[CLUE-Tech] Hyperterminal, minicom, and SSH

David Anselmi anselmi at americanisp.net
Sun Jul 13 19:48:58 MDT 2003


Kevin Cullis wrote:
[...]
> My wife uses Hyperterminal to access from home a separate phone number
> but has recently access the same info via the organization's network the
> MLS system.  Would SSH do the same as Hyperterminal?

Probably not.  To talk to a modem you are, essentially, talking to a 
serial port.  Hyperterminal talks to serial ports.  Minicom talks to 
serial ports.  Telnet and SSH talk TCP (network card, not serial port).

> Why?  My wife says that the "command line" is faster and she's now using
> Linux to access the web based stuff and I'm trying to get her connected
> via possibly SSH to further "disconnect" her from Winblows.
> 
> Any suggestions or hints to follow?

Minicom seems to be exactly what you want.  Is there something about it 
you don't like, or something it does/doesn't do that you want?

<geek stuff><!-- this is kind of long, might want to bail now-->
Like most things in computing, this stuff is built in layers, so let's 
look at it that way.

You have a phone line that runs from your house to the MLS system. 
Phone lines carry sound -- voice conversations.  Two modems can talk to 
each other using sound over a phone line (they send bits of course, not 
speech).  That's the first layer.

Computers connect to modems through a serial port (either from a 
physical port connected to an external modem, or through a port built 
into the modem card).  We'll ignore Winmodems for now, their differences 
are orthogonal to this discussion.  Naturally the serial port sends bits 
between the computer and the modem.  This is the second layer.

At the next layer you have two choices.  Originally serial ports 
connected to teletypes, the keyboard/monitor used to interact with the 
computer.  So the thing sent back and forth over the serial port was 
ASCII, text and control characters.  That still works today.  On your 
Linux box, your virtual consoles are simulated serial ports.  Your 
pseudo terminals (for xterms or telnet or whatever) are simulated serial 
ports.  And you have some real serial ports.

Take a look at your /etc/inittab and you'll see some lines about 
tty[1-6].  Those are your virtual consoles and a program like getty 
listens to them.  Want to log in to your machine over a modem?  Just 
copy the tty1 line and change tty1 to ttyS0, plug your first serial port 
into a modem, and dial away.  getty is a program that waits for 
connections to a serial port and runs the login program when one occurs. 
  In the network world you use telnet (client) and telnetd (server) to 
do this.  In the serial world it's minicom (or similar) and getty.

The other choice you have is to run a network protocol over your serial 
line.  That is what most people do with their modems these days and the 
program they use (taking the place of minicom) is ppp.  Where minicom 
and getty allow you to run a keyboard/monitor over a modem, ppp allows 
you to run a network connection over it.  Your modem, serial port, and 
ppp take the place of your network card.

So with that lengthy background (if you're still reading), if your wife 
is using hyperterminal she's not using TCP/IP so she can't use telnet or 
ssh.  She can use minicom.  There might also be "command line" versions 
of minicom (which is screen oriented like vi) or you can set up the 
command line and scripts to dial numbers for you and so on.  But minicom 
adds modem setup stuff, dialing directory, and so on, so screen 
orientation makes it a little more friendly.
</geek stuff>

Dave




More information about the clue-tech mailing list