Experimenting with Split on my machine

You can do some limited experimentation with Split on my machine, via a telnet window. However, the program was not designed to be interactive, and so will frequently give fatal error messages (instead of warnings). Unfortunately, when this happens, the telnet window will probably vanish, so you will not even see the error message! (You could circumvent this by using telnet directly.) TYPE CAREFULLY!

To connect, click here, and then login as the user "split" with password "seven7". Please do not do monster experiments; for that you should install the program on your own machine instead.

If someone else is similarly logged in when you try, your login will not be accepted (and the telnet window will probably vanish instantly).

Here is a small but amusing example:

    type [24,12,8];
    bound y*;
    exit;

For more complicated examples, you can use the commands

    accept (tables only) code.data1;
    accept (tables only) code.data;

to load my tables. Or you can use a command like

    ??no [25,8,10];

to cause my program to accept without proof that codes with some particular parameters do not exist.

If you do a calculation which uses the linear programming software CPLEX, and I happen to be doing so also at the same time, bad things will happen. There seems to be no way to work around this problem, since my CPLEX license only allows one process to execute at a time.

I suggest reading at least the brief tour.

Please tell me if you have suggestions as to how to make this demonstration work better, or more generally if you have suggestions or comments about Split.

Mail to: jaffe@cpthree.unl.edu