Date: Wed, 4 Aug 93 04:30:07 PDT
From: Advanced Amateur Radio Networking Group <tcp-group@ucsd.edu>
Errors-To: TCP-Group-Errors@UCSD.Edu
Reply-To: TCP-Group@UCSD.Edu
Precedence: Bulk
Subject: TCP-Group Digest V93 #198
To: tcp-group-digest
TCP-Group Digest Wed, 4 Aug 93 Volume 93 : Issue 198
Today's Topics:
Convers + readline compile
ka9q Network Configuration Q's
Micor DVP (2 msgs)
Refused Mail
Send Replies or notes for publication to: <TCP-Group@UCSD.Edu>.
Subscription requests to <TCP-Group-REQUEST@UCSD.Edu>.
Problems you can't solve otherwise to brian@ucsd.edu.
Archives of past issues of the TCP-Group Digest are available
(by FTP only) from UCSD.Edu in directory "mailarchives".
We trust that readers are intelligent enough to realize that all text
herein consists of personal comments and does not represent the official
policies or positions of any party. Your mileage may vary. So there.
----------------------------------------------------------------------
Date: Wed, 04 Aug 93 12:06:26 CET
From: BARRY TITMARSH <BTITMARS%ESOC.BITNET@vm.gmd.de>
Subject: Convers + readline compile
To: TCP-GROUP <TCP-GROUP@ucsd.edu>,
Fred you will see from this included info from the compiler that
readline compiles ok. but there is still a problem... Have you any ideas
or changes you might have made to the readline-1.1 package
the convers 930803 compiles ok with out readline support.
Or is this feature still being developed ?
Thanks. Barry
gcc -traditional -c -g -I. -DVOID_SIGHANDLER -DVI_MODE \
-I../ readline.c
gc
c -traditional -c -g -I. -DVOID_SIGHANDLER -DVI_MODE \
-I../ history.c
gcc -
traditional -c -g -I. -DVOID_SIGHANDLER -DVI_MODE \
-I../ funmap.c
gcc -trad
itional -c -g -I. -DVOID_SIGHANDLER -DVI_MODE \
-I../ keymaps.c
rm -f librea
dline.a
ar clq libreadline.a readline.o history.o funmap.o keymaps.o
if -f /us
r/bin/ranlib |; then /usr/bin/ranlib libreadline.a; fi
gcc -Wall -O -ggdb -I/tcp/lib -DPOSIX_SOURCE -I/usr/local/include -DREADLINE con
vers.c -c -o convers.o
gcc convers.o /usr/local/lib/libreadline.a -ltermcap /tc
p/lib/libutil.a -o convers
readline.c:673 (/usr/local/lib/libreadline.a(readline
.o)): Undefined symbol _bsd_ioctl referenced from text segment
readline.c:1120
/usr/local/lib/libreadline.a(readline.o)): Undefined symbol _bsd_ioctl reference
d from text segment
readline.c:1135 (/usr/local/lib/libreadline.a(readline.o)):
Undefined symbol _bsd_ioctl referenced from text segment
readline.c:1957 (/usr/l
ocal/lib/libreadline.a(readline.o)): Undefined symbol _bsd_ioctl referenced from
text segment
readline.c:2156 (/usr/local/lib/libreadline.a(readline.o)): Undefi
ned symbol _bsd_ioctl referenced from text segment
readline.c:2162 (/usr/local/l
ib/libreadline.a(readline.o)): Undefined symbol _bsd_ioctl referenced from text
segment
readline.c:2190 (/usr/local/lib/libreadline.a(readline.o)): Undefined sy
mbol _bsd_ioctl referenced from text segment
readline.c:2213 (/usr/local/lib/lib
readline.a(readline.o)): Undefined symbol _bsd_ioctl referenced from text segmen
t
readline.c:2221 (/usr/local/lib/libreadline.a(readline.o)): Undefined symbol _
bsd_ioctl referenced from text segment
readline.c:0 (/usr/local/lib/libreadline.
a(readline.o)): More undefined symbol _bsd_ioctl refs follow
make: *** convers|
Error 1
------------------------------
Date: Tue, 3 Aug 1993 15:54:56 +0500
From: eb15@postoffice.mail.cornell.edu (Ned Bade)
Subject: ka9q Network Configuration Q's
To: TCP-Group@UCSD.EDU, Packet-Radio@ucsd.edu
Dear TCP-Group and Packet Radio Group Members,
My name is Ned Bade, I work for the Cornell International Institute for
Food Agriculture and Development (CIIFAD) at Cornell University, and I am
helping CIIFAD design an academic email network in the Philippines. We are
considering using ka9q NOS as an IP mail router for a network that will be
both telephone and packet-based. We have started to tap into some local
expertise here at Cornell, but I have some general questions I would like
to place to peoples' expertise on the net. The network would need to
support between 50-100 people initially, and I suspect that number could
grow to 250-1000 in 3-5 years. You can resond to me personally at:
eb15@cornell.edu or I also get the TCP Group and Packet Radio digests, and
will see responses that show up there.
The issues that we are trying to determine are:
ka9q setup related:
^^^^^^^^^^^^^^^^^^^
1) How well ka9q works with POP mail. Also, what version or variant of
ka9q NOS works best as a POP mail server, and where to get it.
2) How well it works over a dial-up connection, and over an ethernet
LAN, and any tricks that might help in setting it up.
3) Whether one can reliably use a client software package like NUPOP over a
TCP/IP packet radio connection. We are considering using NUPOP or some
equivalent offline reader/composer to lower connect times, and would like
to be able to use the same client package for all users of the network.
4) How much direct supervision ka9q NOS needs to operate reliably, and
how much technical backgound it requires of an operator.
5) How difficult it is to install and maintain a packet radio
installation running ka9q NOS and what the level of technical
training a sysop would need to install/manage it. Also, how well does
it recover from crashes (for example, from frequent brownouts...).
6) What hardware would be recommended (from the computer itself
(processor, RAM, hard disk, etc.) to the TNC, radio, antenna, etc.)
for a network where we would like to be able to support 1-2 modem
lines, a packet radio station, with 100-500 users (54 users
initially, but I anticipate it will expand with both official and
unacknowledged users). The primary use would be simple email, but
will probably also include some larger document transfers
(uuencoded), and it would be nice if we could also set up an ftp site
and remailer for interest groups. Maximum transmission distances will
be no more than 100 km (we would prefer to use UHF or VHF and avoid HF),
but want a set-up that also takes into account local conditions (i.e.
high temperature, high humidity, long rainy season with *heavy* rainfall,
frequent lightning, etc.).
Other Issues (added in case you might have any experience/answers):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1) Whether it is possible to use NUPOP to connect over a serial (dialup)
line in cases where you need to place the call through an operator.
(ie, if it is possible to write the connection script so that you can
take the modem off line while placing the call through the operator,
and then bring the modem (and NUPOP) online again as soon as you hear
the carrier detect.
2) How difficult (or easy) it is to write gateway scripts to forward
mail to and from other networks (ie FidoNet, UUCP, etc.) from ka9q
NOS.
Those are my main questions for now. If anyone has helpful comments,
suggestions or resources to offer I would greatly appreciate them
(particularly from people who have experience setting up such systems in
developing countries). Also, anyone who would like to send me a copy of
their autoexec.nos file (if they think it might be helpful or instructively
related to our project), they would be welcome. You can resond to me
personally at: eb15@cornell.edu or I also get the TCP Group and Packet
Radio digests, and will see responses that show up there.
Thanks in advance,
Ned Bade
eb15@cornell.edu
eb15@postoffice.mail.cornell.edu
CIIFAD, 350 Caldwell Hall
255-3035
------------------------------
Date: Tue, 3 Aug 1993 11:25:04 -0500 (CDT)
From: Mr. Sampson <ssampson@sabea-oc.af.mil>
Subject: Micor DVP
To: TCP-Group@UCSD.Edu
Just a comparison with DES to Motorola's scrambling algorithm, but I noticed
their proprietary chip has 2^71 keys as opposed to DES 2^56. Seems like a
step down, but who knows what goes on in the algorithm (maybe just XOR)...
I didn't know what the voice quality was, but the recent report confirms my
suspicion that it wasn't too good. Even military high dollar systems of the
70's were basically un-usable. Fighter pilots wouldn't use them, so they were
ballast. The recent stuff is a lot better, even combined with Have Quick
hopping radios. Fighter pilots will actually use it, so it must be good :-)
The significant change I think was getting rid of the CVSD. The DSP vocoder
technology packs the better formants into much lower rates (9.6k being the
minimum I would use).
---
Steve N5OWK
------------------------------
Date: Tue, 3 Aug 93 15:39:04 -0700
From: karn@qualcomm.com (Phil Karn)
Subject: Micor DVP
To: ssampson@sabea-oc.af.mil
>The significant change I think was getting rid of the CVSD. The DSP vocoder
>technology packs the better formants into much lower rates (9.6k being the
>minimum I would use).
Sigh. This is what I was afraid of. I know the newer DSP-based coders
provide better sound quality, but I was hoping that CVSD would at
least be usable. My company sells a very good DSP-based CELP vocoder,
but at the price they're asking it's not yet very interesting to hams
or other cost-sensitive users, I'm afraid.
Phil
------------------------------
Date: 04 Aug 93 04:35:21 EDT
From: Steve Dworkin <70730.220@CompuServe.COM>
Subject: Refused Mail
To: Advanced Amateur Radio G <tcp-group@ucsd.edu>
Im not sure who should get this,but here goes... Is there
some way that when this digest gets refused somewhere,
it doesnt make a reentry back to here. Two days ago I
downloaded a 55k message because someone (nameless)
decided to put an entire publication in here rather than
naming an ftp site. Today, I downloaded 75k only to
find that someone refused the 55k digest and thus it
was re-posted. It makes it tough on those of us that
have to pay for our access. Much obliged.
Steve Dworkin, N2MDQ
------------------------------
End of TCP-Group Digest V93 #198
******************************
******************************