Le jeudi 24 février 2011 à 13:53 +0100, Jérôme Hénin a écrit : > Hi everyone, > > On 24 February 2011 13:27, Michael Scherer <[email protected]> wrote: > > This is not the number of click, but the fact that it take a longer time > > to scan the whole menu. Like reading a article with and without carriage > > return. The number of word to read are the same, but it is easier with > > carriage return, because this doesn't require any form of coordination > > ( ie, this is not on the same cognitive level ). > > > > I am not sure if I am clear enough, but on the other hand, I fear that > > explaining the whole topic would requires much more time, and time have > > showed that people hate when I explain them ergonomics ( as seen on > > various rpmdrake bugs ). > > Michael, your point about ergonomics is very true in principle, but > there seems to be a missing link in its implementation. The two-tier > system, with a 'More...' submenu is very sensible, _if_ that submenu > contains the most rarely used entries.
That unfortunately more complex. Ie ideally, you have to track applications usage to make this work. There is work going on with zeitgeist and integration to gnome-shell/unity that could lead to this. So in the mean time, we have to rely on guess made from packagers on what would be the best application to use _if_ people did install more applications and didn't remove the installed default one, nor cleaned/move the entry ( because for some reason, it seems that no one ever do it ). > IIRC, a bunch of proprietary > applications do this with self-masking menu entries. In our case > (mdv/mga), there is no mechanism to ensure this, as far as I know. > That would be a major step forward. Well, I guess that a bug do exist for this in kde bugzilla. If not, then maybe you could open one. -- Michael Scherer
