>"How can anyone invest time developing for them?"

What alternative do you suggest? I suggested they were the best remaining option for an open-source mobile OS now that Ubuntu were dropping out of the race. What other options are there? Sailfish? Anything else?

There are no alternatives.

This ideology of open-source mobile OS goes back to Openmoko.

None of them managed to sell enough devices to keep a profitable business.

>"None of them managed to sell enough devices to keep a profitable business."

Nokia managed it with the N900 and the N9, the only reason they didn't continue is they lost their bottle and sold out to Microsoft.

If you were fine with "open source" on those Nokia devices, you should be happy with "open source" on Android.

Also as a former employee I can tell they didn't sell that much units, specially when compared with Symbian devices.

What he said. On top of it, maybe next failed attempt can try building on one of these for a stronger foundation:

https://github.com/ExpressOS/expressos

http://genode.org/

Tizen is an alternative, whether you like it or not is another matter.

Oh, you make it sound so simple. A platform is an alternative to dominating ecosystems if it exists at all. It having apps, locking you into Microsfot, or even existing 5+ years down the road doesn't matter? If that's true, there should be all kinds of people on all kinds of smartphone platforms right now instead of most of them on two.

Meanwhile, Tizen is available for people wanting whatever their changing offering is. I guess just budget regular rewrites in a new language/platform you had to learn into your FOSS app on Tizen with either few contributors or the more common zero. That's some great ROI.

KDE's Plasma Mobile? Like Ubuntu Phone/Unity 8 it's also using Qt QML to build flexible layout apps that support touch and keyboard+mouse.

https://plasma-mobile.org/

That's even worse than the already-terrible market for FOSS phones. You get something that will have Linux or some standard API but probably disappear eventually. Then add the aggravation of having to change your stack constantly. Yeah, I'd pass on it unless it stabilizes into a standard stack. Preferably not dependent on Microsoft or Oracle tech either to reduce patent/copyright risk a bit.

Oracle aside, .NET core is open source and I don't see today's Microsoft pulling the same crap as Oracle did with Google.

What about tomorrow's? Copyright's, patents, and software last a long time. Better to know you aren't dependent on a malicious party who can turn against you any time.

In today's market, I think Microsoft would rather give a competitor a royalty free permanent license to all their patents than let Google continue to just own the industry. Mobile is only one of the many markets they compete in, but Google is using their mobile dominance to help lead and push their expansion elsewhere. Microsoft needs a new mobile OS option out there, even if it isn't theirs.

Microsoft makes several hundred million to billion on Android royalties alone. They did that with either actual lawsuits or the threat of them. I don't remember what it was in the start but do know they suck a fortune out of Android vendors (esp Samsung) while they did everything they could to put it out of business with Windows Phone. Nothing contributed, but billions taken via patent claims.

https://www.forbes.com/sites/ewanspence/2015/11/01/microsoft...