July 1999 SPECTRUM, on ISS

The US-Russian Space Relationship: Symbolism At Any Cost?

July 1999 SPECTRUM (Institute of Electrical and Electronic Engineers)
http://www.spectrum.ieee.org/spectrum/jul99/features/mir.html

TWO SPACE STATIONS, ONE OLD, ONE NEW, CIRCLE THE EARTH, in orbits deliberately kept as far apart as possible. But the ties that bind the two, Russia's Mir and the multinational International Space Station, which is primarily funded by the United States, are becoming ever more distressingly tangled.

The biggest and most obvious problems have occurred in the station's power and communications systems. A close look at how they were solved, from the component to the system level, reveals how efforts to preserve the unprecedented U.S.-Russian space partnership have engendered equally unprecedented costs, repeated last-minute revisions, and a startling lack of accountability on both sides.

The cast of characters includes the components of the International Space Station (ISS) currently in space or planned for the near future:

Zarya (Russian for Dawn), the 20-ton Russian-built but U.S-financed component of the International Space Station, launched on a Russian rocket 20 November 1998. Also called FGB (for functional cargo block), it is powered by solar arrays and batteries and provides initial control and propulsion for the U.S. node module. The Russian-built Service Module, once launched, will be mated to Zarya and then take over its functions for the lifetime of the ISS. Zarya will then be demoted to serve as storage space and an interconnect structure.
Unity, a U.S. module known as a node, with six attachment ports (one at either end, four around its waist), to which modules launched in the future will be connected. It was carried into space 4 December 1998, aboard the STS-88 Shuttle mission.
The Service Module, recently re-named Zvezda (Russian for Star). This 20-ton centerpiece of the ISS will be launched at the earliest in November 1999. A modified Mir space station module, it is the first Russia-funded component of the ISS and is now almost two years behind in delivery. Its job is to provide long-term life support and control and propulsion functions.
---And, not to be forgotten:
Mir, the bloodied but unbowed champion of Russia's space program. Launched in 1986 on what was supposed to be a five-year mission, it has been continuously manned since September 1989. Mir barely survived a string of catastrophies in 1997, but was revived with new equipment carried up to it on U.S. shuttle missions. Although Russia promised to terminate the Mir program in 1999 lest it distract their efforts from the ISS, some kind of prolongation of its operations appears likely.

In one sense, the funding (or lack thereof) of the Mir gives the most precise picture of the status of the ISS. Its Service Module, without which no long-term crew can survive, is already officially two years behind schedule, due to inadequate Russian funding. In effect, for a year after its late 1998 launch, Zarya will mostly fly unmanned until it is mated with the Service Module.

As a result, the ISS currently aloft will remain unmanned for more than triple the period for which it was designed. It will have to be nursed by remote control by flight controllers at the Korolov Mission Control Center, north of Moscow, and at NASA's facility in Houston.

Even after the crucial Service Module is attached to the ISS, months more might pass while it is outfitted before the long-term three-man crew is able to go aboard--unless, according to a brand-new plan, the crew is sent aloft before those outfitting flights, and makes do with what they find.

The current problem is that the design lifetimes of Zarya's avionics and the amount of rocket propellants aboard were based on schedules that have had to be extended from four to more than 12 months (disregarding the fact that some components were already well into their service lifetime when installed). Although Zarya can be refueled and its avionics boxes replaced during shuttle visits, the recently stretched mission is opening the entire ISS program up to new risks.

As if to prove this very point, the transmitter in the U.S. module broke down early in April. It had been designed to operate for the short interval between the Zarya and Unity launches and the later Service Module launch; and if the original schedule had held (calling for Service Module launch in April), it would have lasted long enough. Fortunately, an extra shuttle mission had been added to the schedule for May, mostly to carry equipment the Russians had been unable to load onto their vehicles, so that the crew was able to fix the U.S. radio.

Pride of place

The ISS now consists of two sections: the U.S.-built Unity node and Zarya, which houses temporary control [Fig. 1]. Zarya is sort of a construction crew trailer on a building site. For essentially U.S. domestic political reasons--something seen again and again in ISS development--the United States paid for Zarya.

The module was "invented" to avoid launching the Russian Service Module first, followed by the U.S. node--which, to ISS planners, would make the station look like a Russian vehicle with subsequently added U.S. appendages. Instead, Zarya was bedecked with a small U.S. flag, declaring itself to be a U.S. module; when the Service Module eventually shows up, the Russians will appear to be in a subsidiary role. (Just to confuse things, the United States does in fact "own" Zarya, having paid for it, yet the Russians do all its remote control and do not even tell NASA its command codes.)

At about the time Zarya reached orbit, in late 1998, the launch schedules agreed upon by NASA and the Russian Space Agency collapsed. The Russian government had cut funding to its own space agency every year through the mid-1990s. When a major economic crisis hit the country in the fall of 1998, the central government not only stopped funding its own space program, but was even taxing all foreign cash contributions, including NASA's.

In effect, the Russian space agency was being required to send to Moscow a cut of the U.S. monies it had received precisely because Moscow had failed to support it well enough in the first place.

Mir's nine lives

Meanwhile, in mid-February, Russia's existing space station Mir plugged away past its 13th year in orbit, and a new crew was launched for a six-month mission. That Soyuz crew comprised a veteran Russian commander, a French astronaut whose government paid cash for his six-month ticket, and a Slovakian "guest cosmonaut" on a week-long visit, whose trip was paid for on credit.

Mir appears to be in adequate health, all things considered. The rash of spectacular breakdowns throughout 1997 [see To Probe Further,] had eased when nine shuttle trips between 1996 and 1998 brought new equipment and supplies. Breakdowns may still be occurring, but without a U.S. presence on board, they probably are just not announced. Occasionally news of problems does leak out. In April, for example, Russia's last geostationary radio relay satellite broke down, cutting off communication with Mir except through a handful of ground stations within Russia.

Still, nobody knows what will happen when this newest Mir expedition ends. The lifetime of Russia's manned Soyuz transport spacecraft reaches its limit sometime in August, when it must return to Earth--with the Mir crew aboard, who do not stay in the station without a Soyuz as a lifeboat. NASA had hoped that the Russians would follow through with their plans to de-orbit the Mir over the South Pacific in August. But that option has become physically impossible: far from letting Mir's orbit decay lower and lower, Russian controllers have been re-boosting it.

Early in June, Russian space officials announced that the current crew would leave the Mir empty when they come back to Earth in August or September. The station would slowly drift lower until early 2000, when it would be steered into the atmosphere over an uninhabited region of the South Pacific.

But this announcement may only be a threat by the Russian space agency so that its own government will provide more money. The agency could still send up a new crew (three teams of cosmonauts are in training for such a mission) if private funding or supplemental Russian government appropriations can be arranged.

The uncertainty frustrates reliable planning for the continued assembly of the ISS, which was supposed to replace the Mir and now must compete with it for meager Russian resources. And paradoxically, the almost bankrupt state of Russia's space establishment has given it a powerful negotiating tool in light of the U.S. investment, political and financial, in the ISS. What is being played out is the old adage: if you owe the bank $5000, the bank owns you, but if you owe the bank $500 million, you own the bank.

Orbits in motion

This past November, as the launches of Zarya and Unity approached, even the relatively simple task of devising orbits for the ISS and Mir stirred up controversy. Russian ground sites cannot rapidly switch back and forth between the frequencies and codes of the two space stations. Russian officials asked that the orbits of the two space stations be designed so that one space station would complete its daily passes over a ground site before the other space station came into that site's range. Practically speaking, during the three years of detailed pre-flight mission design and planning, the Russian orbital experts had insisted that the orbital plane of the ISS be shifted far away from the orbital plane of Mir.

In the final design, this requirement was satisfied by having the points on the equator where each station is heading northbound (the longitude of the ascending node) set 165 degrees apart, with Zarya west of Mir [Fig. 2]. In spherical trigonometry, the two orbital planes intersect at an angle of about 90 degrees--that is, as far apart as they can possibly be. (The circular orbits of the two stations are separated by about 50 km in altitude.)

Then, only a month before launch, the Russians changed their minds. They wanted to transfer equipment from Mir, including perhaps several of its research modules, to the ISS. But this could not be done without another impossibly large 90-degree change in the orbital plane.

NASA experts speaking off the record with IEEE Spectrum suspected that the last-minute demand had more to do with having NASA prolong Mir than with helping to build the ISS. The United States rejected the request. Zarya was launched 20 November into the originally agreed-upon plane, and the assembly of the ISS began.

Since then, the smooth sailing of the International Space Station has had only enough bumps to keep things interesting for ground controllers in both countries. Communications and command sequences have been practiced, and the complicated software interfaces between the two nations' modules have been tested by forwarding signals to each module through the other (until the U.S. radio broke in April).

Coordination between operators in both control centers, in Korolov and Houston, has been polished. There have been the predictable slew of minor anomalies--called "funnies" on NASA documentation--that require some attention and resolution, but there have been no emergencies.

Electrical 'funnies'

On 11 January 1999, the ISS suffered its most serious problem so far: following measurements of dropping voltage, a series of human commands and automated actions in the station resulted in the shutdown of all but the most critical systems, such as the radio link and attitude control. Technically, the situation was manageable, and was eventually isolated and fixed. But both the cause of the problem and the process of solving it are disturbingly symptomatic of the program as a whole.

The trouble began as a result of initially expected operations. In the course of the normal periodic drift in ISS's orbit relative to the Sun, Zarya's solar arrays can operate at less than peak efficiency. During one of these periods, main bus voltage dropped from the nominal 28 V toward an emergency level of 26.5 V, where an automated load shed routine kicks in (load shed is a procedure that shuts down a list of nonessential power users).

Russian controllers noticed the voltage drop, realized that it might soon trigger the load shed routine, and tried to intervene before it started. They sent shut-off commands to heaters and smoke detectors and other noncritical items. But in spite of everything, they could not keep bus voltage high enough; the load shed was activated, and all but the most essential equipment went dead [Fig. 3].

In the course of the next few days, as the station's orbit went through its normal shifting in space, solar illumination and consequent power generation improved. Ground controllers commanded systems back on one by one, restoring the station's normal configuration. They also tried to understand why they were caught flat-footed by the speed of the voltage drop in the main bus.

To start with, they thought that another power problem might have been related. Early on in the flight, Russian controllers had noticed a degradation in performance of Zarya's six nickel-cadmium batteries. The batteries seemed not to be absorbing full charge from the solar arrays, and the problem appeared to be worsening with time.
Deep-discharge cycles, which NiCd batteries need occasionally to maintain peak efficiency, had been planned to take place once a month. But they soon had to be performed more and more frequently, and by mid-January were being cycled every five days.

The 27-kg Russian batteries, each a bit larger than an automobile battery, are rated at 60 Ah. Although designed for a five-year lifetime, they were already three years old when launched, and there was concern that the performance degradation was age-related. NASA's weekly report on 13 January bravely asserted that "the slight decrease in voltage that had been seen is not believed to have been an indication of any mechanical problems." But things were far worse.

Within a week of the automatically tripped load shed on 11 January, NASA engineers began to suspect that the problem lay not with the batteries themselves but with the control circuits that calculated charge levels. The batteries' actual charge was below the calculated charge, one engineer suggested at a weekly status review meeting, "due to premature termination of the charge cycle."

By February, Russian specialists had confirmed NASA's fears: a measurement device on all six batteries called the MIRT, the Russian initials for "integrating ammeter," had a generic design flaw in one circuit. As a result of the flaw, the batteries reported a full charge no matter how low their actual state--and even when it was dangerously low, any further charging from the solar panels was automatically terminated.

Houston flight controllers developed a procedure to "spoof" the MIRT circuits and force full charges on the batteries. They tricked the charge controller circuit into ignoring its erroneous estimate of actual accumulated charge, so that it would stay hooked up to the solar cells for a longer time.

But even with this temporary work-around, the greater concern was over the batteries' lifetime, which was critically affected by how much and how frequently the batteries charged and discharged.

But there was a yet more worrisome aspect of the MIRT flaw, as U.S. experts pointed out. "This problem could have been detected by ground testing prior to flight," one specialist told Spectrum. "But the Russians skipped end-to-end testing--they never put the whole power system through a series of charge-discharge cycles," he continued, attributing the failure in oversight to the lack of time and money.

(In fact, early in Zarya's flight, similar circumstances had led to a different type of electrical failure, traced to another battery controller. The design of the battery called for redundant pass-throughs on a circuit board, but the manufacturer had built only one pass-through. One of the leads broke a few days after the November launch, crippling the circuit that connected the battery to the main bus. The circuit was repaired last December by the STS-88 crew, when they hooked Unity and Zarya together [Fig. 4].)

On manual

After the load shed event on 11 January, another significant oversight in the design of the power system caused trouble. Once the station was returned to its nominal configuration, the ground controllers attempted to reset the load shed routine that had been triggered by the main bus drop below 26.5 V. The routine was still necessary to restore protection during the 10-12 hours a day when the station was out of range of Russian tracking sites.

To their surprise and dismay, ground personnel discovered that there was no ground command to reset the routine. Only an astronaut typing on a keyboard aboard the station could put it back on alert. One manually instigated load shed had in fact occurred before, during the STS-88 flight, a result of still immature coordination between flight controllers in Moscow and Houston. This time, the station was unmanned, which was planned from the very beginning. But the Russian designers had apparently overlooked the need for off-site reset command.

Engineers then realized that the same battery hardware was installed on the Service Module, still on the ground, and had to be replaced and retested. This fairly basic oversight seems to imply seriously inadequate Russian ground testing and system analyses, and raises questions as to whether other undiscovered flaws exist.

The Russians insisted on replacing all six of the MIRT units on the batteries. As with every shuttle flight, the load and task schedule had been prepared long in advance, but a new one was drawn up to accommodate the repair. The timeline of the next shuttle launch coming up, STS-96 (launched on 27 May), was hastily rearranged for the new task, and the units were replaced as soon as the crew came aboard Zarya [Fig. 5].

Test? Launch!

Inadequate ground preparation on the part of Russia's Mission Control Center had been the cause of another recent error, according to French space official Guy Pignolet, who observed a recent Mir experiment from the Russian control center. In the experiment, a thin-film aluminum space mirror, called Znamya, was to be unfurled as part of a program to illuminate regions of the Earth at night with reflected sunlight. In February, as the rotating dispenser unfurled what was supposed to become an aluminum disk 25 meters in diameter, a command was issued to deploy a boom-mounted antenna. The boom extended directly into the space where the disk was deploying; the aluminum wrapped itself around the boom and tore itself into shreds.

After the failure, Vladimir Syromyatnikov, the developer of the mirror, remarked bitterly to a TASS reporter that "Our style of life is responsible--such a complex experiment demands more time, more specialists." When asked why the command to deploy the antenna had not been canceled, he answered, "Because we didn't think of it."

All aboard

In addition to errors in testing, other kinds may arise. The temptation may be growing for the Russians simply to cut corners in any number of areas to speed up the ISS Service Module's long-delayed launch--even though "shortcuts" led to the early-1999 failures of Zarya's electrical power system and to the kind of superficial planning that destroyed the Znamya experiment on Mir. Yet it easily seduces program managers who are obsessed with only the most immediate schedule goals.

Consider the progress of the Service Module, finally assembled and shipped by rail in early May to the Russian launch site at Baykonur, in central Kazakhstan. The Russians were still claiming that the module could be launched by a Proton rocket on 20 September 1999. NASA had prudently adjusted its schedules, expecting a 20 November launch; more realistic officials thought it unlikely to fly before early 2000.

As the clock ticked away and their money was being eaten up, the Russians dispensed with buying flight spares--that is, hardware qualified to replace units that failed in testing. So now the Service Module--the life support of the entire ISS crew--has no backup flight-qualified units for key systems, such as for oxygen generation.

By March, NASA sources were telling Spectrum of a growing desperation to "get it in the air" almost no matter what the equipment's condition, with the hope that the inevitable breakdowns could be repaired on later shuttle flights. Space experts with long memories have told Spectrum that this obsession with sticking to a schedule by overlooking adequate pre-flight testing is frighteningly reminiscent of the push to launch the doomed Challenger shuttle in January 1986. And even if systems do not fail catastrophically, in the long run it is hundreds of times cheaper and easier to find and fix problems on the ground than it is in space.

Just as imprudent as those pushing to launch any hardware at all, seemingly, were those who wanted a manned presence as soon as possible. After the shuttle mission to the ISS in May (the one that carried the MIRT battery replacements), NASA had expected three more shuttle flights and one Russian supply flight before sending up a crew aboard a Soyuz. The four flights were to add equipment for power generation as well as spare parts and backup hardware for critical life support systems.

Instead, responding to the prolonged delays, Russian and U.S. space officials developed a new plan: sending up the three-man crew to the Service Module as soon as it reaches orbit, without waiting for the four preparatory missions. Thus, ISS's first long-term crew--U.S. astronaut Bill Shepherd and Russian cosmonauts Yuri Gidzenko and Sergey Krikalyov--would be aboard a module before the Zarya/Unity complex begins its automated approach and docking to it. The major advantage of this option is that the crew could provide manual backup to the automated linkup if necessary.

But by going to the Service Module so early, the crew would be exposed to the risk of not having the backup systems that would seem mandatory. What's more, it would have to rely on the systems in the Service Module, whose pre-flight verification is likely to have been even less thorough than those that so clearly failed for Zarya and the space mirror. But if worse came to worst, and enough time were available, the crew could abandon ship, flying home aboard the docked Soyuz.

The Soyuz shell game

With the prospect of prolonging the lifetime of the worn Mir space station, plans for the ISS become even more convoluted. If there is to be a next Mir crew, it is to be launched in August, with two fresh cosmonauts. Only one Soyuz will be available, and that one--Soyuz 204--is now being completed in part with U.S. money provided last fall to the Russian space agency.

Once built, the Soyuz-204 is to be allocated to the ISS's first manned expedition with Bill Shepherd and his crew. That launch date, with the ISS's so-called early crew option--and if the Service Module is actually launched on schedule--would be some time in October 1999. Recall that the Soyuz would stay there, as a lifeboat in a space station that is already low on flight spares.

So much for October. The Russians, however, provisionally lined up their own mission to Mir for two months earlier, in August--without ordering another Soyuz from the factory. Now, it takes the Russians 18-24 months of fully funded work to produce a Soyuz. Any next-in-line Soyuz after the 204 would not be ready until February 2000, and then only if NASA hands over more money.

But the Russians wanted a summer launch for the Mir as well as a fall launch for the ISS. So, for the first launch, they finessed the problem: a Soyuz in hand, or almost in hand, is worth more than one in the bush. The Soyuz 204, it was reported in Russia, would go to Mir.

Recall that the 204 is partly paid for by the United States expressly to keep Russia from further delaying the ISS schedule. Yet the next-in-line Soyuz surely will not be ready by the ISS-crew launch date of October. So the ISS schedule has been delayed more, rather than less. In addition--and more maddening, to some people--U.S. funds essentially were diverted to further the purely Russian interests in the Mir project.

Such a shell game might be expected to generate some heat in the United States, and in fact one NASA source told Spectrum that "NASA would not look kindly" on any Russian attempt to divert Soyuz-204 from ISS to Mir. But these are words of the diplomat. In fact, explained a congressional source close to the project, everything is legal and correct. In its latest contracts with the Russian space agency (RSA), NASA carefully avoided specifying how U.S. money would be spent.

According to the congressional insider, "NASA conceded that [the Mir mission] is one of the things it expected RSA to use the money for." Nonetheless, Moscow's announcement in early June that no crew would be sent to Mir to replace the current one may have allowed it to dodge a major confrontation with NASA.

Triple production

Russian commitments to other components of the ISS program also seem to be built on air. In 2000, NASA's flight plans in support of the ISS call for a total of 10 Russian launchings: two of the manned Soyuz shuttle and six of the unmanned Progress shuttle (a modified Soyuz that is used for one-way trips), and two more modified ISS modules based on Soyuz designs.

In reality, as it has kept the Mir program afloat over the past few years, Russia has been able to annually build and launch only about half that number. To satisfy its commitments to the United States, even if it pulls the plug on Mir this month, Russia must double its spacecraft production rate in less than a year. If it wants to keep Mir, its annual spacecraft production rate must triple.

Shortly before he resigned in April, Randy Brinkley, NASA's space station program manager, was asked by Spectrum whether he believed Russia was capable of that flight rate next year. He answered softly and simply, "No."

Official production records from the Progress plant in Samara in the Volga region, which builds the booster rockets for the Soyuz and Progress, confirm Brinkley's skepticism. Russian plans show 18 rockets scheduled for delivery in 2000. Apparently it did not seem worthwhile even to cover up the evidence: only four rockets are allocated to missions for the ISS, and none to Mir. The others are for commercial customers or Russian Ministry of Defense missions.

Clearly the over-ambitious Russian promises of 10 flights--or even 14, if Mir is prolonged--are either delusional or prevaricating. And unless NASA also is delusional or prevaricating, if ISS plans are not severely modified, more delays will catch NASA "by surprise" next year.

Wishful thinking

Without constant double and triple mortgages, so to speak, the Russians could barely provide the manpower, material, and services needed by the ISS outpost. Yet somehow they still want to be a two-house family, holding onto Mir and its $250 million-a-year operating budget. Many in Russia still hope to find that necessary funding for Mir from "off budget" sources.

In late January, then-Prime Minister Yevgeny Primakov authorized the Energia Rocket and Space Corp., Moscow, which owns and operates Mir, to begin soliciting money from private sources to completely finance Mir. Nongovernment funding was nothing new for Mir. Russia began selling seats on manned space missions to the Salyut-7 space station in the early 1980s and by the mid-'90s was earning between $50 million and $100 million per year just from European space organizations eager to fly astronauts aboard Mir.

In January of this year, just when pulling the plug on Mir seemed a done deed, Russian space officials began talking up a "secret foreign investor" who would furnish the Mir's entire operating budget. In return, the mysterious investor would be given in-flight cosmonaut man-hours for research and other activities on Mir.

Speculation was rife about the hoped-for saviors of Mir: a reticent Australian millionaire, the decidedly unreticent billionaire and ex-U.S. presidential candidate Ross Perot, or even the Chinese space program, which wanted a docking site for its planned two-man space capsules. Some pocket money was supposed to come from a film company shooting scenes aboard Mir--a spinoff from earlier deals with advertisers.

But by early February, hopes for financial salvation outside government sources had collapsed. As Yuri Koptev, head of the Russian Space Agency, told newsmen, "Unfortunately, our lives are such that we sometimes consider the desirable already a reality. It was just wishful thinking."
Spectrum has learned that the potential financiers did exist: a group of U.S. businessmen in Florida that had intended to line up commercial users willing to buy time aboard the born-again Mir. But Mir's capabilities fell short of many of their customers' requirements--for instance, less than 10 kW is available for experiments, and even then the power is not assured. After they had looked deeper into the deal, most of the potential investors passed on it.

Then in May, some Russian officials again placed their bets on the British industrialist who supposedly would pay for a ride on Mir next August, and who they thought could be interested in some commercial arrangement. But this scheme also soon collapsed. Finally, Russian space officials asserted they would keep Mir going even with bank loans, if necessary.

The long goodbye

Until now, space stations have had limited lifetimes and died more or less gracefully [see "To dust they shall return"]. In the early '70s, NASA's Skylab and the early Soviet Salyut stations carried supplies for less than a year of operations. Later Salyuts could be resupplied and thus could operate for up to three or four years. And Mir, designed for a five-year lifetime, has just passed 13 years.

Although by April 1999 the Russians were supposed to have made a "final decision" about deorbiting Mir, they were still boosting its orbit. This was a clear signal that they intended to prolong its life well beyond the official termination date, though they had not had not told NASA officially of the plan.

Why this need to prolong Mir, and by so doing wrench apart even more its commitment to its international partners? Perhaps the Russians realize that the longer NASA is kept in doubt about Mir's fate, hoping that it will be terminated as promised, the longer the flow of U.S. money once earmarked for its successor can be continued.

Despite these difficulties, officials in both countries are committed to seeing the ISS through. Russian Space Agency director Yuri Koptev has repeatedly warned his countrymen that Russian withdrawal from its ISS commitments would mean cancellation of Western commercial space contracts, now approaching a billion dollars a year. For its part, the White House still sees the partnership as central to its policy toward Russia, and repeated efforts in Congress to expel it from the ISS grand plan have been roundly voted down.

As the ISS completes its first months in orbit, and Mir completes what may be its last months, the political and technological sparring in both programs has underscored an old lesson: whatever space experts plan and attempt, reality still turns up surprises. The only reliable prediction is that a year from now these projects will look nothing like today's expectations.

To probe further

Up-to-date information on developments in the International Space Station (ISS) and the Mir programs can be obtained from a number of specialized Internet sites. The official NASA view is reflected on the World Wide Web at http://spaceflight.NASA.gov/index-m.html. The best news media compilation of reports is Florida Today's Space Today, on the Web at www.flatoday.com/space/today/index.htm.
The most respected private Web site covering space technology and politics is Keith Cowing's "NASA Watch" at www.reston.com/NASA/watch.html. The site's jumping-off spot for the ISS is www.reston.com/NASA/station.news.html.
Unique insights into the people who planned and performed NASA's Shuttle-Mir program in 1995-98 can be found in Bryan Burrough's Dragonfly: NASA and the Crisis aboard Mir (HarperCollins, New York, 1998). Although riddled with minor factual flaws, its views of the personalities involved are on target.
On the Russian space industry as a whole, see the author's "Russia's space program: running on empty," IEEE Spectrum, December 1995, pp. 18-35. See also his article, "Shuttle-Mir's lessons for the International Space Station," Spectrum, June 1998, pp. 28-37.

About the author
James Oberg, a veteran of 22 years at NASA's Mission Control center in Houston, is now a full-time writer and a consultant to ABC News. His most recent book, Space Power Theory, was released by the U.S. Space Command in April.