My biggest concers surround the "non-threaded" items... (I think it is
easy to determine when a message goes to one window or another...
we've done this before...) Items like the output of a WHO list or a
FINGER command - will I (as a client) still be expected to parse this
information if I want to display it in a different UI format or use
this in some way?
Some examples:
- Picture a list of all users (matching certain criteria) in a
scrollable list.
- Double click to get more information about them (say... the results
of /FINGER at least...) this might indicate they have an info file,
so click on that link to read it... or their web page to open up a
browser... or...
- Checkboxes to control the audio to play when this person sends you a
message...
- Drag and drop to indicate what people (from the above list) are
displayed in what windows? Or use that scroll list to open up a new
window for that person...
- Picture equivalent actions for any discussion... each has a
subwindow for the people in that discussion that is constantly
updated...
I can babble on for many more examples of this... {: I guess the
point is that many of these require that I parse everything that comes
in... and I don't relish that thought...
I've done this for Connect and Clover... and I began to do it for lily
before I ran into the immensity of the problem that parsing posed
under lily... my noets on xconnect, cloverleaf and lhub are available
for anyone that wants them.
Any thoughts?
P