acm-header
Sign In

Communications of the ACM

Viewpoint

Science, Policy, and Service


Science, Policy, and Service, illustration

Credit: Nextgov

For many computer scientists, the thrill of impactful technical inventions and fast-moving innovation is what pulled us into this field originally and is what keeps us here. But as computer systems impact society ever more deeply, questions arise regarding the societal implications of the technologies we are building, and from there the conversation often shifts to questions of policy and regulation. When it comes to privacy, security, and other crucial issues, what should be expected of these computer systems that are increasingly ubiquitous? How should computer systems be managed and regulated? Just as importantly, who should make those decisions? How will policy experts grapple with developing regulatory and governance decisions about the deeply complex technologies we are developing, and who will help them understand what they need to know?

From its roots supporting trajectory calculations and cryptography during the wars in the first half of the 20th century, computing technology has been deeply intertwined with policy and government issues. But the societal and personal impact of our inventions is becoming more and more apparent and urgent. Artificial intelligence and machine learning are now integral to a profound array of real-world applications, from autonomous vehicles to law enforcement. Internet of Things (IoT) devices and systems are being built and sold that affect human health and safety in myriad ways; they control the electrical grid, manage transportation, meter out medicine dosages, and control the front-door locks on houses. More than ever before, CS as a field faces deep responsibility for creating algorithms, devices, and systems that operate in a manner that is reliable, secure, ethical, and fair.

Simultaneously, the deceleration of Moore's Law and Dennard scaling mean that the underlying semiconductor improvements supporting our innovations are slowing and this scaling loss is leading to massive top-to-bottom shifts in how software and systems are designed, programmed, and operated. The low-level system designer's response has been to supplant software with increased use of specialized hardware accelerators for cryptographic routines, image analysis, and so forth. These hardware-accelerator-oriented approaches offer a viable path forward in the short-term, but are not without their challenges. Namely, although both hardware and software are susceptible to bugs and vulnerabilities, the bugs and vulnerabilities our systems "bake into" hardware are much more difficult, expensive, and time-consuming to fix.

uf1.jpg
Figure. U.S. Department of State headquarters in Washington, D.C., USA.

So here we are, in a world where computing devices and systems have more societal impact than ever before, and yet they are simultaneously more difficult to test, debug, patch, or even understand than ever before. And in the midst of that, policymakers are—often with relatively little technical background—deciding what to do ...

From August 2015 to April 2017, I was a Jefferson Science Fellow (JSF) engaged in computing and communications policy within the United States Department of State. For the first 12 months, I lived in Washington, D.C., and worked in-person at the Office of International Communications and Information Policy in the Economics Bureau at the State Department. After I returned to my university position in August 2016, I continued to work remotely on these issues for an additional eight months. The Jefferson Science Fellow program was established in 2003, acknowledging the strong and widespread impact of science and technology trends on foreign policy issues, and with the goal of augmenting the State Department's in-house science and technology expertise by bringing in tenured science and engineering professors for one-year fellowships. It is one of many efforts across the U.S. government to include scientists and technologists in the conversations regarding how "our" topics shape our world and therefore may require technical inputs for their policy and governance.

Sometimes when I mention to computer scientists that I worked at the State Department, they react in surprise, because it does not seem like a tech-heavy arm of our government. But as many news stories over recent years have highlighted, when foreign policy decisions involve topics like data privacy, security, encryption, Internet censorship, or other charged scientific topics like climate change, most of us would agree it is preferable to have those decisions guided by sound science and technological truth wherever possible.


Both domestically and abroad, computing technology is viewed in terms of both societal and economical benefits.


So what did I do as a JSF? I followed a range of computing and communications issues for my group, such as Internet of Things (IoT), smart cities, blockchain, and financial technologies. For these and other broader topics (computer security and privacy) I interpreted and explained the technology behind the topic, and I contributed to broader governmental and multistakeholder processes to formulate U.S. positions on the issues. I also participated in international meetings where these issues were discussed, and I spoke for the U.S. position on these issues, often negotiating with other countries whose beliefs in topics like Internet freedom differ strongly from our own.

Both domestically and abroad, computing technology is viewed in terms of both societal and economic benefits. Internet connectivity catalyzes better education and healthcare, and it also supports economic growth and entre-preneurship. Likewise, issues like harmonizing the communication spectrum are also important worldwide, as they fundamentally define our ability to use cellphones globally. But from the Internet writ large, to cellphones, to tiny IoT devices, different stakeholders can have widely disparate opinions about how they should be regulated and standardized. In my experience, I saw countries and companies pushing to develop broad technical standards requiring heavyweight cost accounting methods built into tiny IoT devices, while others preferred lighter weight methods that would avoid these. I also saw proposals of network protocols that encouraged inspection of packet contents—with considerable privacy and security implications—that concerned other meeting participants. These differences of opinion need to be negotiated; without some broader agreement, computer systems will lack the interoperability and ubiquity that makes them so vital in our world today.

On a nearly daily basis, a range of computing and telecommunications and digital economy issues are being discussed in multilateral forums including United Nations agencies, in the International Telecommunications Union (ITU), in the Organization for Economic Cooperation and Development (OECD), in the G7, and in other similar dialogs. Most researchers are surprised when I describe the frequency of intergovernmental policy meetings on computing topics. One of my goals as a JSF was to help ensure the U.S. policy approaches at these forums were technically sound and viable. My colleagues and I also advocated for approaches and meeting outcomes consistent with U.S. values regarding an open and interoperable Internet governed through multistakeholder processes, and a level playing field for global tech innovation. In addition to learning a great deal about technology and policy, the time I spent representing my country's values at international meetings also made—at a personal level—for a deeply fulfilling and patriotic year.

Although my JSF immersed me in international policy arenas, computing policy permeates many arenas across government, both domestic and broader. The technical and funding relationships between CS researchers and some agencies (Department of Defense, Department of Homeland Security (DHS), National Institute of Standards and Technology (NIST)) are familiar and long term. In addition, other agencies play increasing roles both in terms of funding and in terms of broader policy engagement. For example, the Department of Transportation has large projects related to smart cities and connected vehicles that will benefit greatly from CS input. Another example is increasing inter-agency attention on artificial intelligence policy issues over the past two years, including both technical aspects of trust and fairness, as well as broader issues of impact on the economy and labor markets. A further example is that as a country, we need to decide how to ensure a broad and diverse cross-section of our K–12 student population has access to computing education and affordable in-home broadband Internet access. Given the ubiquity of computing in society and in our economy, there is a corresponding ubiquity of computing-related policy questions in our government. A fundamental issue is: Who will be answering these questions?

Three years ago when I decided to join the State Department, I found it a very hopeful sign that I and many other science and technology researchers were working in Washington either as JSFs or "on detail" within the Office of Science and Technology Policy (OSTP), or as IEEE or AAAS fellows across government.

Now I am much less optimistic. The position of Science Adviser to the President has not been filled and much of OSTP is quite depleted overall. The State Department's reorganization and hiring freeze has left many leadership positions unfilled. This includes the State Department position of Science and Technology Adviser to the Secretary (STAS), which became vacant in July 2017; the Department's leadership has not committed to refilling it. The status of the JSF program in which I participated is also unclear. The program continues but with fewer than half as many fellows, and with placements only at USAID, not at State. More broadly, the news is filled with stories about the diminishment of science within offices at EPA, NOAA, and other agencies.

Talking with colleagues right now about government engagement, I am met with a wide range of responses. Immediately after the U.S. 2016 election and 2017 inauguration, a conversation about my role as a JSF would often be met with "How could you continue given the new administration ... ?". On the other hand, the change of U.S. administration has not lessened the impact of computing on today's world, and therefore the urgency of computing policy discussions has not decreased either. In terms of specific technical policy topics (like spectrum coordination) the U.S. position has not changed much. Some see (as I often was able to) that as long as one is being listened to, and as long as one is not being asked to go against ethics or beliefs, then having sensible tech-savvy people continue to engage with any administration is a good thing.


If technologists do not speak up to educate policymakers, bad policy will result, and we will have to live with it.


When I ended my service in April 2017, it was four months earlier than my original one-year renewal for remote work, but six months after election day. In April, my office at the State Department included many of the same capable and devoted civil servants, covering many of the same issues, as one year earlier before the election. So, my decision to end my service early was, to my considerable surprise, more about how best to have effective impact than about the presidential administration per se, despite my disagreements with it. In essence, my decision was that there is a wide range of impactful and much-needed ways to help steer our policymakers in the right direction. Mailing back my badge in April was not the end of my work in that regard, but just the finalization of a decision to find other ways to do so.

How can computer scientists help shape policy for the better and ensure maximum societal benefit from our breakthroughs in computer science? First and foremost, there are many important policy issues—IoT security, network neutrality, trustworthy and fair AI, drones, and more—where computing practitioners and researchers can and should weigh in from the technical side. In the U.S., the FCC, NTIA, NIST, and others are often inviting stakeholders and the general public to participate in policymaking processes on these and other topics. Just in the past month, I have cringed at statements I have seen (on policy-oriented public email lists and in public meetings) made by policymakers with unclear appreciation for how security patches might be applied to IoT systems, or for how AI algorithms work, and for where opacity or unfairness might lead to undesirable results. If technologists do not speak up to educate policymakers, bad policy will result, and we will all have to live with it, as people and as technologists. There are also many lawmakers in the U.S. and elsewhere who welcome input on technical issues; working with them and their staff to explain technical issues and their implications is another way to be more engaged. In addition to issues being covered at the U.S. federal level, there are also state and local issues around data privacy, CS education, and other topics that are of critical importance. Finally, working with the press—both technical and general—is another avenue for our knowledge to be parlayed into broader understanding and action.

I particularly implore my academic colleagues to speak. While larger tech companies have dozens of public policy advocates engaged full-time in these conversations, the longer-term and more company-neutral technical viewpoints from academia must be heard as well. This includes advocacy for research and research funding, of course. Importantly though, it also goes well beyond funding. The academic CS community must commit time and energy to advocacy for tech policy paths that are technically rational, that benefit society, and that are decoupled from the motives of a particular company or its shareholders.

At this moment in time, with computing central to our world and yet science often being undervalued or even denigrated by some policymakers, computer scientists must continue to pursue ways and expand our efforts to share our technical knowledge for the good of society. Think of one way to share your technical knowledge with the broader public or government today, and continue with these steps each week. Eventually we will either have governance and policymaking that does not make a computer scientist cringe, or we will at least know that it is not for lack of our attention.

Back to Top

Author

Margaret Martonosi ([email protected]) is an ACM Fellow, the H.T. Adams '35 Professor of Computer Science and the Director of the Keller Center for Innovation in Engineering Education at Princeton University, Princeton, NJ, USA.


Copyright held by author.
Request permission to (re)publish from the owner/author

The Digital Library is published by the Association for Computing Machinery. Copyright © 2018 ACM, Inc.


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account
Article Contents: