> he doesn't think it's a relevant bug any more - that desktop is no longer important.
I don't think the former implies the latter. The desktop can be (and is) still important, it's just that such a narrow view isn't healthy for such a large project.
I'd argue its focus and specialization, something Ubuntu lacks. The TV and Ubuntu-on-Android products have not yet seen anything noteworthy happen, the phone might do something but they are jumping in a big pond. Their lack of focus on the desktop and the resulting obsolescence of their software (the software center, from 2008, mainly, but also compiz and upstart) is hurting the primary product people engage with them for, and it is for a grander scheme that isn't playing out.
Yet, at least. I would never write Canonical off from eventually having one OS and thematic style to run on everything from phones to tvs to pcs to cars etc. It just doesn't look like it is coming soon, and their core product suffers for it. (I've switched my relatives to openSuse just because systemd + yast + software.opensuse.org is much nicer than searching launchpad).
If Ubuntu lacks a solid base from which to build the rest of its empire off of, then that's likely worth its own defect. That prerequisite is just kind of common sense in most business models.
I don't use Ubuntu or know enough about it to say for sure, but I don't think probing other spaces for opportunities is a bad idea. It's not difficult to understand the desire to not be a one-trick pony. However if those probes distract from their main offering (which has been the desktop, though this sort of discussion makes it seem like it may not be for long) then they are obviously a bad idea.
The desktop is still relevant and really its pretty close to as relevant as it was in the 90's. It's just there are more ways to use computers now than there were then and those new ways of computing have been largely dominated by new devices. In other words, the desktop doesn't have to lose relevance just because there are other relevant technologies.
I don't think the former implies the latter. The desktop can be (and is) still important, it's just that such a narrow view isn't healthy for such a large project.