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.
Yesterday, President Obama issued an Executive Order establishing the National Strategic Computing Initiative. This new initiative is to ensure the United States’ continued leadership position in High-Performance Computing in the coming decades. As the announcement from the Office of Science and Technology Policy (OSTP) says, “this coordinated research, development, and deployment strategy will draw on the strengths of departments and agencies to move the Federal government into a position that sharpens, develops, and streamlines a wide range of new 21st century applications.” The initiative is designed to, “advance core technologies to solve difficult computational problems and foster increase use of new capabilities in the public and private sectors.”
The announcement includes examples of the type of problems and potential research areas that this strategy can tackle. From the Precision Medicine Initiative to exascale computing at DOE, the potential for High-Performance Computing is almost limitless. Combined with the fact that advances in computing are helping to drive advancements in other scientific fields, this strategy could be similar to the Larry Smarr moment in the early 1980s, when NSFnet, one of the predecessors of the Internet, was established.
How is this likely to be received in Congress? It’s hard to tell, as Congress and the Administration have been at loggerheads over almost every political issue. That being the case, there is a good chance this Initiative will be received positively; long time readers will remember that the first two hearings the House Science Committee had this session were on HPC. This was done deliberately to educate new members of the committee on the field’s importance to the Federal scientific ecosystem.
In closing out this article, I want to quote the last paragraph of OSTP’s announcement, as it is quite concise as to the importance of this strategy, and HPC as a whole, to the country:
By strategically investing now, we can prepare for increasing computing demands and emerging technological challenges, building the foundation for sustained U.S. leadership for decades to come, while also expanding the role of high-performance computing to address the pressing challenges faced across many sectors.
The Coalition for National Security Research (CNSR), a broad-based coalition of 74 members (of which CRA is a member) including industry, research universities and institutes, and scientific and professional associations committed to a strong Defense Science and Technology (S&T) Program, released a statement commending the Senate Appropriations Committee for their work on S. 1558, the Fiscal Year 2016 funding bill for the Department of Defense. CNSR was specifically applauding the committee for their support of Defense S&T at the 6.2 applied and 6.2 advanced research accounts, while also raising concerns about 6.1 basic research funding levels. The statement makes the case that Congress should adhere to the “20/20 Principle,” which calls for investment in basic research to comprise 20 percent of the Defense S&T overall budget and Defense S&T to comprise 20 percent of the RDT&E (or Research, Development, Test and Evaluation) budget. The statement points out that, “the 20/20 Principle is based on the recommendations from the National Academies reports Rising Above the Gathering Storm (2007) and the Assessment of Department of Defense Basic Research (2005).”
The Senate bill would provide the following funding for the different defense research accounts:
6.1 Basic Research – $2.31 billion for FY16, which is an increase of $40 million, or 1.7 percent, over what was appropriated for FY15 ($2.28 billion) and $229 million more than what was in the President’s budget request.
6.2 Applied Research – $4.93 billion for FY16, which is an increase of $280 million, or 6.0 percent, over what was appropriated for FY15 ($4.65 billion) and $214 million more than what was in the President’s budget request.
6.3 Advanced Research – $5.58 billion for FY16, which is an increase of $252 million, or 4.7 percent, over what was appropriated for FY15 ($5.33 billion) and $114 million more than what was in the President’s budget request.
DARPA – $2.87 billion for FY16, which is a decrease of $50 million, or -1.7 percent, over what was appropriated for FY15 ($2.92 billion) and $107 million less than what was in the President’s budget request.
The proposed increases in the 6.2 and 6.3 accounts are quite good, as they will keep the budgets above an inflation increase. However, 6.1’s increase will be mitigated somewhat by inflation and DARPA’s decrease will be amplified.
By way of comparison, the House Defense Appropriations bill (H.R. 2685), which was passed by the full chamber in May, has the following numbers for the research accounts:
6.1 Basic Research – $2.10 billion for FY16, which is a decrease of $177 million, or -7.8 percent, over what was appropriated for FY15 ($2.28 billion) and $11.5 thousand more than what was in the President’s budget request.
6.2 Applied Research – $4.84 billion for FY16, which is an increase of $190 million, or 4.1 percent, over what was appropriated for FY15 ($4.65 billion) and $124 million more than what was in the President’s budget request.
6.3 Advanced Research – $5.73 billion for FY16, which is an increase of $409 million, or 7.7 percent, over what was appropriated for FY15 ($5.33 billion) and $271 million more than what was in the President’s budget request.
DARPA – $2.97 billion for FY16, which is an increase of $56 million, or 1.9 percent, over what was appropriated for FY15 ($2.92 billion) and is exactly what the President’s budget request is.
As our readers can tell, the House numbers are better for 6.3, only slightly less good for 6.2, better than the Senate for DARPA (though it will just keep ahead of inflation), but is absolutely horrible for 6.1.
The next step in the process for S. 1588 will be to go before the full Senate for consideration. It’s likely to hit a brick wall there, as Senate Democrats have vowed to filibuster all spending bills unless their priorities are considered in the budget and the ongoing spending debates. The expectation within the science advocacy community is that this will stall the budget process for the rest of the year. Senate Majority Leader Mitch McConnell (R-KY) has vowed to return the Federal budget to regular order (meaning to pass the budget on time), but he has an uphill battle if Democrats stay united on their filibuster. We will continue to monitor the situation and will report any new developments that arise.
The National Science Foundation has released a new public access plan for scientific journal articles that arise from research wholly or partly funded by the agency. This plan, called “Today’s Data, Tomorrow’s Discoveries,” is an outgrowth of an Office of Science & Technology Policy (OSTP) memo, released in February of 2013, which directed, “each Federal agency with over $100 million in annual conduct of research and development expenditures to develop a plan to support increased public access to the results of research funded by the Federal Government.” Let’s look at the details.
The NSF plan is very much in line with the requirements set out in OSTP’s memo. It sets a January 2016 effective date; all grant proposals submitted on or after this date will be subject to the plan. As well, NSF has identified the Department of Energy’s (DOE) PAGES (or Public Access Gateway for Energy and Science) system as the agency’s, “designated repository.” (Note: Most probably the use of DOE’s PAGES is in order to control cost, rather than create a wholly new system just for NSF; part of the instructions in the OSTP memo was that any new system must be implemented and operated within existing budgets). The plan requires that, “either the version of record or the final accepted peer-reviewed manuscript in peer-reviewed scholarly journals and papers in juried conference proceedings or transactions,” must be submitted to the PAGES system within 12 months of publication. All of this is in line with OSTP’s requirements.
There is a section on data management as well, which seems to be the first indications of an Open Data plan. The agency’s plan defines research data as, “the recorded factual material commonly accepted in the scientific community as necessary to validate research findings.” However, NSF writes in their plan that, “in the future, NSF will explore whether all data underlying published findings can be made available at the time of publication.” Anything further is still in the early stages of conception and not covered in this specific plan.
How does NSF’s plan fit into the larger Open Access debate? Seeing as it’s in line with OSTP’s memo, and by extension Administration policy priorities, and Congress has given OSTP room, at least temporarily, to direct science agencies in this matter, it seems the likelihood of push back is minimal. However, that could change with a new bill in Congress, as the legislature does have final oversight. For example, the original draft of the 2014 COMPETES Act in the House Science Committee tried to speed up the adoption of Open Access provisions at the science agencies; that could easily be repeated, as there are still many Open Access supporters in Congress. On the flipside, it is possible that the publisher community could get a repeal bill introduced by a sympathetic member of Congress, removing any such provisions from the agencies. This issue will have to play out more before any final outcome is assured.
IEEE Computer Society, “the world’s leading membership organization dedicated to computer science and technology,” and a member of CRA, is looking for nominations for three prestigious annual awards. The deadline to submit nominations is next Wednesday, July 1st. The details on each award are below. If you would like to nominate someone, please go to the award nominating page on the IEEE-CS’s webpage.
Established in late 1997 in memory of Seymour Cray, the Seymour Cray Award is awarded to recognize innovative contributions to high performance computing systems best exemplify the creative spirit demonstrated by Seymour Cray. The award consists of a crystal memento and honorarium of $10,000. This award requires 3 endorsements.
Established in 1992 by the Board of Governors of the IEEE Computer Society. It honors the memory of the late Dr. Sidney Fernbach, one of the pioneers on the development and application of high performance computers for the solution of large computational problems. The award, which consists of a certificate and a $2,000 honorarium, is presented annually to an individual for “an outstanding contribution in the application of high performance computers using innovative approaches.” This award requires 3 endorsements.
Established in memory of Ken Kennedy, the founder of Rice University’s nationally ranked computer science program and one of the world’s foremost experts on high-performance computing. A certificate and $5,000 honorarium are awarded jointly by the ACM and the IEEE Computer Society for outstanding contributions to programmability or productivity in high-performance computing together with significant community service or mentoring contributions. This award requires 2 endorsements.
On June 10th, the Senate Commerce, Justice, and Science Appropriations Subcommittee approved their Fiscal Year 2016 funding bill, which funds, among other things, the budget for the National Science Foundation. The bill was passed on a bipartisan basis in the subcommittee, with only three votes (out of 30 total) against it. Unlike the House CJS version, this bill doesn’t include burdensome language restricting how NSF can spend their allocated funds. However, the bill does flat fund the agency for FY16. That will translate into a cut, once inflation is taken into account, but given the budget environment in Congress, this outcome is middle of the road.
The bill provides for $7.34 billion for the National Science Foundation for Fiscal Year 2016. That is $430 thousand (yes, thousand) less than what was enacted in FY15, and $380 million less than what the President requested in his February budget request. For the Research and Related Activities (RRA) account, where most of the research funding is located at the Foundation, the committee provides $5.93 billion for FY16, which is the same as it was funded in FY15 and $253 million below the President’s request. For the most part, the budgets are flat funded in real dollars, but these translate into a cut once inflation is factored in. These numbers aren’t terribly surprising; however, they are disappointing since the CJS subcommittee has so many science champions, on both sides of the aisle, as members. Of course, another way of looking at this is that the numbers could be much lower.
However, the bill is not without its good news, and this time it concerns what isn’t in it. None of the burdensome bureaucratic language on accountability, which is in the House CJS bill, is present in its Senate counterpart. As well, there are no stipulations on NSF directorate level funding, thus sparing Geophysical Sciences and Social, Behavioral, and Economic Sciences Directorates cuts to their budgets. Whether the burdensome language gets included in the final bill is something that will have to get worked out in conference between the House and Senate after each passes their bill.
The bill now goes to the full committee for consideration. It is likely to pass, sometime in late June or July, and then move to the Senate floor for consideration. The bill’s future after that is in question, since Senate Democrats have threatened to filibuster all appropriations bills moving forward unless their priorities on federal spending are considered. Congressional Republicans, and Senate Majority Leader McConnell (R-KY), have vowed to move the legislature to “regular order” (i.e. passing the Federal budget bills on time) but, short of removing the filibuster rule in the Senate, there will be little-to-nothing the Republicans can do if the Democrats make good on their threat. Events will have to play out more in order to have a better idea of what the final outcome will be. Continue to follow us at the Policy Blog for future developments.