People love to say this but rarely provide concrete examples. Google has a habit of discontinuing unsuccessful services, not technologies like SDKs or whole programming languages.
Not exactly abandoned - it seems to have been maintained for 10 years, although no stable release for the last couple of years.
It does feel like it would have been a dead end to get into though.
My worry with Flutter is the same - it won't get large enough adoption by other tech companies, and Google will move on to something else.
I think it'll still hang out as open source project. But will it get the large investment to make improvements as the various OSs change over the long time?
Every major revision of software has breaking changes (or as you call them, "incompatibilities"), some more than others (see Python 2 -> 3).
That doesn't mean it's "abandoned". Google made major API changes and renamed it to better communicate that it's not some simple JS framework – calling that abandoned is disingenuous. It's only abandoned in-so-far as every old version of every software ever gets abandoned at some point.
I have to port, due to the backwards-incompatible changes.
Those changes may well be for the better, but for those of us maintaining stable programs, porting means increasing risk of failure for no perceivable end user benefit.
Microsoft is much better at supporting old tools and software, at least historically.
