[TUHS] A Talk on Early Unix

[TUHS] A Talk on Early Unix

[Warren Toomey]

All, I was invited to give a talk at a symposium in Paris
on the early years of Unix. Slides and recording at:
http://minnie.tuhs.org/Z3/Hapop3/

Feel free to point out the inaccuracies :-)
For example, I thought Unix was used at some point
as the OS for some of the ESS switches in AT&T, but
now I think I was mistaken.

That's a temp URL, it will move somewhere else
eventually.

Cheers, Warren

[TUHS] A Talk on Early Unix

[arnold]

> For example, I thought Unix was used at some point as the OS for some
> of the ESS switches in AT&T, but now I think I was mistaken.

I think that your initial statement is correct. At Georgia Tech AT&T gave
us some 3B20s - vax size and larger, running System V, and I remember some
discussion that they were used inside AT&T as switches.

This, of course, is second hand information; perhaps some of the Bell Labs
alumni can verify for real.

Arnold

[TUHS] A Talk on Early Unix

[Andrew Warkentin]

On 6/30/16, arnold at skeeve.com <arnold at skeeve.com> wrote:
>> For example, I thought Unix was used at some point as the OS for some
>> of the ESS switches in AT&T, but now I think I was mistaken.
>
> I think that your initial statement is correct. At Georgia Tech AT&T gave
> us some 3B20s - vax size and larger, running System V, and I remember some
> discussion that they were used inside AT&T as switches.
>
> This, of course, is second hand information; perhaps some of the Bell Labs
> alumni can verify for real.
>
> Arnold
>


AFAIK the later ESS switches include a 3B machine but it only handles
some administrative functions, with most of the the actual call
processing being performed in dedicated hardware.

[TUHS] A Talk on Early Unix

[Andrew Warkentin]

On 6/30/16, Andrew Warkentin <andreww591 at gmail.com> wrote:
> On 6/30/16, arnold at skeeve.com <arnold at skeeve.com> wrote:
>>> For example, I thought Unix was used at some point as the OS for some
>>> of the ESS switches in AT&T, but now I think I was mistaken.
>>
>> I think that your initial statement is correct. At Georgia Tech AT&T gave
>> us some 3B20s - vax size and larger, running System V, and I remember
>> some
>> discussion that they were used inside AT&T as switches.
>>
>> This, of course, is second hand information; perhaps some of the Bell
>> Labs
>> alumni can verify for real.
>>
>> Arnold
>>
>
>
> AFAIK the later ESS switches include a 3B machine but it only handles
> some administrative functions, with most of the the actual call
> processing being performed in dedicated hardware.
>

I really should stop using Gmail's web interface and find an MUA that
has a "reply to list" feature and defaults to replying to the list for
messages from the list (it would presumably need explicit filters to
detect messages from some lists because not all lists use the
List-Post header). It's annoying when I try to reply to the list and
instead end up replying to only the poster (I know about the problems
with Reply-To munging; I'm not sure why it isn't more common for MUAs
to have a "reply to list" feature to better deal with lists that don't
munge Reply-To).

[TUHS] A Talk on Early Unix

[Steve Nickolas]

On Thu, 30 Jun 2016, Andrew Warkentin wrote:

> I really should stop using Gmail's web interface and find an MUA that
> has a "reply to list" feature and defaults to replying to the list for
> messages from the list (it would presumably need explicit filters to
> detect messages from some lists because not all lists use the
> List-Post header). It's annoying when I try to reply to the list and
> instead end up replying to only the poster (I know about the problems
> with Reply-To munging; I'm not sure why it isn't more common for MUAs
> to have a "reply to list" feature to better deal with lists that don't
> munge Reply-To).

I have to go through a bunch of finagling to reply to some lists, 
including this one, via Alpine.  At least if I select "reply all", I can 
just cut-and-paste the addresses.  I think it's configurable in the list 
manager daemon.

I do seem to remember it being a bigger pain with Google mailing lists.

-uso.

[TUHS] List reply-to

[Warren Toomey]

On Thu, Jun 30, 2016 at 04:39:22AM -0600, Andrew Warkentin wrote:
> I really should stop using Gmail's web interface and find an MUA that
> has a "reply to list" feature and defaults to replying to the list for
> messages from the list.

I'm running a fairly default Mailman setup for TUHS. The reply-to option
is set to be the original poster. I can change it to the list address if
most people are happy for it to be that way.

Mind you, I'm sure either choice won't be suitable for all subscribers.

Cheers, Warren

[TUHS] A Talk on Early Unix

[Nemo]

On 30 June 2016 at 06:30, Andrew Warkentin <andreww591 at gmail.com> wrote:
> On 6/30/16, arnold at skeeve.com <arnold at skeeve.com> wrote:
>>> For example, I thought Unix was used at some point as the OS for some
>>> of the ESS switches in AT&T, but now I think I was mistaken.
>>
>> I think that your initial statement is correct. At Georgia Tech AT&T gave
>> us some 3B20s - vax size and larger, running System V, and I remember some
>> discussion that they were used inside AT&T as switches.
>>
>> This, of course, is second hand information; perhaps some of the Bell Labs
>> alumni can verify for real.
>>
>> Arnold
>>
>
> AFAIK the later ESS switches include a 3B machine but it only handles
> some administrative functions, with most of the the actual call
> processing being performed in dedicated hardware.

I feel dated:  2ESS was just coming in and offices had Strowger
switches when I left that field. Wikipedia
(https://en.wikipedia.org/wiki/5ESS_Switching_System) claims that 5ESS
ran under UNIX-RTR.

N.

[TUHS] List reply-to

[Steve Nickolas]

On Thu, 30 Jun 2016, Warren Toomey wrote:

> On Thu, Jun 30, 2016 at 04:39:22AM -0600, Andrew Warkentin wrote:
>> I really should stop using Gmail's web interface and find an MUA that
>> has a "reply to list" feature and defaults to replying to the list for
>> messages from the list.
>
> I'm running a fairly default Mailman setup for TUHS. The reply-to option
> is set to be the original poster. I can change it to the list address if
> most people are happy for it to be that way.
>
> Mind you, I'm sure either choice won't be suitable for all subscribers.
>
> Cheers, Warren
>

Of course nothing will please everybody. xD

With Alpine at least I think if it's set to reply-to the list address by 
default, then it'll just ask me when I hit reply, whether I want to use 
the reply-to address or not, so I can have it both ways.  Dunno about more 
"civilized" mail clients though - I've barely ever used anything that 
wasn't Pine. ;)

-uso.

[TUHS] A Talk on Early Unix

[Dave Horsfall]

On Thu, 30 Jun 2016, Andrew Warkentin wrote:

> I really should stop using Gmail's web interface and find an MUA that 
> has a "reply to list" feature and defaults to replying to the list for 
> messages from the list (it would presumably need explicit filters to 
> detect messages from some lists because not all lists use the List-Post 
> header). It's annoying when I try to reply to the list and instead end 
> up replying to only the poster (I know about the problems with Reply-To 
> munging; I'm not sure why it isn't more common for MUAs to have a "reply 
> to list" feature to better deal with lists that don't munge Reply-To).

I'm a FreeBSD/Alpine bigot myself, but I've heard good things about 
Linux/Mutt.

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

[TUHS] List reply-to

[Dave Horsfall]

On Thu, 30 Jun 2016, Warren Toomey wrote:

> I'm running a fairly default Mailman setup for TUHS. The reply-to option 
> is set to be the original poster. I can change it to the list address if 
> most people are happy for it to be that way.
> 
> Mind you, I'm sure either choice won't be suitable for all subscribers.

You'll make enemies either way :-)

I once ran a list with Reply-To set to the list, and people used to 
replying to sender posted some sensitive information...

My preference is "list", and lusers will quickly learn after their first 
mistake...  This is a technical mailing list, after all.

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

[TUHS] List reply-to

[Random832]

On Thu, Jun 30, 2016, at 08:17, Warren Toomey wrote:
> I'm running a fairly default Mailman setup for TUHS. The reply-to option
> is set to be the original poster. I can change it to the list address if
> most people are happy for it to be that way.
> 
> Mind you, I'm sure either choice won't be suitable for all subscribers.

My own email works fine with it as-is, it shows a "reply to list" button
based on, I assume, the List-Post header.

I'd worry that changing the Reply-To header to be the list would (in
addition to obliterating the original reply-to for the odd poster that
has it distinct from their from address) make the "reply to sender"
button go to the list and force people to copy the address manually.

[TUHS] List reply-to

[Greg 'groggy' Lehey]

On Thursday, 30 June 2016 at 22:17:30 +1000, Warren Toomey wrote:
> On Thu, Jun 30, 2016 at 04:39:22AM -0600, Andrew Warkentin wrote:
>> I really should stop using Gmail's web interface and find an MUA that
>> has a "reply to list" feature and defaults to replying to the list for
>> messages from the list.
>
> I'm running a fairly default Mailman setup for TUHS. The reply-to option
> is set to be the original poster. I can change it to the list address if
> most people are happy for it to be that way.

I'd be happy to see that change.  Ideally I think that the reply
should be both to the list and to the poster, but if some MUAs find
the concept too difficult, then the list seems the right choice.

Greg
--
Sent from my desktop computer.
Finger grog at FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed.  If your Microsoft mail program
reports problems, please read http://lemis.com/broken-MUA
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: not available
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20160702/860dc472/attachment.sig>

[TUHS] A Talk on Early Unix

[Brad Spencer]

Lyndon Nerenberg <lyndon at orthanc.ca> writes:

[snip]

> Basically, the 3B2s (and the 3B4K by extension) were designed to hang off the side of the 4ESS and collect toll call records for billing purposes.  Anyone remember Tuxedo?

[snip]

> --lyndon


Tuxedo, the database??  FML buffers, etc..  That Tuxedo???

If so, Ya, the system I worked on when I was at AT&T / Lucent in the
mid-90's to early 2000s used it.  The product was a traffic management
system called NTMOS or NetMinder/NTM.  We had the source code to the
version of Tuxedo used by the product, and ran it on the Vax initially
and then ported it to the Mips based Star servers [Tandems], i386 based
NCR systems, probably the Sun Sparc at some point, and lastly the HPs.




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

[TUHS] A Talk on Early Unix

[Steve Simon]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2130 bytes --]

Tuxedo, yes.

I thought, still think it is kinda cool.
I used it at telecom nz, we where trying to userp the hegemony of CICS,
and from what I heard later - failed.

this would have been 1997, it was still going, owned by BEA.

-Steve


> On 1 Jul 2016, at 04:48, Lyndon Nerenberg <lyndon at orthanc.ca> wrote:
> 
> 
>> On Jun 30, 2016, at 5:53 AM, Ronald Natalie <ron at ronnatalie.com> wrote:
>> 
>> Ah yes, the 3B’s.   Running the state university computer department (in NJ) we got a lot of 3B’s (3B2, 3B5, 3B20).
> 
> We had the misfortune of being donated a 3B4000.  
> 
>> The 3B20 was definitely a piece of telephone equipment.
> 
> The 3B4000 had a 3B20 inside that acted as the bootstrap controller (ala the micro-pdp inside the larger Vaxen doing similar duty).  Most impressively, said 3B20, with the CPU and IO power of a match stick, was also the Ethernet portal for the entire 3B4000.  It couldn't even come close to keeping up with its 10 mbit/s NIC.
> 
> AT&T flogged this abomination to us as the core of our "distributed network environment."  Actually, they flogged it to us as a "1 million dollar donation in kind" because they knew nobody would buy that piece of shit, but this way they got the tax write off, and our beloved University president got to grin madly at a press conference.
> 
> When they replaced it with a half-dozen 3B2-xxx servers (about a year later) we at least got a marginal improvement in network throughput.  But they cancelled that out with RFS.
> 
> Basically, the 3B2s (and the 3B4K by extension) were designed to hang off the side of the 4ESS and collect toll call records for billing purposes.  Anyone remember Tuxedo?
> 
>> The 3B5 was an interesting machine.   We found out how rugged it was when a drain pipe broke over the top of it (the Rutgers main computer center was underground under a court yard between the twin towers of the Hill Center).   The thing survived a deluge of water being dumped into it.
> 
> I don't think we ever landed one of those in the shop.  It seemed like an intriguing bit of gear back when I looked at it.  Decades ago, now.
> 
> --lyndon

[TUHS] A Talk on Early Unix

[Lyndon Nerenberg]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1777 bytes --]


> On Jun 30, 2016, at 5:53 AM, Ronald Natalie <ron at ronnatalie.com> wrote:
> 
> Ah yes, the 3B’s.   Running the state university computer department (in NJ) we got a lot of 3B’s (3B2, 3B5, 3B20).

We had the misfortune of being donated a 3B4000.  

> The 3B20 was definitely a piece of telephone equipment.

The 3B4000 had a 3B20 inside that acted as the bootstrap controller (ala the micro-pdp inside the larger Vaxen doing similar duty).  Most impressively, said 3B20, with the CPU and IO power of a match stick, was also the Ethernet portal for the entire 3B4000.  It couldn't even come close to keeping up with its 10 mbit/s NIC.

AT&T flogged this abomination to us as the core of our "distributed network environment."  Actually, they flogged it to us as a "1 million dollar donation in kind" because they knew nobody would buy that piece of shit, but this way they got the tax write off, and our beloved University president got to grin madly at a press conference.

When they replaced it with a half-dozen 3B2-xxx servers (about a year later) we at least got a marginal improvement in network throughput.  But they cancelled that out with RFS.

Basically, the 3B2s (and the 3B4K by extension) were designed to hang off the side of the 4ESS and collect toll call records for billing purposes.  Anyone remember Tuxedo?

> The 3B5 was an interesting machine.   We found out how rugged it was when a drain pipe broke over the top of it (the Rutgers main computer center was underground under a court yard between the twin towers of the Hill Center).   The thing survived a deluge of water being dumped into it.

I don't think we ever landed one of those in the shop.  It seemed like an intriguing bit of gear back when I looked at it.  Decades ago, now.

--lyndon

[TUHS] A Talk on Early Unix

[Doug McIlroy]

> I'm curious if the name "TPC" was an allusion to the apocryphal telephone
> company of the same name in the 1967 movie, "The President's Analyst"?

Good spotting. Ken T confirms it was from the flick.

doug

[TUHS] A Talk on Early Unix

[Dan Cross]

On Thu, Jun 30, 2016 at 8:41 AM, Doug McIlroy <doug at cs.dartmouth.edu> wrote:

> [snip]
>
> Though Unix itself didn't get into switches, Unix people did
> have a significant influence on the OS architecture for
> ESS 5. Bob Morris, having observed some of the tribulations of
> that project, suggested that CS Research build a demonstration
> switch. Lee McMahon, Ken Thompson, and Joe Condon spearheaded
> the effort and enlisted Gerard  Holzmann's help in verification
> (ironically, the only application of Gerhard's methods to
> software made in his own department). They called the system,
> which was very different from Unix, TPC--The Phone Company. It
> actually controlled many of our phones for some years. The
> cleanliness of McMahon's architecture, which ran on a PDP-11,
> caught the attention of Indian Hill and spurred a major
> reworking of the ESS design.
>

I'm curious if the name "TPC" was an allusion to the apocryphal telephone
company of the same name in the 1967 movie, "The President's Analyst"?

        - Dan C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20160630/5d6cdd88/attachment.html>

[TUHS] A Talk on Early Unix

[Ronald Natalie]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 798 bytes --]

Ah yes, the 3B’s.   Running the state university computer department (in NJ) we got a lot of 3B’s (3B2, 3B5, 3B20).
The 3B20 was definitely a piece of telephone equipment.   They way you powered it down was to turn a knob to off and then hold a button down for three seconds until it went twang and turned off.    Anybody remember the original bell 303 modems?  That’s how you’d put them into loopback.   Used to have to do this from time to time on our old ARPANET modems at the request of the NOC.

The 3B5 was an interesting machine.   We found out how rugged it was when a drain pipe broke over the top of it (the Rutgers main computer center was underground under a court yard between the twin towers of the Hill Center).   The thing survived a deluge of water being dumped into it.

[TUHS] A Talk on Early Unix

[Doug McIlroy]

> AFAIK the later ESS switches include a 3B machine but it only handles
> some administrative functions, with most of the the actual call
> processing being performed in dedicated hardware.

That is correct. The 3B2 was an administrative appendage.

Though Unix itself didn't get into switches, Unix people did
have a significant influence on the OS architecture for
ESS 5. Bob Morris, having observed some of the tribulations of
that project, suggested that CS Research build a demonstration
switch. Lee McMahon, Ken Thompson, and Joe Condon spearheaded
the effort and enlisted Gerard  Holzmann's help in verification
(ironically, the only application of Gerhard's methods to
software made in his own department). They called the system,
which was very different from Unix, TPC--The Phone Company. It
actually controlled many of our phones for some years. The
cleanliness of McMahon's architecture, which ran on a PDP-11,
caught the attention of Indian Hill and spurred a major
reworking of the ESS design.

Doug