The reason why a lot of people, especially people leading eq with reaction
times below 100ms and high speed internet connections prefer pss over
batclient's partywindow is because pss is faster.
True, but there is nothing preventing us from speeding up the party window by
using the pss output. This is what this script does.

http://pastebin.com/hqVMUjpS

Now about how the script does that. In the script the most important thing is
the function takeover() in that we try to get ride of the party window and
replace it with our own, slightly extended version. I am a bit supprised that
it works that well, so it wouldn't supprise me if it stops working every moment
and starts throwing null pointer exceptions arounds it. I tried to get ride of
all references to the old party window, but there sill might be the chance for
it to linger somewhere in the background. To reduce the chance of that call
'$opw_s.takeover' with a empty and more important visible party window. After
that everything is straightforward, the trigger defined in bootup triggers on
the pss lines and calls U(), from U() it goes to opw.update() where it gets
merged with the latest real update from the server to preserve information we
didn't got from the pss output and then use it to update the window.

Since this is highly experimental and I didn't tested it in any party with more
than 1 member please drop me a tell if you run into any trouble or just got an
idea for how to improve the script.

I see I missed a small output to stdout, depending on how often you trigger pss
you might want to remove the line "System.out.println("calling super.c");".