It's a familiar pattern. Upstart was killed to fit upstream Debian's systemd base, next Unity is axed in order to use Debian's upstream desktop and cut in house development costs.
Clearly Ubuntu is not where canonical want to put their investment. It's hardly surprising as Ubuntu's popularity has been declining for years. Shuttleworth may have finally realised what many of us knew years ago: Ubuntu was never going to challenge Windows or Mac on desktop/laptop PCs and it was never going to be a revenue stream (any attempts to make it so were badly implemented and failed). In the meantime Android came along and all their efforts seemed futile, the rest is history.
One of the main failings of Unity is that it is open source in the technical sense only. It was created by canonical for it's products and attempts by others to build on other distributions (such as Fedora) ultimately failed.
And this is where canonical sadly missed the point. gnome at least produce a desktop which can be built on other distributions and even ported to other *nix. canonical chose to cast all of this aside and go "completely in house" and lose the contributions of thousands of other developers - all working for free and to start hacking libraries, etc.
This is where Canonical's model failed and where Red Hat's has undeniably succeeded.
The writing is on the wall for Mir - the single biggest point of failure of which is the GPL3 licensing + Canonical's own CLA (Contributor License Agreement). This will ensure it remains an in house project. I can't see the sense in external "contributors" submitting code under those terms, particularly when there is an MIT licensed alternative.
|