On Monday 18 August 2003 10:11, Michel Dänzer wrote: > On Mon, 2003-08-18 at 19:10, Leandro Guimarães Faria Corsetti Dutra > > Em Seg, 2003-08-18 às 18:11, Michel Dänzer escreveu: > > > The logs don't lie > > > > I hope that's true in X. I wouldn't hold it as absolute truth, from > > the experience I have with other systems. > > I meant X server logs specifically. I have been making very good > experience in this regard (I don't remember a single case where I > couldn't trust X server log output, but maybe I'm just good at > suppressing those memories ;).
Actually, in this case, the X server should probably WWarn about the invalid Core Keyboard configuration. The invalid ibook variant should have been noted, I think. (Though that's not really a lie, is it?) > > > the keyboard configuration was always identical on > > > server startup (unless the logs contained a relevant difference you > > > didn't post), the difference was likely caused by a client. > > > > OK, which clients could be those, and how to investigate the > > possibilities? > > Frank made good suggestions for this. As soon as you mentioned that it could be a client, I thought of all the fancy new config stuff Gnome and KDE have. Frank
