As some of you may have noticed, I can't predict development times worth a
hoot. However, QCTerm is approaching releaseable quality -- but only as a
rough-cut, first-draft, alpha copy (it's important for you to repeat this
last phrase out loud three times in order to be sure that it's well
understood :-).

It is my nature not to want to release anything until it is as close to dead
perfection as we can possibly get. However, no software can ever reach that
level of quality without a fairly wide exposure to the real world. We have
been internally testing QCTerm extensively with the most demanding
HP3000-based programs we have -- and it works well on all of them. However,
Gavin Scott, just yesterday, had the honor of being the first to download an
older version of QCTerm and he found a few pecadillos which we had not seen
before (and which will be corrected quite soon).

Even one mistake or mishandling of an escape sequence can lead to a fairly
high disappointment index in a first-time user of a new product -- and the
impression of that level of disappointment can stay with the user for quite
some time. I hate having that occur -- and that is the reason for my
reluctance. Thus, the question is:

     Do you all want to be able to download the "rough-cut, first-draft,
alpha"
     copy and essentially all be alpha/beta testers -- or would you prefer
that
     we find about 50 divergent users and only release when they've all had a
     chance to run it through their wringers?

The second option means that the general release of QCTerm will be delayed by
one or two months. If we release it now, in its "rough-cut, first-draft,
pre-alpha" state, we will want everyone who downloads this version to stand
up every day, without fail, and turn their face towards Las Cruces, New
Mexico at 12:00 noon (local time) and raise up their right hand and repeat
the following mantra:

     "I understand that this is only a rough-cut, first-draft, pre-alpha
version!
      I gladly accept the slings and arrows of brand-new software! I further
      understand that if I find the emulator unsatisfactory that I am
entitled
      to a full refund on any monies that I have paid AICS Research!"

Ultimately, I expect -- and demand -- perfection (within the guidelines of
what we are doing and how we have defined the terminal emulator). But it
isn't likely to happen on this first release.

Perhaps the most disturbing problem with the emulator as it presently stands
for some of the more technically savvy/vocal people on this list is that it
is this same subgroup of users were the first to give up serial
communications and have been switched over to LAN-based communications for
some time now. However, the emulator is currently only serially-based at the
moment. It has always been our intentions of adding TCP/IP, Telnet, and
possibly full NS/VT support at the end of the development process. It is
important that we're doing this development primarily for our customer base
-- and among these people, LAN-only communication is still relatively rare.
Most of our customers still use DTCs, thus please take this consideration
into account. Full LAN communications will eventually be supported, but it
isn't yet.

The original specifications and development schedule are still up on the web
at:

     http://aics-research.com/emulate.html

QCTerm is put together differently than any HP terminal or third-party
emulator. We decided to make it more like an internet browser than
Reflection. And it uses the same CNTL- shortcut keys that are now becoming
standard on most Windows-based programs -- and that have been standard on
Macs since 1984.

The screen is always 200 characters wide, regardless of many characters can
be displayed on your PC's CRT window, and is always 600 lines in depth. We've
now tested (to some degree or another) the code on Windows 3.1, Windows 95,
Windows 98 (beta), and Windows NT 4. The code was written in 16-bit mode to
work on Windows 3.1 as its minimum system -- however it works best with
Pentium-class machines. It was also designed to be, as much as Windows allows
it, to be completely "drop-on" code, requiring no "setup". QCTerm doesn't
modify any .INI file on the system, so it doesn't require a de-installer. All
that you need do to get rid of it is delete the AICS folder and remove its
fonts (using Windows' control panel).

Any comments regarding the question of when to release this version (now or
later?) would be much appreciated.

Wirt Atmar