I was reading the IEEE Spectrum, which is the IEEE general magazine sent to all members. I prefer Spectrum over Computer, the Computer Society (of the IEEE) house publication, because it is aimed at a wide-range of electrical engineers, so the articles explain things better to the uninitiated and try to refrain from the deep mathematical details that turn an entertaining article into a dull, although information packed, and hard to understand article.

The March 2000 issue has a special report U.S.-Russia Nuclear Safekeeping. The whole gist of the report is that Nuclear Arms Control is much worse than at possibly any time during the Cold War. There are several factors for this, most of them being on the Russian side because of their failing economy, although the U.S. has some responsibilities too.

The first article takes a look at Russia's Early Warning System. As it stands, because of financial difficulties as well as the independence of several republics, Russian land based radar coverage has two holes wherein the U.S. could launch an ICBM attack without being detected until the missiles were over Russian soil. There is also a satellite system to supplement the ground system. But satellite coverage has fallen to about 17 hours a day of monitoring U.S. continental missile bases. Russian satellite coverage neglects monitoring the Pacific and Atlantic Oceans, where ballistic missiles would strike from.

One of the problems is that Russia has a 10 minute launch-on-warn scenario. Given that they can detect incoming missiles some 10 minutes from impact, that's how long the Russian machinery has to confirm that this is a real attack, notify and brief the President, get authorization and then launch its missiles. Compare this with the 30 minute launch-on-warn scenario for the U.S., who have much better satellite and radar coverage and so have a earlier warning time.

Both of the above means that the Russian government may be a bit jumpy and might launch a "retaliatory" strike without sufficient information, both because they may not have the information and because they may not have the time to properly check the information. What are the solutions? One possible answer is for the U.S. to help upgrade the Russian Early Warning System. The problem being that the Russians don't trust anything we build and give to them, and we don't trust the Russians to properly use whatever money we give them so that they can upgrade their system themselves.

Another option is a joint missile tracking system, one operated by both sides, although once again the Russians don't trust us. Another suggestion is to provide a slightly cleaned feed of the data NORAD gets, but why should they trust us to provide them with good data?

Another problem is that Russia has some 150 000 nuclear scientists and other personnel who are not being paid regularly. Russian nuclear security of nuclear material is rather spartan. They protect their warheads well, but not reactor fuel and wastes, some of which is weapons-grade, and even the stuff that's not weapons-grade can be refined to weapons-grade with a lot less effort than using natural Uranium. So the U.S. has been funding projects to improve security at nuclear fuel sites, as well as other projects that employ Russian nuclear scientists in their field. Funding for these projects have been reduced this year.

So, after winning the Cold War, we still have a significant nuclear danger. Both frm the current Russian missiles and from other states that can buy Russian nuclear material and expertise to make their own bombs. I'm not particularly worried, as I have a confidence in our various national security organizations. But it is a reminder that the collapse of the Soviet Union doesn't mean that we can stop worrying about our national defense.