Ok, thanks again for continuing the discussion. My conclusion is as follows:
1) dwm will be slightly redesigned, code-wise. I consider having a less suckish font and drawing abstraction in order to be implemented in different ways (which will also be used by st and dmenu). Officially there will only be X font support as is. But due to this abstraction it'll be possible to implement cairo/pango backends if that seems more suitable for non western glyphs without the need to change any vanilla dwm code. 2) Similiarly I will separate the bar bits into a separate file to make it easy to get rid of the bar if someone prefers going with dzen+xdotool. 3) Regarding the broken apps I need to investigate further. If it's really related to reparenting, it's definately a bug in these apps or in the toolkit (JDK?) they rely on. If that's the case I WON'T introduce kludges to handle them, instead we should file bug reports that these apps get fixed. Particularly if these apps are of commercial nature, it should be the vendor's interest to fix their apps that you as a customer can use them in your preferred environment. You or your organization pays someone license fees I guess so that's just a totally valid request that the vendor fixes these apps. 4) Regarding multihead I will start a separate thread soon what a viable solution might be. Kind regards, Anselm