[log in to unmask] writes:

>  I have a Win95 PC that I'd like to connect to an HP 3000 and run an
>  emulation package. I'm looking for technical help and suggestions. Please
>  respond to
>
>  [log in to unmask]

Let me take "gcosta"'s request as an opportunity to announce that a new,
intermediate version of QCTerm is now available for downloading. The URL is:

     http://www.aics-research.com/qcterm/

If you have previously downloaded and installed a version of QCTerm, all that
you need do is download the new copy and unzip it. Otherwise, for a first-time
installation, you'll need to install the fonts to get it to work. Complete
installation instructions are on the page referenced above and a brief manual
for QCTerm is found by following the link at the bottom of the page labeled,
"A Brief Explanation."

As always, QCTerm is freeware. There is no charge, cost, or catch of any kind.
In fact, we don't even keep track of how many people log on to the web page.
It's just there to use if you wish -- but remember, this is an early version.
It is not yet complete.

In that regard, I was hoping to have the block mode/format mode problems
worked out by now -- but there are still a few little pecadillos yet to be
solved -- so don't use this version with VPLUS screens or the visual mode of
QEDIT. (Thanks are very much in order for Mike Shumko of Robelle for providing
a trial copy of QEDIT so that we might debug QCTerm. It is my understanding
that the newest version of QCTerm (not the one on the web page) is now working
properly with QEDIT, although I haven't verified that myself. Nonetheless, we
still have a few problems with some VPLUS screens.

What is new in this release is that it now has Telnet capability -- in
addition to serial port connectivity. The addition of Telnet was in direct
response to the comments made on this list. NS/VT connectivity will eventually
occur (I presume), but its original schedule was quite a ways out -- and
remains so now.

However, let me say that Telnet ain't bad. We've implemented two versions of
Telnet, which we've called "standard" and "advanced." The "advanced" version
is nothing more than a reinvention of a thirty-year-old idea: half-duplexed
communication. Under standard Telnet, when you strike a character on the
keyboard, the character is packetized as a single character packet,
transmitted to the host where it interrupts the CPU, is added to a growing
buffer, and then echoed back to you. Over any distance at all, this leads to
very mushy communications that are occasionally so irritating that you can't
look at the screen as you type.

Under half-duplex, we suppress echo in the host computer and echo all
keystrokes immediately onto the screen. This not only cuts the number of
packets sent and received by half, it makes long distance communication with
Telnet perfectly psychologically acceptable. Many thanks are due to Gavin
Scott and Stan Sieler of Allegro and Rene Woc of Adager for opening up their
HP3000's to let us telnet into their machines repeatedly. After we got things
worked out, two thousand mile half-duplex telnets are now not all that much
worse than working on a machine right next to you. I'm actually fairly
impressed.

The next test for Telnet was really long-distance communication. Having no
friends on the Moon yet, South Africa was as far away as we could get from New
Mexico and still stay on the planet. Neil Harvey was kind enough to open up
his machine and let us Telnet in -- and again I was impressed. Bandwidth to
the RSA is the problem. If you've got it, communication is quite good and
Neil's HP3000 was no less responsive than Rene's or Stan's. However, there are
also periods during the day when South Africa is essentially unreachable by
Telnet. Ten o'clock in the evening in New Mexico is midnight on the east coast
of the US and four in the morning in South Africa. At that time of night, I
found that I basically had that part of the world to myself and Neil seemed as
if he were next door (indeed, telnetting into our ISP, which is only a mile
away, was not greatly all that more responsive; running a trace, I found that
telnetting into our ISP resulted in a 500-mile route, up to Albuquerque and
back).

There are currently problems with the half-duplex, "advanced" Telnet. If the
HP3000 executes the FCONTROL 12/13 sequence to turn remote echo on and off,
MPE doesn't currently inform the Telnet client of the state change -- thus we
get double echoing of characters as a result. As a consequence, and as a
workaround, we put a "echo" suppress button on the bottom of the screen.
Should you get double echoing, simply press the button and the appropriate
Telnet commands will be sent to the HP3000 to re-suppress echo.

We've never found this problem to occur on UNIX boxes. It is only a problem on
the HP3000, but HP is aware of it. Jeff Vance was kind enough to get me in
touch with Jeff Bandle of CSY's MPE Networking Lab. Jeff (the latter) has been
very open and extremely cooperative. I suspect that the problem will be
cleared soon -- although, of course, it will require a patch to MPE.

BTW, we've called the half-duplexing option "advanced" simply because it seems
that way to the person using it. Calling anything "half" just isn't good
marketing -- but in truth, it's not an advance. It is simply a re-adoption of
a technique that was common 30 years ago.

Finally, a year ago, I said that QCTerm was going to be a Christmas present
(of sorts). It is a good thing that I didn't specify which Christmas.
Secondly, you can see for yourself that the HP3000 community spirit is still
alive and well by the number of people I have thanked in this one message
alone.

Wirt Atmar