> [...] I use SLiRP to talk.  This setup generally works, except that I
> can't successfully ftp into my NetBSD box, which is a pain, and I
> can't ftp into many outside sites unless I set my client to "passive"
> mode, whatever that is.

These are inherent in SLiRP.  When you connect with slirp, you (more or
less) can't accept connections, because you don't really have an
address of your own; your traffic takes on the address of the slirp
host.  This is why incoming ftp doesn't work.

As for passive mode, that relates to the way FTP works.  When you fire
up ftp to a host, it establishes a connection from your machine (C) to
the server host (S).  This connection is called the control connection.
When you want to send or retrieve data, such as a file or a directory
listing, it establishes a second connection, called the data
connection, for the purpose.

The reason this is relevant is that there are two ways of establishing
the data connection: S listens and C connects, or C listens and S
connects.  By default, most FTP clients use the second mode, partly for
historical reasons and partly because it's better supported by the
protocol (or at least was last I looked - has anyone updated 959?).
But of course the second mode won't work if C can't accept incoming
connections - as is the case when C is slirping in.  (The first mode is
called "passive" mode, after the PASV command in the FTP protocol,
presumably so called because in this mode the server listens passively
instead of connecting actively.)

> Another, related question: Would running an X connection over a 19200
> baud SLIP connection be too painful to consider? Or would it be okay?

Depends on how tolerant you are.  My home LAN is connected to the
outside world by a 19200 serial line, and I regularly use X over it;
for example, the window I'm typing this letter into is running at home
and displaying at work.  I find it usable.  I know at least one person
who finds the latency it introduces in keystroke echo so disconcerting
as to render it borderline unusable.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B