In the span of one week, two space startups that have made good on far-fetched promises unveiled plans to develop competing global satellite Internet systems.

Whether there is room for two such multibillion-dollar constellations is unclear, but the announcements by Space Exploration Technologies (SpaceX) founder Elon Musk and OneWeb Ltd. – a company financed by Virgin Group, chip-maker Qualcomm and O3b Networks founder Greg Wyler – are backed by people who have demonstrated the ability to attract both capital and engineering talent.

On the one hand is Musk, who announced plans last week to build a spacecraft production plant in Seattle, an effort to transform satellite manufacturing in much the same way he has the space-launch business. Less than a week later, SpaceX announced a $1-billion round of financing with two new investors, Google and Fidelity, which will collectively own just under 10% of the company.

In a Jan. 20 statement, SpaceX said the money would pay for continued innovation in space transport, reusability, and satellite manufacturing. 

“In order for us to really revolutionize space, we have to address both satellites and rockets,” Musk said Jan. 16 during a Seattle event where he unveiled his plans. “We’re going to start by building our own constellation of satellites, but that same satellite bus and the technology we develop can also be used for Earth science and space science, as well as other potential applications that others may have.”

Musk said the satellite plant would start small—about 60 people—and that engineers could move between SpaceX’s core business of rocket manufacturing in Hawthorne, California, and the new satellite venture in Seattle. He said the project would take 12-15 years to complete and cost $10-15 billion to build.

He also said the new company would not be drawing from the Puget Sound’s established space propulsion community to supply the vehicles.

“We’re going to build our own propulsion unit,” he said, adding that the constellation of 4,000 broadband satellites weighing a few hundred kilograms each and orbiting at 1,100-km (683-mi.) altitude will be powered using all-electric Hall- effect thrusters, a technology he says is relatively easy.

In the meantime, OneWeb, formerly WorldVu Satellites based in Britain’s Channel Islands, has announced plans for a startup satellite broadband venture. Unveiled Jan. 15, the new company is to be lead by Wyler, who founded O3b Networks, a constellation of 12 Ka-band broadband satellites that provides Internet trunking to telecom companies and corporate and government customers globally in a band around the equator.

Wyler, who is no longer directly affiliated with O3b operations, has said the amount of overlap between OneWeb and O3b is minimal given the latter’s focus on large telecom companies. He says the OneWeb satellite system would introduce the first-ever telecom-class micro satellites with a fleet of 648 spacecraft providing low-latency, high-speed Internet access to small-user terminals deployed around the world.

OneWeb plans to work with local operator partners to provide this access, though it is unclear what kind of terminals will be used, or how much they will cost.

Wyler also said OneWeb’s first satellite launch vehicle would be Virgin Galactic’s LauncherOne.

Both announcements prompted comparisons with past Internet satellite ventures that flopped, notably Teledesic and Skybridge, two well-financed startups with plans for global constellations of low-orbiting satellites that could deliver high-speed Internet to corporate and individual customers. Both companies spent substantial resources to obtain regulatory licensing with the International Telecommunication Union (ITU), which coordinates orbital frequencies in an effort to avoid radio interference among spacecraft. But neither constellation was ever developed, owing largely to technical issues.

Despite such comparisons, some industry observers are optimistic that these technical setbacks could be overcome today.

“Ten or 15 years ago we had similar business models appear, all of which collapsed,” says Francois Auque, head of the space division at Airbus Defense and Space here. “But today we appear to be in a new phase, with more technical and financial capabilities being brought to bear on these new projects.”

Still, Musk’s presentation, which was posted on youtube.com, left a number of technical questions unanswered.

For example, while OneWeb already has Ku-band slots filed with the ITU that Wyler obtained through WorldVu, it is unclear whether Musk will have access to frequency slots in low Earth orbit.

“There’s the ITU filings, and we’ve done the filings associated with that,” Musk said during his talk. He was also vague regarding the ground system for his constellation, and whether he could make high-throughput ground terminals for low-Earth-orbiting satellites affordable for individuals.

“The user terminals will be at least $100 to $300, depending on which type of terminal,” he said, without explaining the challenge of developing antennas designed to track multiple satellites passing overhead.

Musk has long said his ultimate goal is to establish a permanent colony on Mars, and that his satellite venture could support this in multiple ways. First, he says, because satellites provide a better means to generate revenue than launchers, but also because Mars will need communications, too.

“A lot of what we do in developing an Earth-based communication system could be leveraged for Mars as well, crazy as that may sound,” Musk said.