Over the past few decades NASA and other organizations have attempted to design satellites that would be capable of carrying out multiple missions. The fundamental idea is to create a single spacecraft design that could be duplicated over and over in order to reduce unit costs while satisfying several space missions and applications. To date, the results have been disappointing.

The spacecraft bus tends to become overly complicated, excessively heavy and too costly for most applications. The underlying problem with this approach is the one-of-a-kind market demand that has plagued the satellite community since its beginnings. Low production and flight numbers just do not permit realization of scaling and production advantages that are common in mass production industries.

It has taken five decades to arrive at the conclusion that most missions require unique design features which effectively precluding the use of a one-size-fits-all design.

While it is true that Iridium, GPS and a few other programs have been able to produce large numbers of satellite clones, the majority of satellite purchases involve very few flight vehicles. Fortunately, the situation is not "black and white." There are a good number of "standard" spacecraft buses that are marketed by the satellite manufacturers.

For example, the geostationary Earth orbit (GEO) communications satellite market amenable to some standardization of the spacecraft bus elements and subsystems.

Most GEOs require an Earth-pointing attitude control system, power collected from sun-oriented arrays and stationkeeping requirements that are similar for all equatorial positions. All payloads can be mounted on the Earth-pointing face, and so on. This situation is uniquely convenient for manufacturers of GEO communications birds.

However, standardization is not easily achieved for low-Earth-orbiting (LEO) satellites. The variety of applications and missions requires special satellite designs in almost all cases. The only situation in which a single design can be effective in terms of cost, mass and complexity is one in which many cloned satellites are needed for a unique application.

Although two examples were cited above, the next new opportunity may be a distributed space infrastructure that could replace large monoliths with constellations of small satellites. This idea has been around for some years but has not been tested for a number of technical and political reasons.

However, recent monolithic program shortcomings in combination with advances in small spacecraft technologies and computational techniques have encouraged a refocusing of efforts toward the possible future adoption of distributed space systems.

It is possible that one day soon many of today's space applications will be accomplished using large numbers of inexpensive small spacecraft, all working together to provide functions that are currently the purview of billion-dollar big birds.