Canonical did a lot of good, but i think it was always profit-motivated, thus always at the mercy of a marketplace mostly full of people who didn’t (and largely still don’t) know good software from bad.
Look at the basics, corporate income = (customers * sales) - expenses. And the largest customer-base is Mr and Mrs Average, who are consumers rather than creators. They don’t know just how bad things are, they think what the advertisements tell them to think, that it’s the newest latest technology (even though it’s from the '60s and '70s with a few new i/o devices and faster and bigger everything). And if one of them figures out how to make the thing work, they’re all raving about how great it is because their flip-phone didn’t have it, way cool!
Heh, i’m ranting again innit; sorry. I’ve been unusually tied-up with non-computing “life” issues, if anyone remembers how “life” can mess up your critical-path charts and totally vaporize any ideas you might have had about “release dates” (if you’re dumb enough to publish those in the first place). But i am still scheming on how to set up a “foss” community that’s kind of like the foundation whoosits set up, the guy who started the whole open-source concept and who is now an object of some peoples’ (fake-news) derision, only with some goddamn economic muscle, enough so i could be assured that @wimpy for example was not at the mercy of Canonical which is at the mercy of a bunch of civilians who just want bezel-free smartphones. Sorry, end of rant. For now.
[note to moderators: if you don’t want me ranting, just delete my userid, and when i come back from that particular grave we can discuss the merits of end-user identifiability.] 
