Elon Musk has confirmed SpaceX is getting into the broadband and small satellite businesses, but details remain vague.  Speculation has been driven by a number of reports, the most recent a Wall Street Journal piece last week that Musk took a swipe at.

Rumors started a couple of months ago with WorldVu Satellites first working with Google, then jumping to over to SpaceX.   WorldVu has secured satellite spectrum licenses from the ITU, according to a June Space News piece, and will set up a network of 360 satellites in low earth orbit (LEO) to provide low-cost, high-speed wireless broadband coverage.  Initial operation date would be in late 2019.

Last week's Wall Street Journal piece co-authored by Rolfe Winkler turned up the Internet buzz volume to 11, resulting in a series of me-too pieces that echoed back and forth re-writing the original WSJ piece.

It isn't clear what trigged Musk into breaking silence late yesterday, November 10. He first tweeted "SpaceX is still in the early stages of developing advanced micro-satellites operating in large formations. Announcement in 2 to 3 months."   

This was followed with an exchange between Taras Pater and Musk:

                @elonmusk I hope this is about free and unfettered internet access for the masses.

                @Shmizer1 unfettered certainly and at very low cost

Key bits here are that SpaceX appears to be the firm building small satellites in large numbers to deliver "unfettered" Internet access at low cost.  There's no discussion of how many satellites or where they would be built.

A follow up Twitter exchange between WSJ's Winkler and Musk implies a lot of speculation:

                @elonmusk confirms @WSJ scoop: "SpaceX is in early stages developing advanced micro-satellites operating in large formations"

                @RolfeWinkler No, WSJ was wrong on several important points. The article shouldn't have been written on rumor and hearsay.

Musk, alas, doesn't explain what points the Journal got wrong, leaving everyone to develop more rumor and hearsay for the next two to three months before the next announcement.

Why is SpaceX getting into the small satellite business? I'd say there are five points to consider.

First, SpaceX is developing a partially reusable launch vehicle. It plans to fly back the first stage of its Falcon 9, enabling it to be re-used and thereby drastically driving down the cost of putting satellites into orbit. 

Lower cost also plays into the second point, higher flight rates. If it is cheaper to fly, more customers will want to take advantage of putting satellites into orbit. SpaceX should flying more often so they can maximize the amount of revenue they can get out of their reusable booster.  Running regular flights means you can put up a larger number of satellites at a time, refreshing/replacing them on a standard schedule.

More capable and lower cost satellites are the third key.  Traditional communications satellites are conservatively built to last for 10 to 12 years, and can cost $100 million or more.   Building large numbers of smaller satellites would result in economies of scale, dramatically lowering the per-satellite cost.  Like the consumer electronics industry, satellite manufacturers could focus more on delivering improved features through faster iterations.   Planet Labs, Planetary Resources and Google-owned Skybox Imaging all operate on a philosophy of smaller satellites at lower cost, with the understanding that shorter in-orbital life means you replace hardware more often.

Recurring and diversified revenue comes from operating a communications network. A high-speed, low latency broadband network works equally well if you are delivering phone calls to Africa or Internet of Things (IoT) applications in Canada and North America. 

Finally, there's the whole Mars mantra.  If SpaceX plans to put a colony on Mars, it will need a communications network.  A broadband communications network in Earth orbit today makes a good beta test for MarsNet v1.0.