FLAKY KEYBOARD

[PAUL DE GROOT]

Our keyboards have developed some strange quirks, as have the keyboards of many
of the users I support. It seems as though the Alt, Shift and Control keys
sometimes stick. Since I've reassigned all the function keys to work with these
keys, the consequences can sometimes be dangerous. We use Harris Write, a
mutation of XyWrite, on 386SXs (Philips P3345), on a Novell network (Netware
286 v. 2.15). I'm wondering if my tampering with the keyboard file could have
anything to do with this. To keep the keyboard file shorter, I've eliminated
many of the key assignments. Is this okay? For example, here's a small section
from the Control table in the keyboard file. As you can see, there's no
assignment for 42, 44, 47 or 48. Basically, I've taken these functions out
because I don't want unexpected things happening to our users. 41=CM 43=SD
45=EF 46=BC,c,a,CC,XC 49=$1 But it's not uncommon, particularly after you've
used one of the shifting keys, to find the keyboard no longer responds the way
it should. To save a file on our system, you hit F7. To call a file, you hit
Alt-F7. Sometimes when people try to save a file, they get ``abort, store or
open a new window'' because every time they try to save, the program is trying
to call. It thinks the Alt key is down. Tapping both Alt keys a few times fixes
the problem. These are garden variety 101-key keyboards which come with the
p3345. I use procomm and a variety of other programs and don't have this
problem with them, so it's not the hardware. This problem has plagued us for
some time. I can't identify anything new that might have caused it. Any ideas?