The bit about psychology is "obvious", but the argument falls apart when trying to get from that to the conclusions. It's true that taking away something that

The bit about psychology is "obvious", but the argument falls apart when trying to get from that to the conclusions. It's true that taking away something that people count on makes them unhappy. To declare, however that this is the sole or main reason for the widespread rejection of the Gnome changes requires more evidence. It is definitely false in the case of many users (myself included). It's really just a variant of the old line "you dislike this only because it's different; but try it our way and you'll find it's better". This is a dishonest, and frankly an offensive rhetorical trick. It presumes to know the reader's mind better than the reader does; implies that familiarity is the only reason anyone ever prefers one thing over another; and assumes that all change is always good - all of which is absurd. Similarly in the linked article, the author pretends that there is no objection based on the actual merits of the software: in other words, users object that the removal makes the software less useful, and the writer pretends to misunderstand the complaint as "we now have n - 1 features instead of n features". He even arrogantly says that some features they "shouldn't have had in the first place" - when in reality, people complain because they were using the feature for something they cared about and aren't satisfied with the new way. The really important hidden premise of apologists like this is that there are no intrinsic merits of one design over another - that it's only a matter of style and fashion and what's familiar, and there's never any factor of one way being intrinsically better or worse suited to the interaction of human and device. That's a big assumption, but once they've put in a huge amount of work based on it, they can't even admit that anyone could have any intelligent objection, and defensively blame any dissent on an irrational emotional response. The same observations apply in the case of Ubuntu Unity and Windows 8: the root cause of their failure is a sort of paternalism where the GUI designers disregard what users actually prefer, and can't even admit that the preferences are based on practical workflow considerations rather than emotional reactions.