kibitz starts a new shell (or another program, if given on the command line), while prompting user2 to run kibitz. If user2 runs kibitz as directed, the keystrokes of both users become the input of the shell. Similarly, both users receive the output from the shell.
To terminate kibitz it suffices to terminate the shell itself. For example, if either user types ^D (and the shell accepts this to be EOF), the shell terminates followed by kibitz.
Normally, all characters are passed uninterpreted. However, if the escape character (described when kibitz starts) is issued, the user may talk directly to the kibitz interpreter. Any Expect(1) or Tcl(3) commands may be given. Also, job control may be used while in the interpreter, to, for example, suspend or restart kibitz.
Various processes can provide various effects. For example, you can emulate a two-way write(1) session with the command:
The -noproc flag runs kibitz with no process underneath. Characters are passed to the other kibitz. This is particularly useful for connecting multiple interactive processes together. In this mode, characters are not echoed back to the typist.
-noescape disables the escape character.
-escape char sets the escape character. The default escape character is ^].
-silent turns off informational messages describing what kibitz is doing to initiate a connection.
-tty ttyname defines the tty to which the invitation should be sent.
If you start kibitz to user2 on a remote computer, kibitz performs a rlogin to the remote computer with your current username. The flag -proxy username causes rlogin to use username for the remote login (e.g. if your account on the remote computer has a different username). If the -proxy flag is not given, kibitz tries to determine your current username by (in that order) inspecting the environment variables USER and LOGNAME, then by using the commands whoami and logname.
The arguments -noescape and -escape can also be given by user2 when prompted to run kibitz.
% kibitz user2 kibitz user3
Additional users may be added by simply appending more "kibitz user" commands.
kibitz handles character graphics, but cannot handle bitmapped graphics. Thus,
% xterm -e kibitz will work % kibitz xterm will not workHowever, you can get the effect of the latter command by using xkibitz (see SEE ALSO below). kibitz uses the same permissions as used by rlogin, rsh, etc. Thus, you can only kibitz to users at hosts for which you can rlogin. Similarly, kibitz will prompt for a password on the remote host if rlogin would.
If you kibitz to users at remote hosts, kibitz needs to distinguish your prompt from other things that may precede it during login. (Ideally, the end of it is preferred but any part should suffice.) If you have an unusual prompt, set the environment variable EXPECT_PROMPT to an egrep(1)-style regular expression. Brackets should be preceded with one backslash in ranges, and three backslashes for literal brackets. The default prompt r.e. is "($|%|#) ".
kibitz requires the kibitz program on both hosts. kibitz requires expect(1).
By comparison, the xkibitz script uses the X authorization mechanism for inter-host communication so it does not need to login, recognize your prompt, or require kibitz on the remote host. It does however need permission to access the other X servers.
If the environment variable EXPECT_PROMPT exists, it is taken as a regular expression which matches the end of your login prompt (but does not otherwise occur while logging in). See also CAVEATS above.
kibitz is in the public domain. NIST and I would appreciate credit if this program or parts of it are used.