An interesting article, although I think that in the end the writer threw the baby out with the bath water - even I would take Ubuntu over Windows!! Open/Libre Office is pretty much up to speed and works well in the real world, and there are plenty of other examples of open-source that do a better job than their closed-source counterparts. However he did manage to hit a couple of big nails on the head.
I wear two hats when I do Linux. The first is as a hobbyist, a tinkerer, a hacker. I like to get under the bonnet, fiddle about, build systems, play with artwork, try out different things. My other hat is as a computer user, and as a user I want my system to be rock solid, stable and not messed about with. I like continuity, reliability and I tend to be conservative in my software choices.
Enough has been written about the cavalier attitudes of some developers - the KDE4, Gnome3, consolecrud and sudo debacles are well known. In the real world these projects would go bust or be redeveloped at the behest of the users. Do it or lose market share and go broke. Mind you, some companies like Google and Facebook have been known to treat their user base with utter contempt, probably based on the fact that they know the user doesn't have much choice in the matter. Interestingly within Linux the Openbox scene doesn't seem to suffer from this syndrome - maybe that's due to a strong female presence in the upper echelons, although that's no guarantee of ethics either.
However, I can see where the writer is coming from - I am a great believer in the adage "If it isn't broken, don't fix it". I am also a firm believer in accepting at times that something can't actually be improved - sometimes something just works so well that it doesn't need to be fiddled with. I know this must be particularly galling for developers, but that is one of the points this article is making. Is the product out there for the sake of the developer, or for the user?
If the product is out there for the user then the needs and wishes of the user have to be taken into account. In the real world computer users do a select number of things, using specific software that they have become used to, and almost automating tasks through repetition. I do research, upkeep three websites, do a weekly article for our local newspaper, do some graphic production, and build computer systems. I don't like my system unduly messed about with by bored and pushy developers. I don't want to keep relearning how things work. I don't want to have to keep changing the way I do things because somebody has decided there is a "better" way. The bottom line is that the computer is just a tool and I need it to be predictable, and I need the development base to be non-intrusive.
In the end though Open Source has to be the choice of thinking people. There are plenty of examples on both sides of less than friendly behaviour to the end user, but at least in the Open Source world you can participate. That may not suit the "typical" computer user that the article talks about, but then you have to decide whether you want to be part of the herd or to follow what you think is right. Neither is a particularly easy road - and neither is limited to computer systems either. There are many other things in life that require simliar decisions.