[TUHS] Help request: uucp, mail on 4.2BSD

[TUHS] Help request: uucp, mail on 4.2BSD

[Clem Cole]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1829 bytes --]

On Tue, Mar 7, 2017 at 10:23 AM, Dave Horsfall <dave at horsfall.org> wrote:

> It's been ages since I delved into UUCP; first was the
> ​ ​
> "original", then HoneyDanBer.
>

​Actually this is a great question for this list .. how many
implementations were created?
1.) The original 1978 version that shipped with V7 and 32/V (BSD 4.1 and
4.2)
2,) PC-UUCP for DOS came next -- I never knew how much was ripped off from
the original, because at the time, the Chesson's G protocol was not well
specified.   The authors claimed to have reverse engineered it - I will say
it worked.
3.) Honey-Dan-Ber rewrite - most popular for a long time
4.) Taylor UUCP first real clone that I know of that I do think was done
with out looking at other's source.  G protocol had been publicly
documented by then and the Trailblazer in fact was shipping with the
protocol imbedded in it.

Any others that folks know about and how well were they used?  Did things
like Coherent have a UUCP?   Linux and FreeBSD were able to use to Taylor
UUCP because it became available by then.    Whitesmith's Idris lacked
anything like UUCP IIRC (but was based on V6).   Same with Thoth originally
at Waterloo, but by the time they shipped it as the QNX product it was V7
compliant but I do not remember a UUCP being included in it.    Minux
lacked a UUCP as I recall, but I'm hazy on that has Andy's crew wrote a lot
of the user space.   Coherent was a "full" V7 clone and include things like
the dev tools including yacc/lex and was released much, much before the
Taylor version came out -- so what do they use for uucp if at all?

Does anyone remember any other implementations?

Clem
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170307/f8d811e1/attachment.html>

[TUHS] Help request: uucp, mail on 4.2BSD

[Jaap Akkerhuis]

> On Mar 7, 2017, at 18:17, Clem Cole <clemc at ccc.com> wrote:
> 
> Does anyone remember any other implementations?

The EUUG distributed it's own version, mainly written/maintained
by Piet Beertema.  It introduced the f-protocol.

AT&T shipped an (expensive) commercial version mainly based on
HeneyDanBer.  (Maybe just a different packaged version of that).

	jaap

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 235 bytes
Desc: Message signed with OpenPGP
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170307/b5ee4a23/attachment.sig>

[TUHS] Help request: uucp, mail on 4.2BSD

[Arthur Krewat]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3002 bytes --]

I ran Taylor UUCP on a small BBS system (3 lines!) I ran from my 
apartment in Deer Park, NY, circa 1991-1994 - USENET node kilowatt

It was on System V R4 by Consensys on a 486-33 and later a 486DX2-66 - 
but some things were badly broken, so much so that things like UUCP 
weren't even usable. I don't remember why. Might have had something to 
do with the devices.

So Taylor it was. I have version 1.04 source code.

Hooked up to Motorola (mcdhup) in Hauppauge for USENET news and email, 
using a Telebit Worldblazer, and then disseminated it to various other 
small USENET nodes.

I had quite an interesting routing setup for the mail system included 
with SVR4 (mailsurr?) that would take the UUCP maps posted in 
comp.mail.maps, run them through pathalias and construct a complete path 
to every node in the map. It was the only way I knew of to deal with 
multiple systems I could connect to that would in turn also connect to 
multiple other USENET nodes. I had no definitive "smart host".



On 3/7/2017 12:17 PM, Clem Cole wrote:
>
> On Tue, Mar 7, 2017 at 10:23 AM, Dave Horsfall <dave at horsfall.org 
> <mailto:dave at horsfall.org>> wrote:
>
>     It's been ages since I delved into UUCP; first was the
>     ​ ​
>     "original", then HoneyDanBer.
>
>
> ​Actually this is a great question for this list .. how many 
> implementations were created?
> 1.) The original 1978 version that shipped with V7 and 32/V (BSD 4.1 
> and 4.2)
> 2,) PC-UUCP for DOS came next -- I never knew how much was ripped off 
> from the original, because at the time, the Chesson's G protocol was 
> not well specified.   The authors claimed to have reverse engineered 
> it - I will say it worked.
> 3.) Honey-Dan-Ber rewrite - most popular for a long time
> 4.) Taylor UUCP first real clone that I know of that I do think was 
> done with out looking at other's source.  G protocol had been publicly 
> documented by then and the Trailblazer in fact was shipping with the 
> protocol imbedded in it.
>
> Any others that folks know about and how well were they used?  Did 
> things like Coherent have a UUCP?   Linux and FreeBSD were able to use 
> to Taylor UUCP because it became available by then.    Whitesmith's 
> Idris lacked anything like UUCP IIRC (but was based on V6).   Same 
> with Thoth originally at Waterloo, but by the time they shipped it as 
> the QNX product it was V7 compliant but I do not remember a UUCP being 
> included in it.    Minux lacked a UUCP as I recall, but I'm hazy on 
> that has Andy's crew wrote a lot of the user space.   Coherent was a 
> "full" V7 clone and include things like the dev tools including 
> yacc/lex and was released much, much before the Taylor version came 
> out -- so what do they use for uucp if at all?
>
> Does anyone remember any other implementations?
>
> Clem
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170307/8e4dec71/attachment-0001.html>

[TUHS] Help request: uucp, mail on 4.2BSD

[Jacob Goense]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 782 bytes --]

On 2017-03-07 18:17, Clem Cole wrote:
> 1.) The original 1978 version that shipped with V7 and 32/V (BSD 4.1 
> and 4.2)

 From "Casting the Net":

In 1976, Mike Lesk at Bell Labs came up with a program called UUCP—“UNIX 
to UNIX
copy.” UUCP enabled users to send mail, transfer files, and execute 
remote commands. Lesk
first called it a “scheme for better distribution” (Mini-Systems 
Newsletter, January 1977); but
only a month later it was referred to as UUCP. First designed to operate 
over 300 baud lines,
UUCP was finally published in February 1978.

UUCP was taken up widely and this led to a need for improvements. The 
next version was
written by Lesk and Dave Nowitz, with contributions by Greg Chesson, and 
appeared in
Seventh Edition UNIX in October 1978.

[TUHS] Help request: uucp, mail on 4.2BSD

[Dan Cross]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2785 bytes --]

On Tue, Mar 7, 2017 at 12:17 PM, Clem Cole <clemc at ccc.com> wrote:
>
> On Tue, Mar 7, 2017 at 10:23 AM, Dave Horsfall <dave at horsfall.org> wrote:
>
>> It's been ages since I delved into UUCP; first was the
>> ​ ​
>> "original", then HoneyDanBer.
>>
>
> ​Actually this is a great question for this list .. how many
> implementations were created?
> 1.) The original 1978 version that shipped with V7 and 32/V (BSD 4.1 and
> 4.2)
> 2,) PC-UUCP for DOS came next -- I never knew how much was ripped off from
> the original, because at the time, the Chesson's G protocol was not well
> specified.   The authors claimed to have reverse engineered it - I will say
> it worked.
> 3.) Honey-Dan-Ber rewrite - most popular for a long time
> 4.) Taylor UUCP first real clone that I know of that I do think was done
> with out looking at other's source.  G protocol had been publicly
> documented by then and the Trailblazer in fact was shipping with the
> protocol imbedded in it.
>
> Any others that folks know about and how well were they used?  Did things
> like Coherent have a UUCP?   Linux and FreeBSD were able to use to Taylor
> UUCP because it became available by then.    Whitesmith's Idris lacked
> anything like UUCP IIRC (but was based on V6).   Same with Thoth originally
> at Waterloo, but by the time they shipped it as the QNX product it was V7
> compliant but I do not remember a UUCP being included in it.    Minux
> lacked a UUCP as I recall, but I'm hazy on that has Andy's crew wrote a lot
> of the user space.   Coherent was a "full" V7 clone and include things like
> the dev tools including yacc/lex and was released much, much before the
> Taylor version came out -- so what do they use for uucp if at all?
>

Coherent came with a modified version of Taylor, IIRC. At least in the
later versions; I don't know if they had something else earlier.

Does anyone remember any other implementations?
>

One or more microcomputer BBS (Bulletin Board System) platforms had UUCP
support to bridge their store-and-forward messaging networks to USENET and
send email, etc. The implementation I remember off the top of my head was
Waffle, written by Tom Dell. Waffle later became the basis for the
"influential" (however one chooses to define that...I think they mean that
in the same way as the WeLL is considered influential) Mindvox
Community/early ISP in New York City. I'm sure there were others in that
niche, but I don't know of any off the top of my head. The whole BBS thing
is mildly interesting in its own right, more as a social phenomenon rather
than technically, though.

        - Dan C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170307/89ef7bae/attachment.html>

[TUHS] Help request: uucp, mail on 4.2BSD

[Erik E. Fair]

I have a memory of a "standard" collection of patches for bugs in the version of UUCP before HoneyDanBer UUCP that I think were put together and maintained by Steve McGeady (then of Tektronix, later of Intel and other fame) which effectively became a "version" of UUCP. If one wanted to run a stable, effective site on UUCP (and USENET) one applied the patches to the UUCP sources before starting up.

Unfortunately, my google-fu has not managed to find a reference or repository of those patches ...

	Erik Fair

[TUHS] Help request: uucp, mail on 4.2BSD

[SPC]

2017-03-07 18:43 GMT+01:00 Jaap Akkerhuis <jaapna at xs4all.nl>:
>
>> On Mar 7, 2017, at 18:17, Clem Cole <clemc at ccc.com> wrote:
>>
>> Does anyone remember any other implementations?
>

My experience was a bit weird. I implemented one UUCP bridge for
cc:Mail (Lotus) in my enterprise. More or less between 1990 and 1992.

It worked calling (using a modem) every hour to one phone number in
Madrid and doing an interchange of messages.

Very funny, with some issues with codepages, length of messages, and so.

I think that I keep yet the manuals of this application but I'm not sure at all.

Regards
Sergio Pedraja

[TUHS] Help request: uucp, mail on 4.2BSD

[Dave Horsfall]

On Tue, 7 Mar 2017, Dan Cross wrote:

> One or more microcomputer BBS (Bulletin Board System) platforms had UUCP 
> support to bridge their store-and-forward messaging networks to USENET 
> and send email, etc. The implementation I remember off the top of my 
> head was Waffle, written by Tom Dell. [...]

Was this the UUCP that was available for CP/M?  I found it on the old 
Walnut Creek CD, moved it over to my CP/M box via SneakerNet (I ran CP/M 
for years, carefully avoiding DOS/WinDoze) and it worked; it was overlaid 
to hell and back hence really slow, but it worked.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."

[TUHS] Help request: uucp, mail on 4.2BSD

[John Labovitz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 401 bytes --]

On Mar 7, 2017, at 18:17, Clem Cole <clemc at ccc.com> wrote:

> Does anyone remember any other implementations?

I ported UUCP and either B or C News to the Mac ca. 1989-1990, using the Lightspeed/THINK C compiler on System 6 or 7. Can’t remember exactly how I read news or mail — maybe I wrote a simple GUI? Regretfully, I never released the code, and don’t have it around any more.

—John

[TUHS] Help request: uucp, mail on 4.2BSD

[Steve Simon]

An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170310/8f1ea4b3/attachment-0001.html>

[TUHS] Help request: uucp, mail on 4.2BSD

[Lyndon Nerenberg]

> On Mar 9, 2017, at 4:31 PM, Steve Simon <steve at quintile.net> wrote:
> 
> Maybe its the same one, but I remeber a special UUCP protocol which had its
> MTU and timeouts carefully adjusted to get the best performance across X25.

'f' protocol.  It encoded everything into printable ASCII characters to avoid triggering any PAD escape sequence. 

[TUHS] Help request: uucp, mail on 4.2BSD

[Lyndon Nerenberg]

> On Mar 9, 2017, at 6:04 PM, Lyndon Nerenberg <lyndon at orthanc.ca> wrote:
> 
> 
>> On Mar 9, 2017, at 4:31 PM, Steve Simon <steve at quintile.net> wrote:
>> 
>> Maybe its the same one, but I remeber a special UUCP protocol which had its
>> MTU and timeouts carefully adjusted to get the best performance across X25.
> 
> 'f' protocol.  It encoded everything into printable ASCII characters to avoid triggering any PAD escape sequence. 

Performance wise, it might have been the first "steaming" protocol :-)  All it was concerned about was 1) don't send anything that looks like a PAD escape sequence, 2) just send bytes at the link.

The idea was that the PAD would flow control the session, so uucico just went into dumb mode and shovelled out the bytes.

--lyndon

[TUHS] uucp protocol nits

[Lyndon Nerenberg]

So if we are going to talk UUCP, how can we not bring up the protocol, and it's beloved behaviour, in certain implementations.

'g' protocol was what everyone ran. 64 byte packets, in a three packet window.  By default.  But 'g' could really race along, if provoked.  The window could slide up to seven!  Unless you were running Xenix, where that provoked a core dump.  On most systems, increasing the window size meant binary patching uucico.

I fuzzily remember 'g' implementations that could handle packets up to 256 bytes, but I can't remember now if the basic (pre-HDB) UUCP could deal with that.

HDB cleaned up a lot of things.  While complicating the configuration files to no end.

In parallel to all this, Rick Adams was pounding the living daylights out of the BSD UUCP code. That which ran on seismo. Then uunet.

-- uunet!ncc!lyndon  (so many uucp path sigs ...)

[TUHS] Help request: uucp, mail on 4.2BSD

[Dave Horsfall]

On Fri, 10 Mar 2017, Steve Simon wrote:

> Maybe its the same one, but I remeber a special UUCP protocol which had 
> its MTU and timeouts carefully adjusted to get the best performance 
> across X25.

And there was yet another variation ("s"?) designed for satellite 
circuits.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."

[TUHS] uucp protocol nits

[Larry McVoy]

> In parallel to all this, Rick Adams was pounding the living daylights out of the BSD UUCP code. That which ran on seismo. Then uunet.

Where is Rick these days?  He still doing stuff?  Lots of fond memories
of that guy and that time.

[TUHS] uucp protocol nits

[Lyndon Nerenberg]

> On Mar 9, 2017, at 6:57 PM, Larry McVoy <lm at mcvoy.com> wrote:
> 
> Where is Rick these days?  He still doing stuff?  Lots of fond memories
> of that guy and that time.

The last time I bumped into him was at Interop circa 1993(?) where he was orbiting the BSDi booth that Rob Kolstad was holding down.

I don't know either of them personally, beyond some email interactions.

--lyndon

[TUHS] uucp protocol nits

[Lyndon Nerenberg]

>> Where is Rick these days?  He still doing stuff?  Lots of fond memories
>> of that guy and that time.

But he did mostly vanish after Alternet came to be successful.

I was never sure about the integrity of that whole Usenet loan thing.

--lyndon

[TUHS] uucp protocol nits

[Lyndon Nerenberg]

> I was never sure about the integrity of that whole Usenet loan thing.

s/Usenet/Usenix/

[TUHS] Help request: uucp, mail on 4.2BSD

[Dan Cross]

On Thu, Mar 9, 2017 at 9:51 PM, Dave Horsfall <dave at horsfall.org> wrote:

> On Fri, 10 Mar 2017, Steve Simon wrote:
>
> > Maybe its the same one, but I remeber a special UUCP protocol which had
> > its MTU and timeouts carefully adjusted to get the best performance
> > across X25.
>
> And there was yet another variation ("s"?) designed for satellite
> circuits.


It seems that there was a whole slew of protocols at one time.
Interestingly, the FreeBSD documentation still includes a fairly extensive
description of UUCP (I guess these are really the `info` pages from Taylor
UUCP) that describe a number of different protocols:
https://docs.freebsd.org/info/uucp/uucp.info.Protocols.html

        - Dan C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170309/3ad36b6e/attachment.html>

[TUHS] Help request: uucp, mail on 4.2BSD

[Lyndon Nerenberg]

> On Mar 9, 2017, at 7:45 PM, Dan Cross <crossd at gmail.com> wrote:
> 
> It seems that there was a whole slew of protocols at one time.

g:	the original
G:	a later (HDB?) SVRx version that did 256 byte packets and a seven packet window
f:	X.25 optimized printable-characters-only
x:	similar to above?
z:	Doug Evans wrote this as an alternative to 'f' back when 8-bit paths were not everywhere yet.
i:	'internet'  stream the data. 

g, G, and f, we can get definitions for easily enough I think. For 'z' I can track down Doug, but Taylor UUCP should have the details. 

--lyndon

[TUHS] Help request: uucp, mail on 4.2BSD

[Jim Carpenter]

On Tue, Mar 7, 2017 at 1:30 PM, Dan Cross <crossd at gmail.com> wrote:
>> Any others that folks know about and how well were they used?  Did things
>> like Coherent have a UUCP?   Linux and FreeBSD were able to use to Taylor
>> UUCP because it became available by then.    Whitesmith's Idris lacked
>> anything like UUCP IIRC (but was based on V6).   Same with Thoth originally
>> at Waterloo, but by the time they shipped it as the QNX product it was V7
>> compliant but I do not remember a UUCP being included in it.    Minux lacked
>> a UUCP as I recall, but I'm hazy on that has Andy's crew wrote a lot of the
>> user space.   Coherent was a "full" V7 clone and include things like the dev
>> tools including yacc/lex and was released much, much before the Taylor
>> version came out -- so what do they use for uucp if at all?
>
>
> Coherent came with a modified version of Taylor, IIRC. At least in the later
> versions; I don't know if they had something else earlier.

I don't have a running system right now but Coherent 3.2 used V2
config files (L.sys, L-dev, etc.) according to my 1991 manual. A
Coherent manual with a 1993 copyright confirms that later versions
used Taylor UUCP.

My 80286 with Coherent 3.2.1A (last of the 3's and the last to support
the '286) and its UUCP was how I connected to my first ISP. ($5/month
for mail and news. Who needs SLIP!)

Jim

[TUHS] Help request: uucp, mail on 4.2BSD

[Jaap Akkerhuis]

> On Mar 10, 2017, at 1:31, Steve Simon <steve at quintile.net> wrote:
> 
> Maybe its the same one, but I remeber a special UUCP protocol which had its
> MTU and timeouts carefully adjusted to get the best performance across X25.

Yes. Piet also did the f-protocol

	jaap

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 235 bytes
Desc: Message signed with OpenPGP
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170310/5efb343c/attachment.sig>

[TUHS] Help request: uucp, mail on 4.2BSD

[Erik E. Fair]

The Taylor UUCP is part of pkgsrc (started by NetBSD but supports many, many Unix system platforms; see pkgsrc.org ):

http://cdn.netbsd.org/pub/pkgsrc/current/pkgsrc/net/uucp/README.html

> The Taylor UUCP package provides everything you need to make a UUCP connection.  It currently supports the 'f', 'g' (in all window and packet sizes), 'G', 't' and 'e' protocols, as well a Zmodem protocol, the FX UUCICO 'y' protocol, and two new bidirectional protocols. If you have a Berkeley sockets library, it can make TCP connections. If you have TLI libraries, it can make TLI connections.  It supports a new configuration file mechanism.

The "t" protocol was for use with TCP over the Internet.

	Erik

[TUHS] Help request: uucp, mail on 4.2BSD

[Lars Brinkhoff]

"Erik E. Fair" <fair-tuhs at netbsd.org> writes:
> The Taylor UUCP is part of pkgsrc (started by NetBSD but supports
> many, many Unix system platforms; see pkgsrc.org )
>
> The "t" protocol was for use with TCP over the Internet.

Right.  I tried "t" first, but seismo didn't accept that.  "G" works
better.

[TUHS] Help request: uucp, mail on 4.2BSD

[Joerg Schilling]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2016 bytes --]

Lyndon Nerenberg <lyndon at orthanc.ca> wrote:

>
> > On Mar 9, 2017, at 7:45 PM, Dan Cross <crossd at gmail.com> wrote:
> > 
> > It seems that there was a whole slew of protocols at one time.
>
> g:	the original
> G:	a later (HDB?) SVRx version that did 256 byte packets and a seven packet window
> f:	X.25 optimized printable-characters-only
> x:	similar to above?
> z:	Doug Evans wrote this as an alternative to 'f' back when 8-bit paths were not everywhere yet.
> i:	'internet'  stream the data. 
>
> g, G, and f, we can get definitions for easily enough I think. For 'z' I can track down Doug, but Taylor UUCP should have the details. 

I have:

g:	The original protocol from the 1970s

G:	enhanced g-protocol introduced by Svr4

d:	Protocol for DataKit connections.

e:	Protocol for TCP links from HDB UUCP, similar to t-protocol.
	BSD UUCP used an implementation from Arne Ludwig.

f:	Seven Bit protocol with checksums on the entire file at a time
	No protocol flow control, but XON/XOFF
	It only uses the characters between \040 and \176 (' '..'~')
	Written by Piet Beertema, CWI, Amsterdam, Sep 1984
	Modified for X.25 by Robert Elz, Melbourne Univ, Mar 1985

F:	flow control protocol similar to f-protocol.
	Written by Piet Beertema, CWI, Amsterdam, Sep 1984
	Modified for X.25 by Robert Elz, Melbourne Univ, Mar 1985
	Probably created from f-protocol by Carsten Borman TU-Berlin
	or by me (Jörg Schilling)

h:	A protocol similar to the t-protocol with no error checking.
	Apparently used for HST modems.

s:	High 's'peed protocol based on the g-protocol.
	Variable block sizes 32..4096 bytes, up to 7 windows
	Written by me (Jörg Schilling)

t:	Protocol for TCP links from BSD
	Most likely from Rick Adams

x:	Protocol for X.25 links

Jörg

-- 
 EMail:joerg at schily.net                  (home) Jörg Schilling D-13353 Berlin
       joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.org/private/ http://sourceforge.net/projects/schilytools/files/

[TUHS] uucp protocol nits

[Brad Spencer]

Lyndon Nerenberg <lyndon at orthanc.ca> writes:

> So if we are going to talk UUCP, how can we not bring up the protocol, and it's beloved behaviour, in certain implementations.
>
> 'g' protocol was what everyone ran. 64 byte packets, in a three packet window.  By default.  But 'g' could really race along, if provoked.  The window could slide up to seven!  Unless you were running Xenix, where that provoked a core dump.  On most systems, increasing the window size meant binary patching uucico.
>
> I fuzzily remember 'g' implementations that could handle packets up to 256 bytes, but I can't remember now if the basic (pre-HDB) UUCP could deal with that.
>
> HDB cleaned up a lot of things.  While complicating the configuration files to no end.
>
> In parallel to all this, Rick Adams was pounding the living daylights out of the BSD UUCP code. That which ran on seismo. Then uunet.
>
> -- uunet!ncc!lyndon  (so many uucp path sigs ...)


Back a long time ago, I ran OS/9 on a 6809E Tandy Color Computer 3.  The
relationship to Unix is that it was obviously inspired by it, especially
Vx where x <= 6 [or perhaps 4 or 5, the block diagrams describing OS/9
could have described the older Unix systems ].  One of the items I
worked on quite extensively was the UUCP implementation.  I didn't write
the original C code reimplementation that it used, but modified it quite
a bit and one of the items I added to it was the ability of the g
protocol to handle a bigger packet window and probably to handle bigger
packets.  At the time I dialed it into UUNET once or twice a day for
email and some very small amount of Usenet news.  This all would have
been in the 1992 - 1994 time frame.  So, ya, the UUCP g protocol could
be fiddled with somewhat and it would likely work.



-- 
Brad Spencer - brad at anduin.eldar.org - KC8VKS
http://anduin.eldar.org  - & -  http://anduin.ipv6.eldar.org [IPv6 only]

[TUHS] Help request: uucp, mail on 4.2BSD

[Dave Horsfall]

On Thu, 9 Mar 2017, Erik E. Fair wrote:

> The Taylor UUCP is part of pkgsrc (started by NetBSD but supports many, 
> many Unix system platforms; see pkgsrc.org ):
> 
> http://cdn.netbsd.org/pub/pkgsrc/current/pkgsrc/net/uucp/README.html

FreeBSD also has Taylor UUCP; /usr/ports/net/freebsd-uucp:

    This is a port of the Taylor UUCP Unix-to-Unix Copy Program suite
    of utilities.  This source was formerly a part of the FreeBSD base
    system, and this package is based on the final version of that
    source code, so it includes all previous FreeBSD customizations.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."

[TUHS] uucp protocol nits

[Ron Natalie]

> 'g' protocol was what everyone ran. 64 byte packets, in a three packet
window.  By default.  But 'g' could really race along, if provoked.  The
window could slide up to seven!  Unless you were running Xenix, where 
> that provoked a core dump.  On most systems, increasing the window size
meant binary patching uucico.

Anybody remember the Telebit trailblazer modems that snooped on the g
protocol going through them?

[TUHS] uucp protocol nits

[Arthur Krewat]

Absolutely! We were talking about that earlier in the list, maybe not 
this subject line though.



On 3/10/2017 3:34 PM, Ron Natalie wrote:
>> 'g' protocol was what everyone ran. 64 byte packets, in a three packet
> window.  By default.  But 'g' could really race along, if provoked.  The
> window could slide up to seven!  Unless you were running Xenix, where
>> that provoked a core dump.  On most systems, increasing the window size
> meant binary patching uucico.
>
> Anybody remember the Telebit trailblazer modems that snooped on the g
> protocol going through them?
>
>
>

[TUHS] uucp protocol nits

[Lyndon Nerenberg]

> On Mar 10, 2017, at 12:34 PM, Ron Natalie <ron at ronnatalie.com> wrote:
> 
> Anybody remember the Telebit trailblazer modems that snooped on the g
> protocol going through them?

The company I was at was one of the early resellers of the Trailblazer.  I have fond memories of hauling one out for an on-site demo for a company running some flavour of 3B2.  We hooked it up to a serial port, cu-ed out to our office server, and then I did a simple cat of a large text file to show off the throughput.

While the sysadmins drooled, a growing cohort of office workers started piling up outside the office door asking if the server was down.  The serial port interrupt load was enough to take out all the rest of the terminals in the office ;-)

--lyndon

(We didn't get the sale.  And AT&T stopped referring their customers to us.)

[TUHS] uucp protocol nits

[Corey Lindsly]

> The company I was at was one of the early resellers of the Trailblazer.  =
> I have fond memories of hauling one out for an on-site demo for a =
> company running some flavour of 3B2.  We hooked it up to a serial port, =
> cu-ed out to our office server, and then I did a simple cat of a large =
> text file to show off the throughput.

Anyone remember using cu ~%put / ~%take commands to transfer files across 
dial-up lines on systems that didn't speak uucp? No error correction, 
just blast those bytes. Binary files through an acoustic coupler modem 
were always a particularly tricky proposition. Fun times.

--corey

[TUHS] uucp protocol nits

[Dave Horsfall]

On Fri, 10 Mar 2017, Ron Natalie wrote:

> Anybody remember the Telebit trailblazer modems that snooped on the g 
> protocol going through them?

Not quite sure what you meant by "snooping" (spyware?), but they emulated 
the "g" protocol and went like the clappers.

-- 
Dave Horsfall DTM (VK2KFU)  "Those who don't understand security will suffer."