We already have great, modern and maintained screen alternative: it's called tmux. And yes, it's written in c, but don't immediatelly see any reason why it woul

We already have great, modern and maintained screen alternative: it's called tmux. And yes, it's written in c, but don't immediatelly see any reason why it would be much better to rewrite it in Haskell. And I'm saying that as a XMonad user who knows a little bit of Haskell. But you know, there is nothing stopping you from trying it yourself to show that this idea is actually viable :) Edit: There are actually few reasons why the XMonad model would not be a good fit for screen like program: XMonad requires haskell compiler to be installed on machine you are using it - this is fine for my personal machine running recent fedora (or any recent linux distro), but it would be a problem for all those servers I have tmux installed - like RHEL, Debian stable or FreeBSD, tmux is often already there or easy available in official/semioficial package because (among other things) all it's dependencies (few c libraries) are likely already installed. On the other hand you do not want to install ghc on much servers for obvious erasons :) But as I said, if you like the idea, just go and make it happen - it sounds like a fun project to start. But I'm afraid it would not be much practical.