Re: Connecting to Ethernet

[Toby Reiter <str4552@OBERLIN.EDU>]

On Tue, 19 Nov 1996, David Empson wrote:
> > There are a few Mac Se's which I think are connected in this fashion in
> > another part of the campus. But the one I am typing on right now is 
> > definitely using ethernet, as these ports used to be occupied by vt100
> > dumb terminals (now occupied by Macintosh SE dumb terminals, what a waste).
> 
> Ah - now it makes sense.  There is no way that you could possibly
> connect a VT-100 terminal directly to an Ethernet: the VT-100 only
> supports an asynchronous serial connection (possibly sync serial as
> well, but that is beside the point).
> 
> This means that your supposed "Ethernet port" on the wall is actually a
> serial port which goes to a terminal server.  This server is then
> connected to the Ethernet backbone in some fashion.
Well, all I know is that they're the exact same ports and cables 
used by all Ethernet capable computers--Macs, Ibm's, etc.-- (in dorm rooms, 
computing center, etc.) These other computers all have TCP/IP
capabilities, so these could NOT be serial cables.  I checked 
with the people at the computing center, but they didn't know what it was 
(there is someone who actually does, he just was not there at the moment.)
 
> > Well, actually I'm not sure. These Mac SE's are using Kermit (relatively
> > poor communications package) and are set up in such a way that the system
> > is completely inaccessible.
> 
> If the Macs are using Kermit, then they are definitely running over a
> serial connection.  The terminal server is probably using a Telnet
> connection over Ethernet to the host machine, and the host machine is
> able to support Kermit file transfers.
> 
> You could easily use a IIgs in the same manner - run standard
> communications software on the IIgs (e.g. Kermit, Z-Link, ProTerm,
> ANSITerm or Spectrum), and see what happens.  If the host only supports
> Kermit file transfers, you need a comms program which supports the
> Kermit protocol.
The email host is a VMS Alpha server. The majority of the ethernet capable 
computers use telnet to connect to the network, so I distinctly doubt it 
is sent only be Kermit. (However, access to TCP/IP is only available 
through ethernet, and _not_ by modem connection.)
> > I also don't have direct connection to a Macintosh hooked up to the 
> > network, (the closest Mac is a SE which would have to be connected the
> > same way is this one is now).
> 
> The existing Macs do not have direct access to the Ethernet, so this is
> a moot point.  They are effectively running in a hard-wired terminal
> session to a UNIX host.  It may be possible to run TCP/IP over this
> connection if the host machine is able to pump PPP or SLIP through the
> Telnet connection (freaky!  TCP/IP encapsulated in a protocol running on
> top of TCP/IP), or the terminal server can be configured to provide
> this.
Actually, we have a large number of IIci's, IIcx's (or whatever), and 
assorted PowerMacs attached to the network on both ethernet and localtalk
(depending upon where on campus they are). I am currently typing this on 
a Power Macintosh 5260/100 in another dorm room, which is attached to the 
ethernet network by means of a blue ethernet cable which is identical to the 
one used to attach the Mac SE's to the network. All of these normally 
ethernet attached computers are able to access all of the Mac servers 
over EtherTalk.  Therefore there is no need (as far as I know) to run 
slirp over the network to get TCP/IP capabilities (I wouldn't have enough 
disk space probably, anyway.).
> From your description, it is not at all clear what the Macs are actually
> able to do on this network, and therefore what your IIgs would achieve
> by being connected in some way.
This is because the question I had concerned the few Mac SE's I had seen 
hooked up in this way. The majority of the computers have direct access 
to both the internet over ethernet or a LocalTalk router as well as 
having access to all of the AppleShare volumes.