On Wednesday 16 January 2008, José L. Redrejo Rodríguez wrote: > And the final problem is the desktop: Our students and teachers have > been trained and using GNOME for the last 5 years, so we can not discuss > if GNOME is better or worse than KDE: It just something that can not be > changed and both desktops have to live together. The above is quite understandable from a technical support standpoint, but from a pedagogical perspective regarding computer-literacy this is IMHO just plain wrong and counterproductive: Both KDE, and Gnome (and any other current desktop) is makes up a WIMP interface. Given similar configuration (regarding menu structure and the like) a computer literate end-user should have no real dificulty switching between one and the other for everyday use. Yet practically speaking a lot of people currently do have problems. Mostly because they only have experience with one desktop and thus have no experiental basis to differentiatie between the semantically important interface elements and the purely visual or spatial differences. By giving children (and teachers :) experience with multiple desktops they'll quite naturally start to extrapolate which are the important elements and patterns Thus yielding a more computer-literate person. > Happily, freedesktop.org initiative has made thing easier in this field, > but there still some configurations very kde-related in a Debian Edu > installation. Ideally we would make it easy for a school to give their students experience with multiple desktops (and themes and widget styles) by providing similar setups for multiple desktops. The hard part about that is deciding on how a desktop should be set up in the first place. Once you have a clear idea of "we want it this way, because of this reason" implementing that vision in a particular desktop is a relatively minor effort. The amount of desktop customization that debian-edu-config does is minimal ATM, while my impression is that Linex has a put a fair amount of work in that. => I think documenting the linex vision, and consequently implementing it for both KDE and Gnome (and maybe XFCE) is the way to go. That'll give us one starting point from which to refine things. > After seing things like: First doing an svn export before grepping, and excluding README, changelog, and translations cuts the below numbers seriously, though that still yields a substantial amount of work: > ../src/debian-edu-install# grep -r skole * |wc -l > 282 grep -r --exclude=README* --exclude=changelog --exclude-dir=debian/po skole * | wc -l 60 > ../src/debian-edu-install# grep -r "10\.0\." * |wc -l > 3 same > ../src/debian-edu-install# grep -r "tjener" * |wc -l > 67 24 > ../debian-edu-config# grep -r "10\.0\." * |wc -l > 100 49 > ../src/debian-edu-config# grep -r "tjener" * |wc -l > 298 141 > ../src/debian-edu-config# grep -r skole * |wc -l > 628 251 > Option c) seems the more elegant and flexible solution, from a technical > point of view and thinking of people around the world who will have > different names for the computers and different network ranges, but it's > also the hardest one. If this option is chosen a lot of work need to be > done. I think c) is definately the way to go -- Cheers, cobaco (aka Bart Cornelis)
Attachment:
signature.asc
Description: This is a digitally signed message part.