Tuesday, January 27, 2009

Science and Policy Bootcamp - Day 2

Today my overwhelming feeling is, "Wow. How does government get anything done?" I've known that for years...but I guess I'd always thought that if I was there, things would change. How typically idealistic of me.

Today's focus was on creative innovation teams, and we started with the Department of Defense. I'll be the first to say that DOD makes me a bit nervous. I have family in the military, and while I do support soldiers that protect our country, it just has never felt right to accept the link between the DOD and science.

But, without the DOD, we wouldn't have interchangable manmade parts, assembly lines, electronics, semi-conductors, aviation, radar...the list goes on. But - we can realize that the DOD doesn't see their role as completely military. Many technologies that came from military funding serve very non-militaristic purposes now, and eschewing the importance of that kind of R&D is just naive.

Even so, the fact that economists like Vernon Ruttan (a University of Minnesota professor) even ask the question, "is war necessary for economic growth?" speaks to the role that military has in innovation. Do we need that kind of driving force, that challenge pervading our everyday work culture to actually accomplish things? The distinguishing factor between the military and the National Science Foundation is that promise of deliverables: "The country will be kept safe." versus the idea of "Yes - we'll learn some stuff." This is also known in the innovation world as utilizing a challenge model. And it works. During World War II, a group of engineers was brought to Los Alamos to assist with the Manhattan Project...but they weren't told why they were there, what they were doing, and only recieved vague information from their physics peers regarding what they were creating. Finally, Oppenheimer decides to override Washington's concerns about security, and tells that this project is attempting to build the atomic bomb, in a race to save human life. Productivity increases overnight, and things get done. But is there a way to do this without a doomsday prediction at your door?

Another point with the military is that it is effectively a "dictatorship." If top command sees the utility of a certain technology, there is instant adoption and implementation...there is effectively a built-in market for this new product. Now--turn this around and put it in the public sphere in the attempted legislation on fluorescent light bulbs and alternative energy. Instead of being praised for early adoption and efficiency, congress is accused of "turning socialist." Wait -- what? It's okay for national security, but not okay for other issues like energy independence...which is, oh shoot. National security.

In total, tthe hree main tenets to spur innovation are the capitol (money and physical equipment), people, and the organizational structure. Thus, the government can technically improve innovation simply by tweaking the manner in which it organizes these groups. The key to this group sort of organization materializes in DARPA (Defense Advanced Research Projects Agency) after the Sputnik embarrassment to American progress in 1958. Instead of dealing with three separate space agencies, Eisenhower decided to take away the space program from the military entities and hand them over to DARPA, until space responsibilities were handed over to NASA in 1960. However, this left DARPA free to focus on computing, and most of the early computer science departments (including MIT until a few years ago) were almost exclusively funded by DARPA.

To clarify - DARPA is not a specific physical place - it's colloquially known as "100 geniuses connected by a travel agent." One of the upsides to DARPA was the relatively flat heirarchy. This is something we talked about a lot today - how having only two levels makes for a much more even playing field, as well as easier means of communication, and a small enough work force to be truly effective. Interestingly enough, DARPA makes a point of contracting with both serious veterans of DOD work as well as newbies, for there is some serious benefit in involving an untrained eye and mind in on a project. (As an aside, it's odd thinking that someday I'm going to cross that border from newbie to veteran).

There was a huge problem during the Cuban Missile Crisis of "command and control" (communication between all branches of the military on a real time scale). One of DARPA's new assignments was to effectively create these connections. Ever heard of the internet? The thing that makes DARPA really great is that they saw the product, and went back to the necessary fundamentals to figure it out: they were interested in breakthrough technology and were okay with high risk operations.

We also talked about several other innovation groups: one of the first was Menlo Park in New Jersey, a research facility bounded by Thomas Edison to create the light bulb. They succeeded---but the more amazing part about the story is that they also made the leap to explaining electron theory, and building the entire infrastructure to power their new invention--utility companies, power plants...the scope of what inventing the light bulb actually involved was incredible. And they made pies together at two in the morning and sang songs. My kind of place.

Do egos ruin innovation? If you can't play well with others, should you get rid of them? There are arguments on both sides, but if your innovation system is based on groups, you're not going to have the same amount of group cooperativity of That is important, but the other option is to manage these particularly difficult scientists. Shockley, a researcher at Bell Labs, became known as the industry jerk for breaking up a collaboration at Bell Labs with semi-conductors, which effectively made Fairfield Semi-conductors and not AT&T the predominant semi-conductor manufactorer.

The story of Genentech was also surprising. Apparently the founders of the company met in a bar to talk about the genetic engineering revolution, and this meeting is in bronze in the lobby of Genentech in San Francisco. When you deal with biotech start up companies, there are a couple of different variables that keep you on your toes. First, the 1980 Bayh Doyle Act makes it much more difficult to give founding members stock options in a start up company (one reason that many scientists head abroad, where these options are still intact). Second, the FDA is a pain. And it takes forever...but what they DO is certify that a drug actually WORKS. That's a huge deal. Pharmaceuticals are the only field in which a product is certified to work correctly. Third, unlike the 25 years it took for internet to go from bench to homes, biologics can go from bench to medicine in eight years, a much smaller time period (which makes biotech startups a more viable platform for angel and venture capital).

A point was made to show the differences between MIT and Harvard: the former of which has a very meritocratic system (no honorary degrees), as well as the presence of immigrants in both the faculty and classrooms. The diversity of thought has always made MIT a very distinctive institution, and it is still unique among research universities for its collaborations between science, engineering, and industry.

The speaker today was David Goldston, the former Staff Director of the House Science Committee. He now writes a column for Nature magazine and teaches a course at Harvard. He spoke of his job on the House floor, as well as the strategy and manipulation involved in his day-to-day life. He is a real people person, and it is essential to his job to network and talk to people, persuading him that science matters.

One of his first points was to differentiate science for policy, and policy for science. Both are science policy, but their goals are completely different.

---Science for policy is the use of scientific discovery and innovation to drive policy-based decisions. For example, science for policy is using data on climate change to inform congress on specific actions that could be taken to alleviate the problem.

---Policy for science is how the government regulates science and controls science-based spending. For example, science for policy is setting the budget for different labs or types of science, as well as mandating the use of stem cells in research.

The thing to keep in mind in terms of science policy is "what does the country need?" And then within those restriction, what are the rules, regulations, and politics which allow for fulfilling this need?

For instance, an example he gave was the standards for lowering ozone concentrations in the air in 1997. According to the science, there is no threshold at which ozone is "safe." The rubric provided is a relationship between the level of ozone and the number of hospital admissions. So, when the bill got to the floor, the debate was on, "How many people can be admitted to the hospital?" No one wants to debate this, and there is no really scientific solution. That's where the policy comes in. And things get complicated.

He also commented on the climate issue, something he considers atypical in DC, for congress is debating a purely scientific question (is climate change real?), there is a majority scientific answer in place (not an opinion...an answer by science), and the consensus in science didn't move the debate at all. The issue is now becoming that since the conservatives spent so much time in denial, they have shoot themselves in the foot in terms of defining the actual policy positions that will come out of this discussion. Even if Congress decides "Yes, climate change is real" they still need to determine how this will affect US policy in terms of energy consumption, biodiversity, environmental sustainability, etc...and that is going to be really challenging.

Another interesting thought to keep in mind is that members of congress are not chosen via merit. They are elected public officials, and may or may not feel that understanding science or knowing US economics is essential to their job. Fortunately, senators do have staff members to help bring them up to speed, but is this good enough?

Frankly, when hearing all of his stories, Congress reminds me of grade school. The interpersonal relationships seem very juvenile, with things changing based on what someone said in the elevator.

Apparently, when you actually visit with congresspeople, you have two jobs: convince them that this will help them become re-elected, and it will also bring money to their district. That made me kind of sad...is that all they think about? And legislation just takes a long time. It is tedious, yet also an critical part of a democracy.

I also asked him how he keeps up with the news: his recommendation for reading materials is the New York Times, the Washington Post, and the Wall Street Journal, as well as online sources (BBC) and NPR. Bills and memos are also available online, which is a resource that hasn't been around for too long, but it makes the process a lot more transparent.

All the better to get more familiar with the legislative process...

No comments: