Canonical might have problems with the current de'facto toolchains of the GNU/Linux ecosystem, or it is to large or unmaintainable for what they want to do with it. Like you said maybe its a good idea, maybe its not.
Starting clean you can take all the knowledge gained from previous open projects and create something fresh that is more lightweight, faster and easier to maintain. But will this new components live up to it expectations? Only time will tell, and Canonical is willing to back that venture.
I also believe one need to stop calling linux, linux or the lesser adopted but more correct term for me GNU/Linux. Call the operating system you use by its distribution name, that is the actual ecosystem you using. Which yes, is part of a greater Open Source, Free Software ecosystem.
But to achieve inovation one needs to boundaries, and at the moment Ubuntu is the OS on the desktop with most commercial and adoptions success.
1) Make the GNU/Linux stuff work on Android. That's what Tizen does, AFAIK. It's a lot of work. Advantage: everything works with it, and it's well understood.
2) Glue stuff up to work with whatever Android does. That's what they're aiming for. Advantage: less work.
Why didn't the Android team make the GNU/Linux stuff work on Android; Because it is a lot of work?
Why should another company then spend resources to do that work.
I don't understand what you mean with your second statement. But that might just be because I'm dead tired. But I can not stop myself from reading and typing!!1
> Why didn't the Android team make the GNU/Linux stuff work on Android; Because it is a lot of work?
Because it's GPL licensed. That's it.
Google has spent a lot of time and work ripping out perfectly good pieces of code that are GPL licensed and replacing it with (sometimes inferior) BSD or Apache licensed pieces. At some point, the only piece of GPL code will be the Linux kernel.
Device manufacturers' lawyers are very paranoid. That's (part of) the reason Android has its own BSD-licensed libc, "bionic", instead of glibc.
When Google ported Chrome to GoogleTV, which runs Android, they ported glibc, too. Apparently, it was easier for them to port glibc to Android than port Chrome to more limited bionic.
Yet in the end, you pull the entire Linux FOSS community towards whatever whimsy Google is going on at any given time. If they suddenly redid the graphics API and threw out their current binary blob batch, the FOSS space would have to get back up and try again after getting the rug pulled out from under them.
I don't see it as good if Google can lead FOSS desktop development around by a leash.
Starting clean you can take all the knowledge gained from previous open projects and create something fresh that is more lightweight, faster and easier to maintain. But will this new components live up to it expectations? Only time will tell, and Canonical is willing to back that venture.
I also believe one need to stop calling linux, linux or the lesser adopted but more correct term for me GNU/Linux. Call the operating system you use by its distribution name, that is the actual ecosystem you using. Which yes, is part of a greater Open Source, Free Software ecosystem.
But to achieve inovation one needs to boundaries, and at the moment Ubuntu is the OS on the desktop with most commercial and adoptions success.
My 0.02c.