Computing Research Policy Blog
The Computing Research Association (or CRA) has been involved in shaping public policy of relevance to computing research for more than two decades. More recently the CRA Government Affairs program has enhanced its efforts to help the members of the computing research community contribute to the public debate knowledgeably and effectively.
[With this post, CRA welcomes Kayla Holston, our new Eben Tisdale Science Policy Fellow, who will be working with CRA policy staff this summer. Kayla is a rising second-year Rodman Scholar at the University of Virginia, pursuing majors in biomedical engineering and cognitive science. She’s particularly interested in computing as it relates to neuroscience research, and plans of a future practicing neurosurgery. We’re thrilled to have her on staff this summer and expect to see much more from her on the blog in the weeks to come! – Peter Harsha]
The Senate is considering S. 1398, or the “Energy Title of America COMPETES Reauthorization Act of 2015,” a bill aimed at “extending, improving, and consolidating energy research and development programs” at the Department of Energy, specifically at the Office of Science (SC) and the Advanced Research Projects Agency-Energy (ARPA-E). In simple terms, it sets funding policy for these two science agencies for the next three years. This bill was introduced by Senator Lamar Alexander (R-TN) in mid-May and is currently awaiting consideration by the Senate Energy Committee. The bill calls for an increase in funding for the DOE SC and ARPA-E and the funding numbers are aligned with the 2007 and 2010 COMPETES Acts. In addition, it proposes the elimination of a number of duplicative programs at the Department, a number of which are research fellowships. Let’s look at some of the details.
The funding numbers for both the DOE SC and ARPA-E are quite good and are true successors to early versions of COMPETES. As our readers will recall, the DOE SC is home to the Advanced Scientific Computing Research (ASCR) program, a program that encompasses much of the Department’s computing research. S.1398 calls for the DOE SC to be authorized at $5.271 billion for Fiscal Year (FY) 2016, a 2.7 percent increase over the pervious Fiscal Year budget ($5.131 billion for FY15)—it’s also worth noting that this is almost identical to the $5.247 billion authorized (but never appropriated) for FY11 in the 2010 COMPETES Act. After FY16, the proposed funding for the SC continues to exhibit an upward trend, with about a 4 percent increase each year:
FY17: $5.485 billion
FY18: $5.704 billion
FY19: $5.932 billion
FY20: $6.178 billion
It is important to note that, while the proposed FY16 funding is greater than what was appropriated in FY15, it is still below President Obama’s requested budget for FY16 ($5.34 billion).
S. 1398 proposes a similar funding pattern for ARPA-E. The Agency was authorized for $300 million in the original 2007 version of COMPETES, but has been mostly flat funded since its establishment. In S. 1398, Senator Alexander proposes a yearly ~4 percent increase in funding, aligning the numbers with what was originally proposed in the COMPETES Acts:
FY16: $291.2 million
FY17: $303.3 million
FY18: $314.7 million
FY19: $327.3 million
FY20: $340.6 million
It’s worth pointing out that S. 1398 only authorizes funds—it does not appropriate funding for the agencies. In other words, this bill would set policy for these parts of the DOE and state how much they should be funded. It is the job of the Appropriations Committees and separate bills to determine how much money DOE SC and ARPA-E actually get to work with.
Alexander also proposes the consolidation of a few programs. Namely, S. 1398 would combine funding for the DOE Early Career Awards for Science, Engineering, and Mathematics Researchers and the Distinguished Scientist Program. The bill also calls for elimination of the Protecting America’s Competitive Edge (PACE) Graduate Fellowship Program, with the idea that those grants will instead fall under the Science Education Program.
So what are the bill’s prospects? That’s unclear at this point. It’s not exactly clear whether the bill will move in the Senate as a free-standing bill, or get folded into a broader DOE authorization, or a Senate version of the COMPETES Reauthorization. What we do know is that it represents a marked change of approach from the DOE provisions in the House version of the America COMPETES Act Reauthorization, which we’ve previously discussed here. At some point in the legislative process, whether its with a COMPETES Reauthorization or an Energy Bill, the House and Senate will have to resolve their differences and find something they can both pass. What that final bill will look like is anyone’s guess, but from the community’s perspective, we hope it embodies more of the Senate’s approach than what we find in the House COMPETES Reauthorization.
We will continue to monitor S. 1398 for additional developments; be sure to check back for updates!
On May 20th the full House Appropriations Committee passed the Commerce, Justice, Science funding bill; this is important to our community because it is the bill that contains the funding for the National Science Foundation, which funds 89 percent of all university-led fundamental computer science research in the U.S. First, the not-so-bad news: NSF doesn’t exactly get a budget cut in actual dollars; it in fact gets a small bump (though when inflation is considered, that bump may go away completely). The worse news: NSF gets some onerous language on how to spend the tax-dollars it’s allocated. Let’s get into the details.
NSF received $7.34 billion in Fiscal Year (FY) 2015, the present budget we are working under. For FY16, the President requested $7.72 billion, or an increase of $379.3 million (or 5.2 percent) over FY15 levels. In the House Appropriations bill, the agency would receive $7.39 billion in FY16, an increase of $50 million (or 0.7 percent). Again, it’s an increase in actual dollars – so things could have been worse. But the bill contains some disheartening language restricting the agency’s activities.
The most significant section concerns restricting how money in NSF’s Research and Related Activities account, where most of the money for research resides, can be spent. The relevant section reads:
The Committee directs NSF to ensure that Mathematical and Physical Sciences; Computer and Information Science and Engineering; Engineering; and Biological Sciences comprise no less than 70 percent of the funding within Research and Related Activities. Further, the Committee directs that NSF allocate no less than the fiscal year 2015 levels for the Office of International Science and Engineering; Integrative Activities; and the U.S. Arctic Commission.
This language would force Geosciences (funded at $1.3 billion in FY15) and Social, Behavioral, and Economic Sciences (funded at $272 million in FY15) to share roughly $1.32 billion in funding. The combined budget of the two directorates in FY15 was $1.58 billion, so this language translates into a cut of about $257 million to the two directorates, or 16.3 percent. This is in line with the authorization levels passed by the House Science Committee in their bill H.R. 1806 the America COMPETES Act of 2015, which CRA has stated it opposes.
Of almost equal concern, the Appropriations Committee included the following section concerning, “Transparency and accountability:”
The Committee supports section 106 of the America COMPETES Reauthorization Act of 2015, H.R. 1806, as reported, that enhances transparency and accountability of NSF grants by requiring that public award abstracts articulate how the projects serve the national interest. The Committee understands that NSF has already taken steps to implement these transparency processes. NSF is directed to comply with section 106 and provide periodic updates to the Committee on its transparency activities.
As our readers will recall from last year, a similar section to Section 106 is one of the areas that CRA has had significant concerns over, as it puts unnecessary bureaucratic burdens on the agency and is already covered in legislation which founded NSF. While the original language in the section was much worse than what the committee ended up with in COMPETES, we believe it still creates an unnecessary level of bureaucracy and provides a “fix” to a process that isn’t truly broken.
The only other issue of note to our community is a short section on high-performance computing. The committee’s language urges NSF to continue its commitment to the field by modernizing its facilities. Here is the relevant text:
The Committee urges NSF to continue its commitment to modernizing its world-class big data and high-performance computing, which support all areas of scientific research and education, including the most demanding scientific challenges.
So, where do things go from here? The bill now moves to the House floor for consideration; it is likely to pass along a party-line vote, just as it did in committee. It is expected on the House floor as early as today but probably take a day or two to get to a final vote. On the other side of Congress, the Senate Appropriations Committee still needs to complete its own CJS bill. When that will happen also isn’t yet known; the Senate has been the bottleneck in the Appropriations process for the last number of years. But with a new Republican majority, there is a chance that the chamber will move more quickly. Even if that happens, we would not expect a Senate CJS bill until the end of June, at the earliest. We can hope for a better outcome there, but that is not guaranteed. Assuming the Senate gets their bill together, the two chambers will then have to come up with an agreed upon bill in conference, which would then have to be passed by both chambers, before being sent to the President’s desk. That’s still a long way away, so we will have to wait to see how events unfold over the coming summer.
Be sure to check back for more updates on the Fiscal Year 2016 budget.
The House of Representatives today will consider the America COMPETES Reauthorization Act of 2015 (H.R. 1806), a bill designed to set policy at three key science agencies and authorize funding for the next two fiscal years. The bill is being brought to the floor by House Science, Space and Technology Committee Chairman Lamar Smith (R-TX), who secured passage of the bill through his committee on a strict party-line vote. CRA raised concerns about the bill at the committee level citing issues with the funding levels authorized and a disagreement over significant cuts the bill would make to authorizations for the Social, Behavioral and Economic sciences as well as the Geosciences — concerns that remain as the bill moves to the House floor.
The bill is a reauthorization of the America COMPETES Acts of 2007 and 2010, bills which sought to authorize robust and sustainable investments in fundamental research at three key Federal science agencies: the National Science Foundation (NSF), National Institute of Standards of Technology (NIST), and Department of Energy’s Office of Science. They were inspired by a conclusion reached by the National Academies that the U.S. was woefully underinvesting in the physical sciences (which, in DC parlance, is anything that’s not life science) and that underinvestment was constraining U.S. innovation and competitiveness in an increasingly competitive world. They set the three agencies on a funding trajectory that would authorize a doubling of their research budgets over seven years. They were both strong, bipartisan statements about the importance of investments of fundamental research and the priority the Federal government ought to place on them. Unfortunately, the current bill does not share that priority and does not set these agencies on a path to robust, sustainable investments in research.
We’ve written about the concerns we raised with the committee. Though the bill provides a healthy increase to the NSF Computing and Information Science and Engineering directorate in the first year of the authorization (FY 2016), it does so at the expense of the geosciences, and the social, behavioral and economic sciences. We believe that these cuts are detrimental to not just the research portfolios in those disciplines, but detrimental to computer science as well. Work in computer science — in cyber security, in human-computer interaction — is informed by work that comes from the social, behavioral and economic sciences. Understanding human interactions and motivations is crucial in understanding how to build hardware and software that is more secure and easy to use.
CRA is not alone among scientific societies in opposing H.R. 1806. Forty-three other societies, universities, coalitions and task forces joined in writing letters of concern about the measure. Despite that, it’s likely that the bill will pass, on a largely party-line vote, in the House today. This would be a disappointing result, not just because of the content of this particular bill, but because it signals a particularly partisan approach to setting science policy that Congress has largely avoided in past years. Federal science policy is too important to pass on a party-line vote.
On April 29th, the Coalition for National Science Funding (CNSF), an alliance of over 140 professional organizations, universities, and businesses, held their 21st Annual Capitol Hill Exhibition. CNSF supports the goal of increasing the federal investment in the National Science Foundation’s research and education programs, and the exhibition itself is a great way to show members of Congress and their staff what research the American people have funded.
This year the Computing Research Association, a member of CNSF, sponsored three students, two PhD candidates and a defended PhD candidate, from Johns Hopkins University to come to Washington to demonstrate their work. Kelleher Guerin (the defended PhD candidate) and Amanda Edwards demonstrated their collaborative robot for manufacturing, called CoSTAR; while Colin Lea demonstrated a virtual reality interface that can be used to more easily program robots by novice, non-technical users. All three young researchers are advised by Greg Hager, professor and chair of the Department of Computer Science at Johns Hopkins University and Chair of the Computing Community Consortium; Dr. Hager was also in attendance at the event and fielded questions.
CoSTAR, a system for human instruction and collaboration with robotic systems, has been developed to enable experts skilled at a task, but not skilled with programming, to instruct the robot as an apprentice. In apprenticeship, a teacher understands the capabilities of the apprentice, and builds on those capabilities until the apprentice can perform the desired tasks. Their system allows for a similar instruction of robots by representing the capabilities of a robot as a set of easy to understand building blocks. A novice end-user can then use these building blocks to create a plan, which the robot follows to accomplish a task. This also allows for the robot to reuse information it is taught across many different tasks. Their system also allows for the robot to collaborate with humans, and respond to dynamic events just as a human would. In addition, they showed how advances in virtual reality could provide an environment for intuitive robot interaction and teaching.
With regard to the virtual reality interface, it was noted that most robots used in factories are old and not intended to be used directly with humans. Users must stand behind shielding where they spend a substantial amount of time programming the robot. The researchers’ work showed how advances in virtual reality could provide an environment for intuitive robot interaction and teaching. Using a virtual reality headset, and a pair of 3D joysticks, a user can virtually move the robot around as if they were performing the task themselves.
All of this work is supported from the CISE directorate at NSF. Both projects were well received by the attendees of the exhibition; in fact, the students fielded questions from members of Congress, Congressional staffers, NSF Program Officers, and even the NSF Director, France Córdova.
A number of other organizations had displays and were demonstrating NSF funded research at the event. From the University of Illinois’ National Center for Supercomputing Applications “NSF Blue Water” supercomputer, to the American Political Science Association’s “American National Election Studies: Understanding the Changing American Electorate,” to the American Astronomical Society’s “Disruptive Technology & Cosmology at 17,000 Feet,” the exhibition was a great display of the different types of research being supported by NSF. Look here to see a list of some of the participating organizations and what a few of the exhibitors were presenting.
The House Committee on Science, Space and Technology is currently marking up the America COMPETES Reauthorization Act of 2015 (H.R. 1806), introduced by Committee Chair Lamar Smith (R-TX) — a bill designed to provide three key science agencies with authorizations for funding for FY 2016 and FY 2017 and implement other policies. The bill is cast as a reauthorization of the original America COMPETES Act of 2007, which, inspired by the seminal National Academies report Rising Above the Gathering Storm, was a bipartisan attempt to buttress Federal support the National Science Foundation, Department of Energy’s Office of Science, and the National Institute of Standards and Technology in recognition of the critical role they play in fostering U.S. innovation and competitiveness in an increasingly competitive world. That original bill attempted to put those agencies on a path to doubling their research budgets over seven years — an important symbolic goal that demonstrated Congress’ commitment to support the physical sciences (which, in DC parlance, is anything not in the life sciences) after years of relative underinvestment.
H.R. 1806 also would authorize funding at NSF, DOE’s Office of Science, and NIST, as well as the White House Office of Science and Technology Policy, but does not provide for the steady and real growth in the Federal investment in research called for by the Gathering Storm report or enacted by past COMPETES authorizations. While the bill includes authorized increases for some programs in FY 2016, the bill holds agency funding flat in FY 2017 — and that means a cut in real dollars when inflation is considered.
Though the bill also shows strong support for investments in computing research at NSF and DOE, it does so at the expense of other science disciplines, including the social, behavioral and economic sciences (SBE) and geosciences (GEO). CRA, along with many partners in the science community, opposes this approach, especially given the importance of research in fields like SBE in informing computing research — particularly in computing areas like cyber security research and human-computer interaction. The insight into human behaviors provided by SBE is critical to understanding how best to design and implement hardware and software that are more secure and easier to use.
For these reasons, CRA opposes passage of H.R. 1806 as introduced. CRA released the following letter to Committee Chair Lamar Smith and Ranking Member Eddie Bernice Johnson (D-TX) explaining our lack of endorsement.
April 21, 2015
The Honorable Lamar Smith
House Committee on Science, Space and Technology
2321 Rayburn House Office Building
Washington, DC 20515
The Honorable Eddie Bernice Johnson
House Committee on Science, Space and Technology
394 Ford House Office Building
Washington, DC 20515
Dear Chairman Smith and Ranking Member Johnson,
As an organization representing over 200 PhD-granting departments in computing, 16 industrial computing research labs, and 6 affiliated computing societies, we commend you both on your long-standing efforts to support the Federal investment in fundamental computing research. While we are pleased to see elements of that support continue in the America COMPETES Reauthorization Act of 2015 (H.R. 1806), other concerns — including the overall level of support provided to key science agencies in the act — prevent us from offering our endorsement of the bill.
In particular, we are disappointed to note that the bill, by flat-funding science agencies in the second year of authorizations, fails to provide for steady and real growth in the Federal investment in research, something we believe is critical to our Nation’s ability to compete, prosper and be secure in the coming years and decades. Indeed, when inflation is considered, the authorizations for FY 2017 represent real reductions in research investments, including investments in computing research.
We are also disappointed to note that research at the National Science Foundation in the Social, Behavioral, and Economic (SBE) sciences, along with the Geosciences, would be curtailed under this authorization. As you are aware, research in several key areas of computing — including cyber security and human-computer interaction (HCI) — is significantly informed by work emanating from the SBE directorate. The insight into human behaviors provided by SBE-funded work is critical to understanding how best to design and implement hardware and software systems that are more secure and easier to use. In cyber security work, where the human is often the weakest link in the chain, it is especially crucial to understand the varying motivations and usage patterns that dictate how people interact with their machines, and the expertise in studying those issues in large part resides in the social, behavioral and economic sciences. In HCI work, expertise in social, behavioral and economic sciences is critically valuable in creating workplace systems that foster collaboration and creativity, creating disaster response systems that influence people to effectively find shelter and assistance, and creating systems that motivate medical adherence and compliance with medical treatment.
We would be happy to work with you and your staff to help address these concerns and create legislation we could support wholeheartedly. However, as the act stands, we are unable to offer our endorsement.
J Strother Moore
Information on the ongoing markup, including a live stream, can be found here. CRA joins with a large number of academic, scientific and industry groups in opposition to H.R. 1806. We’ll have more detail about the results of the markup soon.