TUHS December 2017

tuhs@tuhs.org

82 participants
93 discussions

by Warner Losh

>From the tuhs web site: "This is a set of addenda to Seventh Edition Unix, possibly put out by the Labs." and "The identity of the person who donated them is unknown." Two questions: Was this put out by the Labs? Second: There was recently a discussion about a tape found at some public location during an early Unix user group meeting. Is this that tape? Warner

6 years, 4 months

Re: [TUHS] Date madness

by jnc＠mercury.lcs.mit.edu

> From: Arnold > So how did you manage to lose a whole year? Time travel? :-) No, I just haven't booted the V6 often in the last year, and when I did, I never bothered to reset the time; so it was still on Nov, 2016. Noel

6 years, 4 months

Re: [TUHS] Date madness

by jnc＠mercury.lcs.mit.edu

> So, I looked at the code, and the bug is not obvious. Because there _was_ no bug!!! :-) It was saying December 12 was a Monday because the _year_ was still set to _2016_, and in 2016, December 12 _was_ a Monday! {Smacks forehead!} Noel

6 years, 4 months

Re: [TUHS] Date madness

by jnc＠mercury.lcs.mit.edu

> Guess the date code is probably missing the '400 year' exception. So, I looked at the code, and the bug is not obvious. (The routine for detecting leap years in V6's ctime.c is so minimalistic it actually works for 2000!) The version I'm currently running had been worked on some, to have a hack fix for the overflow of the number of 8-hour periods since the epoch ("hack" since the 'fixed' version will break in 2029 or so), and I was worried that 'fix' was wrong and caused the week-day problem, but I don't think so. Noel

6 years, 4 months

The French SOL Project

by Clem Cole

In the early 1980s, a bunch of French researchers set out to build a clone of UNIX/V7 in Pascal (using a ukernel IIRC). The project was the SOL project [Gien, 1983]. I believe it eventually begat the Chorus system (which was C++); which UI was going to use for System V/R5 before it all blew up. 1) Does anyone know what happen to SOL? Was it finished, deployed, used for anything? 2.) Did the sources and doc survive (and who owns the IP)? I think those should be in the TUHS archives, as I think this was the first attempt at a rewrite of UNIX in something other than C (or assembler). 3.) On a similar thought, did the Chorus code survive and who owns the IP? Clem [Gien, 1983]*“The SOL Operating System*”, Michel Gien, USENIX Association, 1983, Proceedings of the Summer ’83 USENIX Conference, Toronto, On, Canada, July, 1983, Pages 75-78

6 years, 4 months

Re: [TUHS] Early Clones

by Rudi Blom

actually not an early clone but a late(r), Philips MPX (Multi-processor UNIX) based on SystemV running on Moterola CPUs. Mostly sold in the (Retail) Banking Industry as branch server. The "M" was more related to having various boards running their own UNIX. That would be an Ethernet board and one Networking board for SDLC and X.25. These boards run their own copy of unix to 'offload' the main CPU. I've worked with this UNIX but more a system tester, interface to National Sales Organisations. No idea where the source is, but guess it would be IP of Philips (or DEC, or Compaq, or HP, or HPE) if anyone still remembers that is. For a (very) short while there was also an Intel CPU based version, but dropped in favour of SCO UNIX when DEC bought Philips Computer division. A very little bit in this 1987 article. https://www.cbronline.com/news/better_late_philips_enters_the_uk_unix_marke… Cheers, uncle rubl

6 years, 4 months

Re: [TUHS] Date madness

by jnc＠mercury.lcs.mit.edu

> From: Warner Losh > So at some point you had to hack things in to make 20xx work, no? I had a number of date issues with V6; more here: http://mercury.lcs.mit.edu/~jnc/tech/V6Unix.html#Issues That one was the easiest to fix! (Although this one will probably also be pretty easy, too.) Noel

6 years, 4 months

Re: [TUHS] V7 Addendem [ really lawyers and AT&T consent decree ]

by jnc＠mercury.lcs.mit.edu

> From: Clem Cole > IP and datagrams were very much built on no central control Well, yes and no. One can easily have a centrally controlled datagram network (q.v. the ARPANET) - although it's true that its path-selection algorithms, etc were not centrally controlled - but other aspects of the network were. (Interestingly, after various routing disasters the Internet caused by improper configuration, some aspects of path selection in _parts_ of it are now effectively centrally controlled; but I digress.) The IP Internet was designed with no _overall_ central control, but as a collection of autonomous entities. > In the end, it was MetCalfe's law (which was formulated on observations > about the phone system) that caused IP to win Over any and all comers, including other decentralized datagram networks like CLNP. MetCalfe's law doesn't talk about decentralized, it's just about 'first to field'. > all want to see the net neutrality go away This whole 'net neutrality' campaign drives me completely crazy. If all people wanted was a rule saying 'ISPs can't give third parties _worse_ service, or - more importantly - deny service altogether, unless those parties pay up' (i.e. what would amount to targeted extortion), I'd be _all for_ a rule like that. But the 'net neutrality' aficionados (most of whom, I'm fairly sure, are not aware of/thinking about these details) are all signing up for a much more expansive rule, one that says 'no ISP can offer anyone _better_ service for paying more money' - which is quite different. My problems with this latter form are two-fold. First, what's wrong with that anyway? Do we have a rule saying you can't get better road service if you pay? Absolutely not - restricted toll lanes are becoming more and more common. So there's clearly no societal agreement on this principle. (I suspect this 'net netrality' campaign has as a goal some sort of 'forced equality' thing - unless the people behind it simply don't even understand the difference.) Second, that rule is, with a little extra work on the ISPs' part, ineffective anyway. All they have to do is build _two_ networks, one better provisioned than the other - and priced accordingly. You want better service? Sign up for the second network; you'll pay more, but it's your choice. Noel

6 years, 4 months

Re: [TUHS] V7 Addendem [ really lawyers and AT&T consent decree ]

by jnc＠mercury.lcs.mit.edu

> From: George Michaelson > I didn't mean to disrespect the people who did the models or the > protocol or standards work btw. Oh, I personally have no problem with you saying hard things about this work, as long as it's what you really think. One never finds the truth unless one is willing to look hard, neh? And I'm pretty sure Dave Clark (who was a leading light in IntServ) would be OK with you doing so too (I worked very closely with him for years, to the point where I deputized for him at meetings). I honestly don't remember exactly what my take was on IntServ and RSVP; I'd have to go look. I recollect seeing the case that _if resources were limited_, certain classes of application (I guess we called them 'inelastic') would need reservations to work properly. So I was probably susceptible to the argument that 'if we've got bandwidth to light our <insert flammable objects> with, we don't need resource reservation'. And I remember being not _thrilled_ with RSVP, but I don't recall exactly why. But, as, you said, wrong list. Maybe internet-history: http://mailman.postel.org/mailman/listinfo/internet-history if you did want to delve into it. Noel

6 years, 4 months

Re: [TUHS] MERT?

by jnc＠mercury.lcs.mit.edu

> From: Jon Steinhart > Probably the best person to ask is Heinz Lycklama. I checked my email log for my previous MERT enquiries, and that's the exact same advice I got from Brian Kernighan when I asked him (a couple of years back) if he had any suggestions for tracking it down. I guess I need to finally get a 'round tuit'... unless there's someone else here who knows him well, and wants to give it a go. Noel

6 years, 4 months

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

TUHS December 2017