wales@valeria.cs.ucla.edu (05/10/89)
The following material summarizes my experience with Telix 3.11. I am
posting it in the hopes it may be useful to other Telix users.
Also included below is a modified VT102 "termcap" entry which seems to
work well under Telix.
I have previously passed most of these comments on to the Telix author,
Colin Sampaleanu (though I never received any response acknowledging my
letter). However, if someone on this newsgroup who is in frequent touch
with him could forward this material to him again anyway, it probably
couldn't hurt.
Also, I would be very interested in knowing when a new release of Telix
(3.12?) will appear. I've checked the Exis Inc. BBS in Toronto every
so often, but as of last week there was no sign of a new version and no
announcement of the imminent coming of a new version.
-- Rich Wales // UCLA Computer Science Department // +1 (213) 825-5683
3531 Boelter Hall // Los Angeles, California 90024-1596 // USA
wales@CS.UCLA.EDU ...!(uunet,ucbvax,rutgers)!cs.ucla.edu!wales
"This is yet another example of how our actions have random results."
========================================================================
GOOD POINTS
==> Built-in ZMODEM support.
==> Telix's scrollback buffer size can be set to up to 64K bytes -- much
larger than PROCOMM PLUS's 10K bytes. (I find I do a lot of scroll-
ing back, so this is a very welcome feature in my book.)
==> The SALT programming language is *much* more flexible than Procomm
Plus's ASPECT language. One feature which looks very promising --
though I haven't tried it yet -- is the "track" facility, which lets
you look for multiple chracter strings from the input concurrently.
========================================================================
OBVIOUS BUGS
==> ANSI/VT102 emulation: "^[[@" (insert character) is treated as a
no-op (presumably because it is being given a default count of 0).
However, "^[[1@" (with an explicit count) works.
==> Dialing directory editing: SPACE parity gets reset by default to
NONE if you pass through the existing setting via RETURN -- even if
the entry already has some other value. This is serious: if you
aren't careful, you could easily end up with an "N71" terminal mode
(which generally won't work *at all*).
==> Default initial terminal mode: If you try to set the default mode
(in the global configuration) to be 7 bits with parity, the parity
will be forgotten and will always change back to NONE the next time
you run Telix. The only workaround seems to be to set an 8-bit
default mode (N81). This is a serious problem, not only because
most hosts won't respond to an N71 mode, but also because modems
generally won't understand N71 mode and thus won't process the ini-
tialization string.
========================================================================
MISFEATURES
==> Dialing directory editing: There is no way to go back to change a
previous parameter in a dialing directory entry. (I.e., there
should be a key that moves opposite to RETURN.) As it is now, you
just have to hit RETURNs to step all the way through to the end of
the entry -- then edit it again from the beginning.
==> Dialing directory editing: When a new entry is added, it gets the
default baud, parity, data bits, and stop bits (which is good). But
it always gets "terminal = ANSI-BBS", "local echo = OFF", "add line
feed = OFF", "strip high bit = OFF", and "destructive BS = ON",
regardless of default settings for these attributes. It would be
better if these defaults were copied into a new dialing directory
entry.
==> Terminal setup: If you change parity from "space" to "none", it
ends up showing on the screen as "Nonee" (extra "e" is not cleared).
This is just a cosmetic flaw (though note my earlier comment about
how the parity is not remembered from one session to the next).
==> Terminal setup: The expected responses when one wants to change the
"Add LF after CR?" setting are shown as "Yes/No" -- but, in fact,
the program expects N ("on") or F ("off") instead of Y/N.
==> BREAK length: There is no global setting for default BREAK length
(though whatever you get when you press ctrl-END seems to work fine
on the systems I talk to). Also, it is painful to have to specify
the BREAK length in *every* "send_brk" call in SALT scripts. There
should be a default global BREAK length setting -- changeable both
from the keyboard and from SALT -- and there should be a way to tell
"send_brk" to use the default length (possibly via a zero argument).
==> Date/time display: Although the SALT language lets you change dis-
play formats for time (12 vs. 24 hour mode) and date (MM/DD/YY,
DD/MM/YY, YY/MM/DD), there is no way provided to change these set-
tings from the keyboard. If, for example, you want 24-hour time
display mode as a default, the only way to do it seems to be to run
a SALT script with "_time_format = 1;" every time you use Telix.
This deficiency is especially annoying when you realize that Telix's
default time display format uses 12-hour mode (with no accompanying
"AM" or "PM" indication). If the author can't or won't add a way
to change the default date/time display format, he should at least
change the time display default to 24-hour mode.
==> "Underlined" text on color displays: Some people have reported that
Telix shows underlined text with a blue background (making it hard
to see with certain color combinations). Since I have a monochrome
display right now, I haven't noticed this problem myself. I assume
it can be worked around by either selecting different screen colors
or making a customized "termcap" entry (with some other display
attributes for the "us" and/or "so" parameters).
==> Video attributes for new lines: When the screen scrolls, new lines
at the bottom of the screen are *not* given the proper foreground
and background video attributes from the setup menu. (On a mono-
chrome monitor, I tried black foreground on light grey background.
New lines were still created with white on black.)
========================================================================
WISH LIST
==> Dialing directory: There should be a way to copy an existing entry
to a new slot, in case two or more entries are almost identical.
==> Dialing directory: There should be a way to refer to entries sym-
bolically (instead of just by number). This is especially critical
with SALT scripts: as things now stand, if you do "dial" calls from
scripts, you will have to recompile all these scripts if you ever
insert or delete entries from the middle of your dialing directory.
One workaround for this problem is to put each phone number which
you are going to dial from a script into a separate dialing direc-
tory. Then, the script can select the dialing directory (by name)
and just dial entry #1.
==> Dialing directory: A dialing directory file name without an exten-
sion should be automatically augmented with an extension of ".FON".
The Telix author apparently thought he was adding flexibility by not
enforcing a standard filename extension for dialing directories, but
in the DOS world I think it is better to add default extensions when
none are specified by the user.
==> Dialing directory: Current dialing directory file name should be
converted to all capitals for display on the screen (e.g., in the
alt-Z status display, as well as the "load new dialing directory"
screen).
========================================================================
# VT102 termcap entry
# Modified for Telix 3.11 (1/11/89) by Richard B. Wales, UCLA
#
# Added: :dc=\E[P: (delete character)
# :ic=\E[1@: (insert character -- see notes)
#
# Removed: All null-padding
#
# Changed: :li=#25: (was :li=#24:)
# Changed: :so=\E[1m: (was :so=2\E[7m:)
# Changed: :us=\E[1m: (was :us=2\E[4m:)
#
# NOTES:
#
# "\E[@" should work as "insert character" -- but Telix 3.11
# treats this sequence as a no-op and will do the operation only
# if a count of "1" is explicitly inserted ("\E[1@").
#
# Turn off the "status line" if you use "li=#25"; otherwise, keep
# the original "li=#24".
#
# The "so" and "us" video attributes shown generate high-intensity
# video.
#
dy|telix|telix-vt102|dec vt102:\
:bs:\
:cd=\E[J:\
:ce=\E[K:\
:cl=\E[;H\E[2J:\
:cm=\E[%i%d;%dH:\
:co#80:\
:cs=\E[%i%d;%dr:\
:dc=\E[P:\
:do=^J:\
:ho=\E[H:\
:ic=\E[1@:\
:is=\E[1;24r\E[24;1H:\
:k1=\EOP:\
:k2=\EOQ:\
:k3=\EOR:\
:k4=\EOS:\
:kb=^H:\
:kd=\EOB:\
:ke=\E[?1l\E>:\
:kl=\EOD:\
:kr=\EOC:\
:ks=\E[?1h\E=:\
:ku=\EOA:\
:le=^H:\
:li#25:\
:mb=\E[5m:\
:md=\E[1m:\
:me=\E[m:\
:mr=\E[7m:\
:nd=\E[C:\
:pt:\
:rc=\E8:\
:rs=\E>\E[?3l\E[?4l\E[?5l\E[?7h\E[?8h:\
:sc=\E7:\
:se=\E[m:\
:so=\E[1m:\
:sr=\EM:\
:ue=\E[m:\
:up=\E[A:\
:us=\E[1m:\
:ve=\E[?7h:\
:vs=\E[?7l:\
:vt#3:
========================================================================
-- Rich Wales // UCLA Computer Science Department // +1 (213) 825-5683
3531 Boelter Hall // Los Angeles, California 90024-1596 // USA
wales@CS.UCLA.EDU ...!(uunet,ucbvax,rutgers)!cs.ucla.edu!wales
"This is yet another example of how our actions have random results."