Teledesic was the first company to plan to offer broadband connectivity using a constellation of low-earth-orbit (LEO) satellites. Craig McCaw, who had sold McCaw Cellular to AT&T, founded Teledesic in 1990 and it got a big visibility and credibility boost when Bill Gates made a small ($5 million) investment in the company.

McCaw and Gates were able to attract capital—$200 million from a Saudi Prince, $750 million from Motorola, and $100 million from Boeing, which signed on as the prime contractor. When Boeing and Teledesic finished the final design, the constellation had been reduced from the originally planned 840 to 288 satellites. (Later, Motorola replaced Boeing as prime contractor). The FCC approved Teledesic’s Ka-band spectrum application in March 1997 and 37 counties submitted supporting proposals for the December 1997 World Radiocommunication Conference. Teledesic hoped to provide “fiber-like” connectivity to an “Internet in the sky,” but was unable to deliver and gave up in 2002.

I don’t know what motivated Gates’ investment in Teledesic, but today the Gates Foundation is devoted to fighting poverty and providing health care and education in developing nations. Nearly 20 years after the demise of Teledesic, satellite, launch, and communication technology are vastly improved, the entire world is aware of the Internet, we have applications that can utilize “fiber-like” speed and latency, and Gates is clearly aware of the value (and downside) of connecting the unconnected.

Bill Gates might be thinking that it is time for another try.

Last September, Microsoft announced that customers of Viasat, Intelsat, and SES would be able to access Azure cloud services. Their focus is on government, enterprise, maritime, and airline applications and the announcement states that “each of the partners brings different strengths, for example, choices between Geostationary (GEO), Medium Earth Orbit (MEO) and, in the future, Low Earth Orbit(LEO) satellites” so it seems they are talking with possible LEO partners. (Maybe not with Amazon given its recent challenge to Microsoft’s JEDI defense contract).

Earlier this month, the FCC authorized Microsoft to establish a proof-of-concept connection between two ground stations in Washington and DEIMOS-2, a Spanish imaging satellite. If successful, the test will demonstrate satellite connectivity to Microsoft’s Azure cloud services as well as the rest of the Internet. They plan to run the demonstrations before, during, and after the Ignite conference, which starts on Sept. 22, and if the demonstration results in significant market interest, they will apply for regular ground-station authority, which would put them in direct competition with Amazon’s ground-station service. (Microsoft may have a fear of missing out on space).

Terminals with electronically steerable antennas are a critical LEO broadband component. High-end fixed and mobile users will be able to justify relatively expensive terminals, but success in the consumer market will require user-installed, reliable, low-cost terminals. It turns out that Bill Gates was the lead investor in electronically-steerable antenna manufacturer Kymeta at the time of its launch in 2012 and he is now leading a new $85 million investment round in support of a new high-end mobile service using Kymeta’s new LEO-ready U8 terminal. The expensive U8 is sold for high-end fixed and mobile applications today, but they will surely be able to produce a low-cost fixed-service terminal in the future.

If the LEO broadband business case turns out to be viable, these are early days and there is room for competitors. The Gates Foundation endowment is nearly $50 billion, Bill Gate’s net worth is $115 billion and Microsoft is on a roll. Might we see another broadband LEO constellation?