I've seen that happen before with a startup, they actually shutdown just because Google launched a superior product first.
Then less than 1 year later Google cancelled that project and now there is nobody to provide that service.
Then less than 1 year later Google cancelled that project and now there is nobody to provide that service.
This made me imagine some poor, bright but new-hire Google PR employee realizing this popular (and quite correct) view of Google ("CANCEL ALL THE THINGS" [1][2]) frantically trying to figure out how to actually communicate to upper management (without getting fired) the blunt truth:
This ongoing strategy of announcing projects and promptly canceling them after a few years is incredibly damaging to the company. Google's "ADHD-like low attention span" to projects is anything but confidence-inspiring. :(
1. http://memegenerator.net/X-All-The-Things
2. HN likes numbered references and I do too
My take is that Google is less developer-friendly than Microsoft and Apple. Google does not rely on external developers as much, because so much revenue comes from web based, directly consumer facing products. Google is always trying new things, because they are looking for the next big user facing product. This might annoy developers who build against Google's APIs, but that is worth it because Google relies on users, not loyal developers. It is a fundamentally different approach from Windows or iOS.
I can tell my odd experience about Microsoft: my company reverse engineered many of its products and offered APIs around products without APIs. One day someone from QA sent us an e-mail offering help if some of these products have issues with Windows 7!
A few examples
>AngularJS
>AOSP
>Google Maps
>GCN
>Chrome Dev Tools
>Polymer and Material Design
>Go
>Dart
At the end of the day, all large companies have to dip their toes in new technologies, or die. Some of these projects will work out and others won't. Small companies do the exact same thing, except instead of announcing they are discontinuing a project they go bust.
My approach is not to get too tied to any vendor's technology, because if you do, you're setting yourself up for a fall. The rise of open source and standards has made this much easier than it once was, luckily.
In a sense it is a return to the MS that was before the IBM PC, when they were supplying office and development tools to all comers (MS Office got its start on Apple computers after all).