the work though is driven through the appropriations process. Appropriators are connected to leadership, so when leadership says we want to do X, Y, or Z, the leadership is involved in the Bulk Data Task Force. It will make it over to appropriators. But we found that advocacy aimed at the leg branch appropriations process has been a huge driver of this change, as well. There have been some from the authorizers, although their attention is inconsistent, sometimes it’s good, and sometimes it’s not. But the appropriators are the place where you can go in and say, “Would you just put a paragraph in this report language, to get a study, to get a test case to build a model,” and persistent advocacy over time moves these things forward.
Hudson Hollister
Tell us a bit more about how the modernization activities pursued by the Congressional Data Coalition created capacity that proved crucial during the pandemic.
Daniel Schuman
There are a number of related issues.
One is, is your email in the cloud or is on your local machine? Are you Office 365 or is it on your local device? They had just finished the migration to 365. They were very lucky.
Before the pandemic, if you wanted to introduce a bill or submit an extension of remarks, the way it’s traditionally done, is you take the paper, you walk down to a box called “the hopper,” you put the paper in there, and then they take the paper, retype it, and that’s how it becomes a bill. Which is nuts. What they changed was they built a slightly better system where you can now email it in. Now, they still print it out on the other side and do a whole bunch of things that are kind of ridiculous. But since being there in person was deadly, you don’t have to do that anymore.
Another tool that’s come online, is the ability to [co-sign letters] electronically. In days of old, when you and I worked in Congress, when I was an intern, you’d get a Dear Colleague, you’d walk around office to office and go to a staff assistant who would take out their pen and they would sign it for the Member. Then at the end of the day, with your feet really sore, you’d have all the signatures from the offices that were supposed to sign on [to the letter]. Now you have an online tool, where you send it to the office, they’re like, “Yep! Done.” They’re signed on, it’s authenticated, and that’s it. It’s beautiful. It’s called Quill. It works now in the Senate and in the House, it’s just been turned on. The documents look perfect and it saves a ton of time.
Other things, tools like Zoom were not approved for use by Congress. Zoom is now approved, WebEx is approved, Skype for Business, I think that’s the thing, that’s approved. All these that were FedRAMP certified, but not in use, are now in use everywhere.
Then, of course, there was the learning curve because Members are not known for being technologically advanced. We held a mock hearing at this time last year. We had General Petraeus as a witness, Marcy Harris at PopVox, myself and others, and a former representative from Washington State played the clerk. We had 30 or 40 former Members of Congress, play Members of Congress, which was a role for which they were well suited. We demonstrated that you can have a remote hearing and a remote markup.
Hudson Hollister
Without any problem.
Daniel Schuman
Without any problem!
The Members could figure out how to raise their hand to vote, they could verify the votes, and we demonstrated that it was cyber secure and feasible that Members, in fact, could figure out how to turn on and unmute themselves. And that helped pave the way for the House in the Senate, to a lesser extent, to drop hybrid and fully remote proceedings where they conducted a lot of business. That’s a consequence of modernization plus the coalition Members coming together in partnering to make sure that the legislative branch wouldn’t fail.
Hudson Hollister
It sounds like there is a lot of patience on the part of the coalition Members.
Daniel Schuman
Honestly, it was terrifying. It was really hard. When the House left they had failed to put in place a resolution that would have allowed them to deliberate remotely.
Hudson Hollister
Which meant that the legislative branch of the country could not function.
Daniel Schuman
Yeah. We warned them and they won’t listen to us. They wouldn’t listen to us for two months. Until finally enough of them came back to change the rules. And they still have to do it right and the Senate hasn’t done it, really. But the legislature would die should this happen again, probably. The Senate might, because the Senate still requires a quorum in person. So if there were to be something really bad, like another plague, or air travel were to end like on 9/11. There would be problems with the Senate being able to deliberate, but it’s not like the Senate is known for its rapid pace of deliberation to begin with.
Hudson Hollister
How close are we to the specific digital improvements like digital native drafting and automated self-executing amendments? How close are we to those in Congress and what are some of the barriers that you see?
Daniel Schuman
I think we’re very close to self-executing amendments … We’re very close.
Hudson Hollister
And you’re suggesting that being able to compare is the same thing as being able to amend?
Daniel Schuman
Well, it’s three parts. It’s bill against bill, amendment against bill, and it’s bill against law. You can show how the amendment would change the bill, it will show you how we change either the organic law or the positive law, if it’s codified or otherwise. You will see what it will actually look like and it will save the drafters a ton of time, because you know what the end result is. And since people were relying on the end result, as a result of the technology, people will expect how we look up against the U.S. code or against the organic law.
They are also moving to the next version of XML for drafting purposes and they’re digitizing a lot of their data. The House has fully digitized the U.S. code. It’s digitized much of, but not all, public laws. There’s still some work to be done.
It’s important to talk about the stages. There is digitization, a scanned picture. They are still doing some of that, or perhaps a scanned picture with optical character recognition. If you go online and you look up older U.S. law, that’s what it is. It’s not data, it’s OCR. Then there’s the stage where it’s being typed in, but it’s not sufficiently structured. So you can’t actually see how everything relates. Then there’s finally a structured format.
We have all of those different projects happening simultaneously with the different pieces being treated in different ways by different siloed actors. But we’re getting there. In five years, all the law will be online. All the committee reports will be online. All of the introduced bills, maybe, we’ll have many of those online. We’ll have all the congressional records online. So you’ll be able to see what the committees did with the committee’s report. All that information will become electronically available for free. It won’t necessarily be available in a useful format, but some of it will. Some of it’s being transformed into giant databases where you can then run something up against something else.
Now, there are still things that need to be mashed up. We’re working on a project that is in alpha right now, that pulls together a bunch of disparate data sets that Congress hasn’t pulled together. And one of the big problems right now, is that the entities think of themselves as like what they want to do. They don’t use a user-centric approach. If you’re a staffer and you’re trying to figure out who do I want to have as a co-sponsor of this bill, or how does this work, or what are the related bills? None of the Congressional tech is built to pull together two different data sets to do this.
There needs to be a shift in mindset away from siloed entities. GAO reports, CBO reports, CRS reports, Library of Congress reports, committee reports, subcommittee reports, other documents, testimony — you have to think about how you can make it easier for the human being on the other side of all that stuff to be able to process and understand it in context. That shift in mindset towards customer-centric design towards user-centric design, towards thinking through the different use cases and building towards that, towards the willingness to break down silos. That has not yet been accomplished.
Hudson Hollister
That makes me think that perhaps some of the early forays towards modernization that simply came from a perspective of, “this should be automatic,” without applying the harder work of user-centered design might have been premature, or they might have been cutting corners.
Daniel Schuman
I would say it differently. I think that people build things for a reason that’s relevant at the time. Putting all your stuff onto microfiche made a lot of sense when it was done, because it saves you storage capacity, although they did it badly, and a lot of turns to acid later on. The creation of Thomas.gov was to solve a particular problem. I don’t think there was thought given to transformative uses elsewhere. I think there was a lot of fear around transformational uses. “What if someone misunderstands something that we published, we shouldn’t publish it!”
And that’s ridiculous. People can draw the wrong conclusion from no information just as much as they can draw the wrong conclusion from some information. There’s also, well, if someone else republishes it, then it’s not authentic. You test the authenticity against the original document, that’s how you know it’s authentic. But if you don’t have the original documents available, then you won’t have any way to test it. And what you end up creating are systems where people with money and access, get access to things that allow them to influence the political system, to the disadvantage of everyone else, including the members of Congress who can’t pay for those tools.
Hudson Hollister
You have been working on the modernization of Congress and of its processes and of its document content, for the better part of two decades. You are clearly still motivated. What keeps you motivated?
Daniel Schuman
Our democracy has to work. That’s what it comes down to.
There’s a couple of plays that are working through the back of my head. First, staff are woefully underpaid and there aren’t enough of them. It’s taken 14 years of working on this issue to change enough minds where we have a real shot to actually bring staff pay back to their levels of 2010. They’re down by an average of 20% per position. My theory is that, if pay is a hard and intractable issue, how do we make staffers into super staffers? How do we make it so they don’t have to waste their time doing a bunch of the mindless drudgery? Let’s bring the information revolution — that made it to businesses in the 90s — to the legislative branch.
You hire these staffers and these members, because they’re good with people, because they’re good with relationships. You don’t want them to be wasting their time figuring out, “How does this thing change this thing?” They should be dealing as much as possible in the realm of ideas, and reflecting the will of their constituents and their stakeholders and hammering out deals. All this other stuff should not get in the way.
So the question is, how can we take those other things out of the way so that we can engage in policymaking and political processes? At the same time, since they’re so strapped for time, how do we lessen their comparative disadvantage to lobbyists who have all the money and all the resources in the world to influence them. There’s nothing wrong in the world with having your own opinion. But everyone is entitled to the facts. We need to make sure that everyone can know what’s going on so you can’t get the wool pulled over your eyes. That you have a reasonable chance to make your case and to be heard, and to be heard by people who can actually listen to you and respect what you’re saying. That’s what drives me. I like building systems so that they work better. This is one of the most complex and important systems that you can possibly imagine. It needs to be strong, it needs to be resilient and given the constraints that we have, we have to optimize the hell out of it. Because that’s the only way that our political system is going to be able to survive.
Hudson Hollister
Yes, quite true. Daniel, as a former colleague in the policy world, and hopefully again sometime soon, in different iterations, and also as a voter, I thank you for all the work that you’ve done and for being at the center of this. I’m excited for our audience to get a sense of your work too.
