But that's far from being its only area of expertise. Canonical is also top class when it comes to spin. Any criticism levelled at the company is met by a much bigger reaction from its owner Mark Shuttleworth who not many dare to contradict. Often, following is his steps is his acolyte, the company's community manager, Jono Bacon.
This time, the criticism has been based on statistics presented in a study at the GNOME Users' and Developers' European Conference by senior GNOME project member Dave Neary.
These statistics, presented as the GNOME Census, show that Canonical, despite tying its release schedule to GNOME and being quite dependent on the project, contributes only 1.03 percent of the commits made to upstream.
The criticism of Canonical for being something of a parasite - this isn't the first time that figures have shown that the company's contribution to upstream projects is very low, if any - has come from many sources, one of them being Greg DeKoenigsberg, a former employee of Red Hat.
Red Hat, let me add here, is the top company contributor to GNOME, as per Neary's statistics, with 16.3 percent. Neary has also pointed out that of 11 of the top 20 GNOME contributors of all time are either present or past Red Hat employees.
So what does Shuttleworth do? He brands this criticism as being driven by "tribalism" - which he claims is a bad thing. (He conveniently forgets that if it were not for tribalism then there would be very few Ubuntu users).
Ultimately, tribalism is present in every human being - that is the reason we bond or join this group or that. Follow a football club from any code? That's tribalism. Belong to a religion? There's tribalism again. Member of an old students' association? There it is again. I could go on - but I won't belabour the point.
Tribalism exists because it is part of the human make-up - we want to belong, we don't like being different, we need to identify ourselves in some way in a world that increasingly makes people feel on the outer.
How do you give something so basic to the human race a bad name? Simple. Say that it is based on things like "the other guy has never done anything useful" and "evidence contrary to my views doesn't count."
Tribalism is definitely not based on things like this - it is based on a need to belong. It goes no further - and for Shuttleworth to link tribalism to sexism and racism is just another way of blackballing a fundamental trait of the human race.
But this is classic spin: define something in your own terms, throw a lot of high-faluting language around it, try to confuse the hell out of everyone, and, in a style that Glenn Beck would admire, you can damn anything.
The human race banded together in groups in days gone by for shelter, food and to be safe from attack from others; today we do it based on other reasons simply because the human being is a social animal.
Shuttleworth builds up this smokescreen about tribalism - and much, much later links the criticism of Canonical - which, let me remind you, dear reader, is based on the lack of upstream contributions to GNOME - to this trait. Nothing could be more irrational but he has obfuscated sufficiently by this point to be able to attribute anything to tribalism.
The lack of contributions to GNOME cannot be hidden - but does Shuttleworth deal with it head-on? No, there are various ephemeral excuses trotted out - we are a big community, we have a code of ethics (which is observed in the breach quite often), we are bound to be criticised more often than others because we are bigger, and more and more such irrelevancies.
Answer the question - why is Canonical taking so much from GNOME and giving back to little? That. I'm afraid, Shutteworth conveniently avoids. Even the former Australian master leg-spinner, Shane Warne, would have been unable to generate this amount of spin.
Close on his heels comes his lieutenant, Bacon, another accomplished spinmeister. The first and last time I encountered him, he tried to kill a story which I was doing, by suggesting in the most friendly manner that there was nothing in it. He is brilliant at playing the good cop and then quickly flinging some mud before running away.
His method, this time, is the oldest in the book - relocate the argument on a pitch which is friendly to us. We can't fight on that territory, we'd be unmasked. Let's move the fight over here.
In classic spinmeister mode, Bacon starts out by praising Neary's efforts and creating turbulence so that he can move the argument to an area where Canonical can be, in his mind at least, exonerated.
Neary's census, he says, "only takes into account upstream contributions to GNOME itself." Sure, it does, Neary never purported to do otherwise. No-one who has criticised the company has claimed otherwise either. Canonical's contribution is beggarly; pray why?
But Bacon is now in full flow. "What the report doesn't take into account are upstream contributions that are built on the GNOME platform but (a) not part of official GNOME modules, and (b) hosted and developed elsewhere, such as Launchpad. As such, while the report is accurate for showing code and contributions accepted into GNOME, there are also many projects built on GNOME technology that are not taken into account due to non-inclusion in GNOME modules or being developed outside of GNOME infrastructure."
And then he goes on to list, in great detail, the contributions that Canonical has made to the areas which he specifies. But Canonical is not being criticised for this - it is being criticised for not making contributions to GNOME itself. Why not provide a plausible answer for that?
Neary has not said a thing about Canonical's contributions elsewhere. As a GNOME member, he has looked at one area, that of upstream contributions to GNOME itself, and listed the contributors by the quantum of their contributions.
Why is it that a company run by a multi-millionaire (the global financial crisis would, one presumes, have removed the title of billionaire from Shuttleworth) and one that takes so much from a multitude of projects - Debian, the kernel and GNOME to name just three - so parsimonious when it comes to giving back?
Canonical may have been criticised by a former Red Hat employee - but it's also been slammed by Brian Proffitt, formerly (corrected) of the Linux Foundation, a very fair commentator. But neither Shuttleworth nor Bacon want to go into that territory - all their arguments would collapse, for to which "tribe" does Proffitt belong?
Free and open source software projects thrive in a symbiotic situation. A parasitic relationship will end up harming both the host and the pest. Some food for thought for these spinmeisters at Canonical.