House Publishes Its Rules, Jefferson’s Manual, & More Online as Structured Data

Today the Government Publishing Office published the House Manual — which contains Rules of the House of Representatives, Jefferson’s Manual, and other important legislative documents — online in a structured data format on GitHub. GPO did so pursuant to direction from the House Rules Committee, which was acting in accordance with the rules package passed at the beginning of the 114th Congress, which declares:

The House shall continue efforts to broaden the availability of legislative documents in machine readable formats in the One Hundred Fourteenth Congress in furtherance of the institutional priority of improving public availability and use of legislative information produced by the House and its committees.

The online publication of key legislative documents as structured data is a welcome development. The Congressional Data Coalition has for a long time requested the enhancement, which empowers further analysis and reuse of the information in many different context. I, for one, will be glad to be able to automatically track revisions in the House Rules from Congress to Congress. I know others will find much more insightful uses.

All the offices and agencies involved with the project deserve congratulations, including: the House Rules Committee, the House Parliamentarian, the Clerk of the House, and the U.S. Government Publishing Office. (I’m sure there are more.)

The House Manual, as helpfully explained in the user guide to using the electronic version of the manual, includes:

  • the U.S. Constitution
  • Jefferson’s Manual
  • the Rules of the House of Representatives
  • Provisions of the Legislative Reorganization Acts
  • Congressional Budget Act laws
  • Legislative Procedures

All of the files can be found here. I have not had an opportunity to fully review what’s online — for example, I’m trying to find an unannotated version of the House Rules— but GPO has helpfully requested feedback on the GitHub page.

House Rules Committee Chairman Pete Sessions hailed the move in a statement:

Technology plays an important role in our daily lives, and it is necessary that the House keep up with the most efficient and effective ways to provide information about Congressional activities. As Chairman of the House Rules Committee, I am committed to the advancement of sharing legislative data online and am confident that our efforts will result in a better informed public.

GPO explained, in a press release, the value of publishing the documents in a more flexible format:

Making Government information available in XML permits data to be reused and repurposed not only for print output but for conversion into ebooks, mobile web applications, and other forms of content delivery, including data mashups and other analytical tools by third party providers, which contributes to openness and transparency in Government.

Good job. And thank you.

Continue Reading

2016 Legislative Data & Transparency Conference Set for June 21

The Committee on House Administration will host its fifth annual Legislative Data and Transparency Conference on June 21, from 9-4 in the U.S. Capitol.

Free registration is now open. Continue Reading

A Guide for Appropriators on Opening Up Congressional Information and Making Congress Work Better

For the fifth year in a row, today members of the Congressional Data Coalition submitted testimony to House Appropriators on ways to open up legislative information. The bipartisan coalition focused on tweaking congressional procedures and releasing datasets that, in the hands of third parties, will strengthen Congress’ capacity to govern.Continue Reading

Congress open up the tap for Legislative Data!

Today the Government Publishing Office and Library of Congress completed a full revolution in public access to legislative information. Information about legislative actions in congress–the bills, summaries of the bills, and their status–is now available online, in bulk, in a structured data format. As I wrote in December, this has great significance:

  • It marks the publication of essential legislative information in a format that supports unlimited public reuse, analysis, and republication. It is now possible to see much of a bill’s life cycle.
  • It illustrates the positive relationship that has grown between Congress and the public on access to legislative information, where there is growing open dialog and conversation about how to best meet our collective needs.
  • It is an example of how different components within the legislative branch are engaging with one another on a range of data-related issues, sometimes for the first time ever, under the aegis of the Bulk Data Task Force.
  • It means the Library of Congress and GPO will no longer be tied to the antiquated THOMAS website and can focus on more rapid technological advancement. (At least for data from the 113th and 114th Congresses).
  • It shows how a diverse community of outside organizations and interests came together and built a community to work with Congress for the common good.

Many more good things are on the horizon, and we have hopes for many more from Congress and civic coders, but for now, congratulations to our tireless friends in Congress and the legislative support offices and agencies that have labored to make this happen. Read their announcement here.

Why Congressional Technology Spending Matters and How to Improve the Situation

An architect's rendering of the U.S. Capitol Dome from between 1853 and 1865 from the Library of Congress.
(Architect rendering via Library of Congress)

Crossposted from The OpenGov Foundation blog.  Read The OpenGov Foundation’s prior analysis of Senate and House tech spending.

As our analyses have shown, without complete, accurate and timely spending data, the public is effectively guessing at where their tax dollars go and what value is returned. It is reasonable to expect that Congressional decision-makers are in a similar situation. If you’re shooting in the dark like this, it is difficult — if not impossible — to hit your target. Serving and staffing in Congress is already hard enough without these expensive additional headaches.

Is there another Aaron Schock situation or other Member malfeasance hiding in plain sight? Is one Senator paying more than another for the same tech support service, or same web development work? Are there better, cheaper open source options available? Is Congress getting the best price and best value, or is it getting ripped off? Are there opportunities to pool resources and save significant taxpayer money?

The public doesn’t know. We couldn’t find out. And it is reasonable to expect that decision-makers on Capitol Hill don’t always know either.

The answer is clear: Congress needs to fully overhaul its accounting systems and financial oversight processes. Everything from the software and data formats used, to internal reporting and controls, to how and when it publishes the results all appear to be largely outdated, inefficient and paper-based. From our perspective, every step of the process — from a staff assistant typing in receipts to a citizen sifting through thousands of PDF pages — is a major pain in the posterior and in need of an upgrade. The root causes of this problem must be addressed with nothing short of a full overhaul. Until that transformation happens, Members of Congress, their constituents and the news media will either remain marooned in ignorance or continue to make incredibly important decisions based on a mere sliver of the truth.

Going Further: Garbage In? Garbage Out

Many would wash their hands of all this after suggesting a better way forward. We did not and we will not. Because from our perspective the crucial question to be answered is: why is the data so unreliable and low-quality? It simply is not enough to highlight some of the public-facing problems, most of which have been identified at some level inside Congress. It is not enough to suggest a solution. The missing piece, crucial to anyone invested in doing something about these problems, is understanding why the situation is as it is and how it got that way.

To this end, we went as far upstream as possible in the congressional disbursement workflow: the point where real staffers in real congressional offices document their office’s expenditures for reimbursement out of each Member’s representational allowance (MRA). To put it mildly, going all the way upstream explained pretty much all of our frustrations.

The main menu of the Congressional Accounting & Personnel System (CAPS).
(Main menu screenshot via CAPS User Manual)

Introducing: Congressional Accounting & Personnel System (CAPS). This is the software platform most used by congressional staff comply with their boss’ public reporting requirements and internal compliance procedures.

In brief, here’s how spending data input process can work. A Member or staffer buys a thing or a service that supports official business. If and when the Member or staffer remembers to do so, they hand junior staff paper receipts and handwritten voucher cover sheets for submission to the Finance Office and entry into the internal finance system called PeopleSoft. One by one, interns or staff assistants painstakingly type each expenditure into CAPS, which eventually feeds into PeopleSoft, deciding how to classify each expenditure, vendor and purpose. Recalling one of our earlier examples, one staff assistant could type in “LEXISNEXIS” (no space) while another types in “LEXIS NEXIS” for the exact same purchase. Given this reality, Congress just doesn’t have the right tools, or sufficient staff, to resolve the countless issues that then arise downstream.

The voucher cover sheet menu in the Congressional Accounting & Personnel System (CAPS).
(Voucher cover sheet via CAPS User Manual)

Alternatively, offices can fill out a paper voucher form with receipts. Those documents can be scanned and sent or physically dropped off to the Finance Office for entry into PeopleSoft. Some offices directly enter their data into PeopleSoft with scanned receipts. Among the options, most offices submit through the CAPS system. All of these methods involve significant data entry and records conversion that can frustrate even the most diligent staffers.

Want to learn more? We obtained the full 200 page user manual for the Congressional Accounting & Personnel System (CAPS), which you can read here.

Congress Can Fix This

This critical situation is not the fault of any one person or office. But it is a growing problem that must be addressed as soon as possible. Fortunately, current leaders in both the House and the Senate should have a firm grasp on the issues at hand. After all, this is a Congress that recently spent four years securing passage of the watershed DATA Act, which requires that the Executive Branch transition its grants, loans and contracts spending information to modern, open data formats produced by modern software. Before that, Members of Congress and staff marveled at how well Recovery.gov fulfilled its oversight responsibilities to stop waste, fraud and abuse of billions of dollars of stimulus spending. And it is safe to say that everyone on Capitol Hill is familiar with the benefits of personal online banking.

Congress clearly understands the problem, and how modern technology and open data formats can solve it. And while there are many off-the-shelf products and services that would mark a vast improvement over the current CAPS system, we recognize the challenges posed by such sweeping changes to key congressional systems, not to mention its culture of compliance. That’s why there are many outfits like ours that stand ready to assist in upgrading these systems and procedures. The time is now, for the problem grows larger–and the solution, more costly–with each passing day.

The good news is that, within the House, change is coming. From reading 2015 semiannual report, it appears that the Chief Administrative Officer (CAO) of the House is making progress towards implementing a new system, Hyperion. From the CAO report:

“Staff who handle finances in House offices will have a powerful new tool to help them. Hyperion software is a faster and more efficient way to plan office budgets with real-time numbers that can easily be adjusted to offices’ changing needs. The Office of Finance is currently testing the system. It is scheduled to go live with phase 1 of the project on October 1, 2015. The system will launch with the Office of Budget Policy and Planning (BPP) along with all House Fiscal Year Offices. Phase 2 will be an expanded rollout for Member, Committee and Leadership (MCL) offices and other non-CAO offices for the replacement of the Congressional Accounting and Personnel System (CAPS) budgeting functionality.”

We count that as progress; however, if our Congress is spending taxpayer money to implement a modern, custom-made accounting and finance system, we believe that the public should be able to access the source code without restriction, and that other government bodies — first and foremost the U.S. Senate — should be able to reuse the code and build off of the system developed by the House. Unfortunately, the owners of Hyperion software, Oracle, places such ridiculously heavy and straight-jacketing restrictions on all of its customers that even the most powerful government of the most powerful country in the history of humanity doesn’t appear to be able to break free.

Still, this is a positive development; moreover, on page 1047 of the December 2015 omnibus spending bill, it appears that $1,300,000 in funding was set aside “for upgrade of the Legislative Branch Financial Management System.” We hope that this amount is sufficient to allow rapid deployment of the software by the House CAO. Completing the overhaul of the House’s financial systems should be a, if not the, top priority of whomever becomes the next Chief Administrative Officer of the House. And if the Senate hasn’t started its long-overdue upgrade, it should get moving without any further delay.

In the year 2016, no one, especially those in charge of the legislative branch, should have to fight as hard as we have to find out how Congress is spending taxpayer money. No Member or staffer should have to spend countless hours on financial compliance, when there are so many better and faster ways to fulfill their fiduciary responsibilities. And no one should cast aspersions on those who serve in Congress, or make negative assumptions on their spending decisions, without first taking the time to understand how incredibly hard it is to work on Capitol Hill in this age of shrinking budgets and growing workloads.

Now more than ever, all Americans must take a step back and reflect on the potentially dire consequences of the larger problem before us. We are continuing to force the men and women serving on our behalf in government to try to tackle our shared 21st century challenges without providing them the basic 21st century tools to get those wickedly hard jobs done.

Seamus Kraft is the Executive Director and Co-Founder of The OpenGov Foundation.

Seven years ago today, WikiLeaks published 6,700 CRS reports…and nobody was hurt

Feb. 8, 2009 began like any other Sunday for me. I was up early taking care of a young child, gulping coffee and scanning the news. The New York Times turned my stomach with its report that a Polish engineer had been beheaded by the Taliban. The Grey Lady also gleefully described the tribulations of Republican Party Chairman Michael Steele. A happier bit of news came fromThe Washington Post. Its sports section reminded me that my Cleveland Cavs were playing the Los Angeles Lakers on television at 3:30 that afternoon.

At some point that morning, I managed to get online. (I did not have a smartphone back then.) The Internet was abuzz with speculation about the 51st Grammy Awards, which were that night. (Robert Plant and Alison Krauss would fare well.) And there sat an email from a colleague of mine at the Congressional Research Service, asking if I had heard that WikiLeaks had published a trove of our reports.

I hadn’t.

Julian Assange’s outfit announced its work online with relish:

Change you can download. WikiLeaks has released nearly a billion dollars worth of quasi-secret reports commissioned by the United States Congress. The 6,780 reports, current as of this month, comprise over 127,000 pages of material on some of the most contentious issues in the nation, from the U.S. relationship with Israel to the financial collapse. Nearly 2,300 of the reports were updated in the last 12 months, while the oldest report goes back to 1990. The release represents the total output of the Congressional Research Service (CRS) electronically available to Congressional offices. The CRS is Congress’s analytical agency and has a budget in excess of $100M per year.

My stomach fell, and various thoughts ran through my mind. What would happen at the office? Were my reports in the trove? (Answer: yes.) My email address and 5-digit phone extension were on our reports. Would lobbyists and angry members of the public start contacting us to complain about our work?

That week at the office, I and many other CRS employees jabbered anxiously in the hallways and over coffee about how this could have happened and what it meant for the agency.

As it happened, the WikiLeaks massive report dump had no effect on us. In my succeeding six years at the agency not once did I hear any employee report any harm had resulted. The reason was not difficult to discern: many, if not most, of the CRS reports already had been put online by the Federation of American Scientists and various agencies of the U.S. government itself. Lobbyists and other hyperpolitical sorts inclined to pressure CRS already had our reports, often acquired via pricy subscription services.

This bit of history is important to keep in mind as Congress considers legislation to expand public access to CRS reports. There are individuals on Capitol Hill who claim bad things will happen if this commonsensical reform is enacted. The WikiLeaks report dump demonstrates that posting CRS reports to a central public repository (like GPO.gov) will do no harm to either CRS or its employees.

Today, more than 27,000 copies of CRS reports are scattered about the Internet. The average citizen, however, does not know where to look for them or what keywords will bring them up in Google. (Compare this search versus this search.) It’s way past time for Congress to give the public some value for the $100 million it spends on CRS each year.

Will the needless secrecy surrounding CRS reports end this year?

Not quite a year back, Rep. Mike Quigley, D-Ill., sought to do a little good for the American public. He offered an amendment to an appropriations bill that would require the Congressional Research Service to post publicly a list of the titles of its reports. Advocates for taxpayers and proponents for government transparency were delighted.

The CRS is an agency in the Library of Congress. Its staff of civil servants produce 1,000 or more reports each year. CRS reports describe government agencies (e.g., the Federal Election Commission); explain policies (e.g., SNAP/food stamps); and tally government spending (e.g., Department of Defense appropriations). Congress does not release these nonpartisan reports as a matter of course, but those within the Beltway know where to find copies. More than 20,000 congressional staff have access to CRS reports, so access is not an issue for lobbyists and policy-insiders.

Quigley explained to House appropriators that CRS reports “often are extraordinarily well-done” and that making the reports available to the public “would empower our constituents with extraordinary information about key issues, policies, and the budgets we are debating here in Congress.” Taxpayers pay $107 million per year to fund CRS and none of its reports contain classified or confidential information. Additionally, two other legislative branch agencies – the Congressional Budget Office and the Government Accountability Office – already release their reports publicly. (See here for a dozen other arguments for public release of CRS reports.)

Quigley’s amendment would not have made the contents of the reports public. All it would have done is enable the public to better know what reports exist. Having a list would enable John Q. Public to more easily request copies of CRS reports through their members of Congress. Quigley’s measure also would have been helpful to congressional staff, who labor to respond to constituents’ vague requests. (“I heard there was a report on agriculture that talked about subsidies. Can you get it for me?”)

Sadly, Quigley’s amendment never got a vote or even a discussion. Someone in the room disliked the amendment and Quigley reluctantly withdrew it.

The whole scene was bizarre and out-of-character for the House, which has shown great willingness to open legislative information and data to the public. No possible harm could come from publishing a list of CRS reports. In fact, such lists have been produced for decades. The CRS publishes an annual report for Congress that details its achievements and enumerates all its new reports, complete with their titles and authors’ names. It would be very easy for the CRS to post these reports on its website. Instead, the agency devotes time and expense to post a redacted version of its annual report that omits the list of reports. Why? Possibly, the House Committee on Administration requires it. But nobody will say publicly.

However, all is not lost. One can find the lists of CRS’s nonredacted annual reports online. Older copies placed long ago in federal depository libraries have been digitized in recent years. (See here for example.) Copies of CRS’ lists of reports for Congress from the past 20 years have been posted online by government transparency advocates (see here and here). These workarounds are not ideal, but they are helpful.

And what about Rep. Mike Quigley? He has not reintroduced his amendment thus far. But that does not mean he has given up the fight. Rather, he joined forces with Rep. Leonard Lance, R-N.J., who wrote legislation calling on CRS to publish entire CRS reports—not just a list of them— on the House Clerk’s website.

Congress will not be in DC much this year, thanks to the election. One hopes it can muster the wee bit of energy needed to put an end to the needless secrecy surrounding CRS reports. The public supports open government and would be grateful for increased access to honest information about their government’s doings.

Annual Report of the Congressional Research Service 2014 by R Street Institute

What America needs in the next House Chief Administrative Officer

The Chief Administrative Officer is one of the most important but low profile staffers that makes the House of Representatives work. The CAO runs an office with a huge mandate. It manages critical technology services, oversees and coordinates the financial reporting system, ensures that the press have smooth access to legislative proceedings, handles physical resources like maintenance of certain office equipment, and much more. Ed Cassidy, the current House Chief Administrative Officer, will retire at the end of this year. His departure leaves a strong legacy of open government and public engagement initiatives and highlights the need for a strong leader to fill this role and continue to bring Congress into the 21st century.

While the Speaker of the House gets to pick next CAO, the Congressional Data Coalition has specific traits we would like to see in whomever fills this vital role.

The most important trait for the next CAO to possess is experience in creating transformative change in a large institution. Congress needs an overhaul from the paper and poorly automated systems of yester-century that drives down efficiency and snarls everything in red tape. It’s time for the House to move fully to digital processes by rethinking the ways things are done and how they are assessed. Congress needs a leader who understands this issue and can drive the change.

The ideal candidate for the House’s CAO should also be a:

  • Strong manager and team builder.
  • Advocate for public-private partnerships.
  • Compelling communicator to the public and those in Congress.
  • Champion of technology who is familiar with the importance of machine-readable, structured data.

The goals for the next House’s CAO should be to:

  • Recognize that providing more information to the public and better information to legislators helps Congress do its job effectively.
  • Explore opportunities to move paper and physical operations to digital systems, along with making underlying processes more efficient and effective.
  • Overhaul the Office of Finance with modern budgeting, reporting and compliance software to empower legislators to better manage and report their finances, including producing information about activities in open data formats.
  • Embrace open source software so the public can review and contribute to the tools that run their Congress.
  • Make it radically easier for small/medium-sized companies to do business with the House.
  • Modernize the hardware in all committee hearing rooms and the recording studio.
  • Review and reassess CAO’s mission in 21st Century.

We are eager to see who fills this important role and work with them to seize the opportunity to continue modernizing congressional operations.

Open Government Data Aids Congressional Oversight

According to the U.S. Constitution, Congress is supposed to maintain watchful eye over the executive branch. These days, there is just no way that Congress can do that. The executive branch has 4.1 million employees, 180 agencies, and spends $3.9 trillion each year. The people’s branch needs help, and open government data enables individuals outside Congress to help with oversight. Read more in my article for the Public Administration Times.

Congress Poised for Leap to Open Up Legislative Data

Following bills in Congress requires three major pieces of information: the text of the bill, a summary of what the bill is about, and the status information associated with the bill. For the last few years, Congress has been publishing the text and summaries for all legislation moving in Congress, but has not published bill status information. This key information is necessary to identify the bill author, where the bill is in the legislative process, who introduced the legislation, and so on.

While it has been in the works for a while, this week Congress confirmed it will make “Bill Statuses in XML format available through the GPO’s Federal Digital System (FDsys) Bulk Data repository starting with the 113th Congress,” (i.e. January 2013).  In “early 2016,” bill status information will be published online in bulk– here. This should mean that people who wish to use the legislative information published on Congress.gov and THOMAS will no longer need to scrape those websites for current legislative information, but instead should be able to access it automatically.Continue Reading