Campbell Corner Language Exchange

Panic in Proportion: The Case for "Open Source" Science
by Shuba Gopal

>What do you think about the hubbub over public info for the generation
>of epidemic causing viruses? Is this a real issue, or some overblown
>reportage to sell newspapers?

I think it is a semi-real issue in that it is a genuine concern that we may hand terrorists some terrible weapon by not being aware of the consequences of the research we do. On the other hand, I think the newspapers, and our beloved John Ashcroft, are far and away the bigger culprits here.

First, let me try to put things in context. If I remember correctly, this has to do with the publication a few months ago of the results from a genetic modification of the influenza virus which created a
"super-virus" capable of somewhere in the vicinity of 90-100% mortality rates in mice. Essentially, the researchers identified a way to switch off a key element in the immune response to influenza,
and by extension, to many other viral infections. After publication of the work, they were bombarded with emails and government censure for having published such "sensitive" work in a public, peer-reviewed journal. As I recall, there were harangues even within the scientific
community about how this information would now be exploited by Al Qaeda and other terrorist networks to effect horrifying death tolls on civilian targets.

All well and good. However, what I think many people missed was the point of the paper, which is that an immune response can be switched to a highly effective clearance of virus by the ACTIVATION of this protein. In other words, it was the absence of the protein that caused the high mortality, and turning it back on reduced mortality back to expected levels of
somewhere around 10-20%. So now we have two pieces of information - removing the protein creates killer viruses, a tool that could become a terrifying weapon. But we also know what is involved in resurrecting the immune response. In subsequent papers this summer, researchers have already shown preliminary reductions in mortality rates by switching on other immune system proteins that can counteract the devastating effects of the loss of the original protein. So science has already found ways to counterbalance the potential threat.

I must digress at this point to comment that the mind boggles at the average Al Qaeda terrorist reading these highly technical scientific journals in the first place. And then, hiding in a dusty cave in some remote region of central Asia, one wonders who would have access to a molecular biology laboratory! This is not the sort of work you throw together with a couple of sticks of dynamite or some box cutters. It costs millions to set up and operate a molecular biology laboratory of
the scale necessary to even begin modifying the influenza virus. Sometimes we forget that the NIH annual budget is well over $3 billion, and that just because I can buy a PCR machine and get some
influenza virus doesn't mean that I can then make a killer virus before going home to supper with the kids and goats in my friendly Afghan (or Iraqi, North Korean, Russian, Chinese, etc) neighborhood. So this is the sort of neurotic over-hyping that degrades the American press and turns us all into morbid curiosity mongers.

However, I think the larger question - how much should we make public and how much should we keep secret in these supposedly more dangerous times - is a valid one. I would argue for as much public knowledge as possible, and I would draw on two examples from a different area of
technology, the computer industry.

About a decade ago, enthusiasts of encryption algorithms fought a long and rather bitter battle with the US government to allow public use of encryption methods that had heretofore been the exclusive province of secretive government agencies. Many would argue that it was the
release of these algorithms into the general public around 1994/1995 that fueled the boom of the internet, which for all its inherent problems, has in some small way contributed to an ongoing shift in the way we interact with the world. During the long court battle to release encryption algorithms to the public, government officials argued that the release of the encryption algorithms would allow
terrorists and criminals to effectively bypass any sort of monitoring and free them to act with impunity. In the past eight years, I think there have been fewer than a dozen cases of criminals hiding data behind encryption, and of those, all but one that I know of were caught, tried and sentenced in spite of their efforts to conceal their activities. So it turns out that we gave the criminals a little bit too much credit in this instance.

What releasing encryption to the public domain did do, however, was suddenly expose a number of serious security issues that the government agencies were horrified to discover had existed all along. Why was this? When you open an idea or a technology to extensive scrutiny by a multitude
of people demanding a variety of tasks from a piece of software, you immediately expose all the problems with that piece of software. A collective investigation and improvement effort has now yielded security software that is considerably stronger than it was when it was developed
in the shadowy cradle of government protection.

I would argue that "open source" as such publicly edited software is known, yields stronger, better programs in every instance. Another case in point is the tremendous stability and success of the Linux operating system when compared to privately developed, protected operating systems such as Windows. Linux is rapidly taking over in areas where stability, security and reliability are critical. For instance, a recent survey found that upwards of 60% of all the world's internet servers are running Linux or a related operating system. The stability and security features of Linux are a consequence of its open source development. Individuals from across the planet contribute to
its development and often anticipate problems prior to their appearance. This has led to a truly powerful, incredibly efficient operating system. It has the added benefit that, since it was developed by everyone, few feel disenfranchised by the system. Linux has only one known virus and the security hole it exploited has long since disappeared. How many other operating systems currently on the market can claim that?

I believe that, just as the computer industry thrives in the open source environment, so does science. One cannot do good science, or even any science, in isolation. It is in the discourse and
communication between many individuals, through dialogue which under the best circumstances at least parallels Socratic dialogue, that science progresses. One mind, no matter how great, cannot conceive of an entire system on its own. Even Einstein needed help - he needed Planck's observations to solve the photoelectric effect and he needed others, who came afterwards, to provide experimental proof of some of his predictions. Scientists are fond of the cliche "standing on the shoulders of giants" when they describe their work; we cannot stand on nonexistent shoulders and we cannot climb past those shoulders if we don't know where to go next.

If anything, I would argue that the recent papers on modifications to viruses are critical if we are to prepare for a bioterrorist attack. They are even more critical if we are to find a way to assuage the
terrors of the world as they exist today - a world in which millions die from disease, famine and state-sponsored terror every day. Let us not lose sight of the real problems here in a temporary hysteria over a world that is as fraught with terror, misery and inequality today as it was just a little over a year ago.