Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.misc > #76788 > unrolled thread

Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins

Back to article view | Back to comp.os.linux.misc

Contents

  Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-28 20:03 +0000
    Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-10-28 14:10 -0700
      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Allodoxaphobia <trepidation@example.net> - 2025-10-28 22:43 +0000
        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-29 01:45 +0000
          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-10-28 20:45 -0700
            Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-29 17:43 +0000
              Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Chris Ahlstrom <OFeem1987@teleworm.us> - 2025-10-29 14:01 -0400
                Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-30 02:45 +0000
              Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-10-29 22:00 -0400
                Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-30 03:00 +0000
                  Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins John Ames <commodorejohn@gmail.com> - 2025-10-30 08:12 -0700
                    Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-30 20:19 +0000
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Robert Riches <spamtrap42@jacob21819.net> - 2025-10-31 03:00 +0000
                        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-10-30 23:48 -0400
                          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins The Natural Philosopher <tnp@invalid.invalid> - 2025-10-31 08:33 +0000
                        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-31 04:43 +0000
                        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Richard Kettlewell <invalid@invalid.invalid> - 2025-10-31 09:02 +0000
                          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins John Ames <commodorejohn@gmail.com> - 2025-10-31 08:59 -0700
                          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-31 20:24 +0000
                Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Rich <rich@example.invalid> - 2025-11-01 03:40 +0000
                  Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-11-01 00:45 -0400
                    Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Rich <rich@example.invalid> - 2025-11-02 03:15 +0000
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-11-01 23:35 -0400
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins The Natural Philosopher <tnp@invalid.invalid> - 2025-11-02 11:25 +0000
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins "Carlos E.R." <robin_listas@es.invalid> - 2025-11-02 21:08 +0100
                        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-11-02 21:51 +0000
              Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins "Carlos E.R." <robin_listas@es.invalid> - 2025-10-31 14:33 +0100
                Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-31 23:16 +0000
                  Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-11-01 05:41 +0000
                    Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-11-01 02:18 -0400
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-P  icking Origins vallor <vallor@vallor.earth> - 2025-11-01 07:06 +0000
                        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-P  icking Origins rbowman <bowman@montana.com> - 2025-11-02 01:44 +0000
                      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-11-02 01:34 +0000
                  Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins "Carlos E.R." <robin_listas@es.invalid> - 2025-11-01 14:50 +0100
      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-28 22:59 +0000
        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins c186282 <c186282@nnada.net> - 2025-10-30 01:29 -0400
          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-30 06:25 +0000
          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins The Natural Philosopher <tnp@invalid.invalid> - 2025-10-30 06:59 +0000
      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-29 02:54 +0000
        Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins rbowman <bowman@montana.com> - 2025-10-29 17:32 +0000
          Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Nuno Silva <nunojsilva@invalid.invalid> - 2025-11-01 10:30 +0000
      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins The Natural Philosopher <tnp@invalid.invalid> - 2025-10-29 11:22 +0000
      Re: Ken Thompson Recalls Unix’s Rowdy, Lock-Picking Origins Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-30 22:45 +0000

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

#76891

On 10/31/25 23:40, Rich wrote:
> c186282 <c186282@nnada.net> wrote:
>> On 10/29/25 13:43, rbowman wrote:
>>> On Tue, 28 Oct 2025 20:45:45 -0700, Bobbie Sellers wrote:
>>>
>>>>       Copied to my clipboard then used the editor to correct the line
>>> break.
>>>>
>>>> <https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-
>>> origins/>
>>>>
>>>>       Old eyes at least in this person sometimes miss such subleties
>>>> in small print.
>>>>       That is an interesting article by the way.
>>>
>>> Yeah, those were the days.
>>>
>>> "Thompson credits Richard Stallman with developing much more of the open
>>> source philosophy. “But Unix had a bit of that.” Maybe it grew out of what
>>> Dennis Ritchie was remembering, that fellowship that formed around Unix.
>>> “For some reason, and I think it’s just because of me and Dennis,
>>> everything was open…”
>>>
>>> It was just the way they operated. “We had protection on files — if you
>>> didn’t want somebody to read it, you could set some bits and then nobody
>>> could read them, right? But nobody set those permissions on anything … All
>>> of the source was writable, by anybody! It was just open …"
>>>
>>> While I appreciate some of the work of the FSF Stallman's talmudic
>>> legalisms rub me the wrong way. I remember when 'it was just open'.
>>
>>    Yep, that's how it was. Even in DOS *nobody* used
>>    the 'protection' bits.
> 
> DOS had no protection bits on its files in its filesystem.


   Um ... you could mark 'read only' as I remember (been
   awhile). As for real 'protection' ... no. Nobody had
   the idea for something as simple as DOS and how they
   envisioned it being used. No user names/IDs/groups.
   Not sure ANY of that came in before Win-3.11.

   Central archives/backups ... for THAT you used something
   UNIX/VMS/etc.

>>    Of course tangible THREATS were a LOT fewer back
>>    then, just 'theoretical' if you weren't a spook ...
> 
> With DOS they were almost non-existant.  Single user, single tasking.
> As long as you didn't insert every disk you found lying around and run
> every program thereupon willy-nilly, you'd also be safe from the
> eventual viruses that arose.

   Took a little while, but I do remember the "Pakistani
   Brain Virus" that lived in the boot records of floppies.

   But MOST users ... no ... no worries.

[toc] | [prev] | [next] | [standalone]

#76939

c186282 <c186282@nnada.net> wrote:
> On 10/31/25 23:40, Rich wrote:
>> c186282 <c186282@nnada.net> wrote:
>>> On 10/29/25 13:43, rbowman wrote:
>>>> On Tue, 28 Oct 2025 20:45:45 -0700, Bobbie Sellers wrote:
>>>>
>>>>>       Copied to my clipboard then used the editor to correct the line
>>>> break.
>>>>>
>>>>> <https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-
>>>> origins/>
>>>>>
>>>>>       Old eyes at least in this person sometimes miss such subleties
>>>>> in small print.
>>>>>       That is an interesting article by the way.
>>>>
>>>> Yeah, those were the days.
>>>>
>>>> "Thompson credits Richard Stallman with developing much more of the open
>>>> source philosophy. “But Unix had a bit of that.” Maybe it grew out of what
>>>> Dennis Ritchie was remembering, that fellowship that formed around Unix.
>>>> “For some reason, and I think it’s just because of me and Dennis,
>>>> everything was open…”
>>>>
>>>> It was just the way they operated. “We had protection on files — if you
>>>> didn’t want somebody to read it, you could set some bits and then nobody
>>>> could read them, right? But nobody set those permissions on anything … All
>>>> of the source was writable, by anybody! It was just open …"
>>>>
>>>> While I appreciate some of the work of the FSF Stallman's talmudic
>>>> legalisms rub me the wrong way. I remember when 'it was just open'.
>>>
>>>    Yep, that's how it was. Even in DOS *nobody* used
>>>    the 'protection' bits.
>> 
>> DOS had no protection bits on its files in its filesystem.
> 
> 
>   Um ... you could mark 'read only' as I remember (been
>   awhile).

Not a "protection" in the context of what Ritchie was discussing.  He 
was refering to the fact that files can be owned by users/groups and 
marked to only be read or written by those users/groups.

DOS's 'read only' bit was actually a "do not overwrite" bit.  The only 
user was handed an error message instead of being allowed to overwrite 
the file contents.  Of course nothing stopped that same single user 
from flipping the 'read only' bit off and then overwiting the file 
contents.

>   As for real 'protection' ...  no.  Nobody had the idea for 
>   something as simple as DOS and how they envisioned it being used.  
>   No user names/IDs/groups.  Not sure ANY of that came in before 
>   Win-3.11.

Unix had all that long before DOS ever existed.  The concept of more 
than one user and "file ownership" did not appear in the windows world 
until NT arrived.  And even then, the portion exposed by the win APIs 
was a far cry from what Unix had years before DOS existed.

[toc] | [prev] | [next] | [standalone]

#76940

On 11/1/25 23:15, Rich wrote:
> c186282 <c186282@nnada.net> wrote:
>> On 10/31/25 23:40, Rich wrote:
>>> c186282 <c186282@nnada.net> wrote:
>>>> On 10/29/25 13:43, rbowman wrote:
>>>>> On Tue, 28 Oct 2025 20:45:45 -0700, Bobbie Sellers wrote:
>>>>>
>>>>>>        Copied to my clipboard then used the editor to correct the line
>>>>> break.
>>>>>>
>>>>>> <https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-
>>>>> origins/>
>>>>>>
>>>>>>        Old eyes at least in this person sometimes miss such subleties
>>>>>> in small print.
>>>>>>        That is an interesting article by the way.
>>>>>
>>>>> Yeah, those were the days.
>>>>>
>>>>> "Thompson credits Richard Stallman with developing much more of the open
>>>>> source philosophy. “But Unix had a bit of that.” Maybe it grew out of what
>>>>> Dennis Ritchie was remembering, that fellowship that formed around Unix.
>>>>> “For some reason, and I think it’s just because of me and Dennis,
>>>>> everything was open…”
>>>>>
>>>>> It was just the way they operated. “We had protection on files — if you
>>>>> didn’t want somebody to read it, you could set some bits and then nobody
>>>>> could read them, right? But nobody set those permissions on anything … All
>>>>> of the source was writable, by anybody! It was just open …"
>>>>>
>>>>> While I appreciate some of the work of the FSF Stallman's talmudic
>>>>> legalisms rub me the wrong way. I remember when 'it was just open'.
>>>>
>>>>     Yep, that's how it was. Even in DOS *nobody* used
>>>>     the 'protection' bits.
>>>
>>> DOS had no protection bits on its files in its filesystem.
>>
>>
>>    Um ... you could mark 'read only' as I remember (been
>>    awhile).
> 
> Not a "protection" in the context of what Ritchie was discussing.  He
> was refering to the fact that files can be owned by users/groups and
> marked to only be read or written by those users/groups.
> 
> DOS's 'read only' bit was actually a "do not overwrite" bit.  The only
> user was handed an error message instead of being allowed to overwrite
> the file contents.  Of course nothing stopped that same single user
> from flipping the 'read only' bit off and then overwiting the file
> contents.
> 
>>    As for real 'protection' ...  no.  Nobody had the idea for
>>    something as simple as DOS and how they envisioned it being used.
>>    No user names/IDs/groups.  Not sure ANY of that came in before
>>    Win-3.11.
> 
> Unix had all that long before DOS ever existed.  The concept of more
> than one user and "file ownership" did not appear in the windows world
> until NT arrived.  And even then, the portion exposed by the win APIs
> was a far cry from what Unix had years before DOS existed.

   Which was a big reason I switched to Linux file servers
   as soon as practical. Did my best to make it all transparent
   to the office users.

   'Reliability' was another biggie.

   UNIX ... sorry, we didn't have the cash for that.

[toc] | [prev] | [next] | [standalone]

#76948

On 02/11/2025 03:15, Rich wrote:
> DOS's 'read only' bit was actually a "do not overwrite" bit.  The only
> user was handed an error message instead of being allowed to overwrite
> the file contents.  Of course nothing stopped that same single user
> from flipping the 'read only' bit off and then overwiting the file
> contents.

You don't need a condom to have sex with yourself.

-- 
Climate is what you expect but weather is what you get.
Mark Twain

[toc] | [prev] | [next] | [standalone]

#76969

On 2025-11-02 04:15, Rich wrote:
> Unix had all that long before DOS ever existed.  The concept of more
> than one user and "file ownership" did not appear in the windows world
> until NT arrived.  And even then, the portion exposed by the win APIs
> was a far cry from what Unix had years before DOS existed.

On the other hand, the permissions available on Windows 2008 (maybe 
earlier, but that's when I studied them) are very rich and complex. Much 
more than in Linux, which had to add ACLs.

-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#76973

On Sun, 2 Nov 2025 21:08:13 +0100, Carlos E.R. wrote:

> On the other hand, the permissions available on Windows 2008 (maybe
> earlier, but that's when I studied them) are very rich and complex. Much
> more than in Linux, which had to add ACLs.

Linux has ACLs based on a POSIX proposal which, I recall, only made it to 
draft status before being withdrawn. So it’s not actually conformant to 
any official standard.

Windows has a somewhat more complex ACL system, with some additional 
capabilities. There have been submissions to add support for that to the 
Linux kernel, but last I saw they were rejected on the grounds that the 
Windows ACL system can produce results that are often hard to predict. (I 
think seasoned Windows Server sysadmins are familiar with this.)

One ability missing from the Linux/*nix system is the option to assign 
owner rights to some entity (e.g. a group of users) other than the actual 
creator of the file. This can be very useful in scenarios where multiple 
users need read/write access to the same set of files.

[toc] | [prev] | [next] | [standalone]

#76863

On 2025-10-29 18:43, rbowman wrote:
> On Tue, 28 Oct 2025 20:45:45 -0700, Bobbie Sellers wrote:
> 
>> 	Copied to my clipboard then used the editor to correct the line
> break.
>>
>> <https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-
> origins/>
>>
>> 	Old eyes at least in this person sometimes miss such subleties
>> in small print.
>> 	That is an interesting article by the way.
> 
> Yeah, those were the days.
> 
> "Thompson credits Richard Stallman with developing much more of the open
> source philosophy. “But Unix had a bit of that.” Maybe it grew out of what
> Dennis Ritchie was remembering, that fellowship that formed around Unix.
> “For some reason, and I think it’s just because of me and Dennis,
> everything was open…”
> 
> It was just the way they operated. “We had protection on files — if you
> didn’t want somebody to read it, you could set some bits and then nobody
> could read them, right? But nobody set those permissions on anything … All
> of the source was writable, by anybody! It was just open …"
> 
> While I appreciate some of the work of the FSF Stallman's talmudic
> legalisms rub me the wrong way. I remember when 'it was just open'.


I just read the article, and I have to say that it happened because of 
the culture in Bell Labs, and that money was not a problem. And they had 
that money because the Bell company had that monopoly on phone service 
and charged what they charged.

Unix had that open mind because Bell Lab worked with an open mind. Other 
teams there invented things that later the entire world benefited from, 
like the transistor or the chip. Bell had some limitation about not 
patenting and withholding things they could not directly use on phones.


-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#76883

On Fri, 31 Oct 2025 14:33:46 +0100, Carlos E.R. wrote:

> I just read the article, and I have to say that it happened because of
> the culture in Bell Labs, and that money was not a problem. And they had
> that money because the Bell company had that monopoly on phone service
> and charged what they charged.

It is interesting to compare Bell Labs at AT&T, with PARC (the Palo Alto 
Research Center) at Xerox. The former was the birthplace of information 
theory, and Unix, among other things; the latter was the birthplace of the 
GUI, and Smalltalk, and Ethernet, among other things.

Both were essentially research playgrounds for bright people to try out 
risky new ideas, run at the expense of a vast parent corporation that made 
huge profits from a Government-granted monopoly (the US phone monopoly in 
the case of AT&T, patents on photocopiers in the case of Xerox).

And once the parent company lost that monopoly (the AT&T breakup, the 
expiry of those photocopier patents), the glory days at the research 
playground were over.

[toc] | [prev] | [next] | [standalone]

#76895

On Fri, 31 Oct 2025 23:16:04 -0000 (UTC), Lawrence D’Oliveiro wrote:

> Both were essentially research playgrounds for bright people to try out
> risky new ideas, run at the expense of a vast parent corporation that
> made huge profits from a Government-granted monopoly (the US phone
> monopoly in the case of AT&T, patents on photocopiers in the case of
> Xerox).

IBM had their little skunk works too. In the early days of Java they 
released a compiler that was a lot better than the official and had some 
other neat tools. Some middle manager may have said 'Take a look at that 
Java stuff' and they grabbed it and ran.

[toc] | [prev] | [next] | [standalone]

#76897

On 11/1/25 01:41, rbowman wrote:
> On Fri, 31 Oct 2025 23:16:04 -0000 (UTC), Lawrence D’Oliveiro wrote:
> 
>> Both were essentially research playgrounds for bright people to try out
>> risky new ideas, run at the expense of a vast parent corporation that
>> made huge profits from a Government-granted monopoly (the US phone
>> monopoly in the case of AT&T, patents on photocopiers in the case of
>> Xerox).
> 
> IBM had their little skunk works too. In the early days of Java they
> released a compiler that was a lot better than the official and had some
> other neat tools. Some middle manager may have said 'Take a look at that
> Java stuff' and they grabbed it and ran.

   A lot of good stuff did - and still does - come out
   of Big Blue. You don't HEAR about it much anymore
   however, but they're still into *everything*. They
   are not a leading chips name, but their research
   goes into all the new chips. They are not the center
   of the net, but a lot of their stuff goes into the net.
   They are not a leading "AI" company, but their stuff
   is deeply entwined in "AI".

   And they still sell Big Iron for Big Projects/corps.

   IMHO, they have positioned themselves perfectly for
   the early 21st century. Being just off the radar scope
   by and large they will make their money yet not be
   subject to the storms of the more popular tech stocks.
   WHEN "AI" mostly crashes, IBM will still be there.

[toc] | [prev] | [next] | [standalone]

#76898 — Re: Ken Thompson Recalls Unix’s Rowdy, Lock-P icking Origins

At Sat, 1 Nov 2025 02:18:30 -0400, c186282 <c186282@nnada.net> wrote:

> On 11/1/25 01:41, rbowman wrote:
> > On Fri, 31 Oct 2025 23:16:04 -0000 (UTC), Lawrence D’Oliveiro wrote:
> > 
> >> Both were essentially research playgrounds for bright people to try out
> >> risky new ideas, run at the expense of a vast parent corporation that
> >> made huge profits from a Government-granted monopoly (the US phone
> >> monopoly in the case of AT&T, patents on photocopiers in the case of
> >> Xerox).
> > 
> > IBM had their little skunk works too. In the early days of Java they
> > released a compiler that was a lot better than the official and had some
> > other neat tools. Some middle manager may have said 'Take a look at that
> > Java stuff' and they grabbed it and ran.
> 
>    A lot of good stuff did - and still does - come out
>    of Big Blue. You don't HEAR about it much anymore
>    however, but they're still into *everything*. They
>    are not a leading chips name, but their research
>    goes into all the new chips. They are not the center
>    of the net, but a lot of their stuff goes into the net.
>    They are not a leading "AI" company, but their stuff
>    is deeply entwined in "AI".
> 
>    And they still sell Big Iron for Big Projects/corps.
> 
>    IMHO, they have positioned themselves perfectly for
>    the early 21st century. Being just off the radar scope
>    by and large they will make their money yet not be
>    subject to the storms of the more popular tech stocks.
>    WHEN "AI" mostly crashes, IBM will still be there.

"Where would I be without IBM?"

https://www.youtube.com/watch?v=D04TRaXFuVU

(Information Society)

-- 
-v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090Ti 24G
   OS: Linux 6.17.6 D: Mint 22.2 DE: Xfce 4.18 
   NVIDIA: 580.95.05 Mem: 258G
   "Oxymoron: Subsequent Initiatives."

[toc] | [prev] | [next] | [standalone]

#76936 — Re: Ken Thompson Recalls Unix’s Rowdy, Lock-P icking Origins

On Sat, 01 Nov 2025 07:06:16 +0000, vallor wrote:

> "Where would I be without IBM?"

Comfortably at home...  We used to say IBM stood for 'I've Been Moved'. 
One of my cousins cycled through New Jersey, New York, Vermont, and 
Arizona. 

The Vermont one was the best. IBM was short of space and had leased an old 
grocery store. It was low key with no big signs on the front so people 
would wander in looking for bananas.

I know the feeling. Where I work had the same approach, just a street 
number on the building. There was a state office plaza down the street so 
people would wander in looking for the DMV, VA, and other random stuff. We 
started locking the doors as telecommuting meant not many people 
physically on site and the homeless perceived that as an invitation.

 

[toc] | [prev] | [next] | [standalone]

#76934

On Sat, 1 Nov 2025 02:18:30 -0400, c186282 wrote:

>  They
>    are not a leading chips name, but their research goes into all the
>    new chips.

I haven't followed the latest episodes in the drama, but IBM sold off 
their fabs. 

https://en.wikipedia.org/wiki/IBM_Microelectronics

A friend interned with IBM in college and went on to work for them all his 
life. Fortunately he was of retirement age when they pulled the plug on 
Essex Junction. Many people weren't that lucky. According to Wiki IBM and 
Global Foundries have settled their suits and countersuits. 

[toc] | [prev] | [next] | [standalone]

#76911

On 2025-11-01 00:16, Lawrence D’Oliveiro wrote:
> On Fri, 31 Oct 2025 14:33:46 +0100, Carlos E.R. wrote:
> 
>> I just read the article, and I have to say that it happened because of
>> the culture in Bell Labs, and that money was not a problem. And they had
>> that money because the Bell company had that monopoly on phone service
>> and charged what they charged.
> 
> It is interesting to compare Bell Labs at AT&T, with PARC (the Palo Alto
> Research Center) at Xerox. The former was the birthplace of information
> theory, and Unix, among other things; the latter was the birthplace of the
> GUI, and Smalltalk, and Ethernet, among other things.
> 
> Both were essentially research playgrounds for bright people to try out
> risky new ideas, run at the expense of a vast parent corporation that made
> huge profits from a Government-granted monopoly (the US phone monopoly in
> the case of AT&T, patents on photocopiers in the case of Xerox).
> 
> And once the parent company lost that monopoly (the AT&T breakup, the
> expiry of those photocopier patents), the glory days at the research
> playground were over.

Exactly.

Bell Labs survived for a while after the AT&T breakup, but the coup de 
grâce came with the crash of Lucent in 2000.

-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]

#76795

<https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-origins/>

MULTICS didn’t die. Bell Labs pulled out, but the project continued,
and did finally reach production, obviously too late to satisfy some
parties. GE sold its hardware business (along with their rights to
MULTICS) to Honeywell who didn’t know what to do with that weird OS.
They charged an arm and a leg for it, but that didn’t seem to deter a
small cohort of loyal customers.

There are some documents at Bitsavers -- look at the brochure here
<https://bitsavers.trailing-edge.com/pdf/honeywell/large_systems/multics/>,
for example, and see how advanced the capabilities were that it
offered, compared to any other early-1980s large-system OS.

By the way, Thompson doesn’t use Unix any more. He gave up on Apple a
couple of years ago, and uses Linux now
<https://forums.freebsd.org/threads/unix-pioneer-ken-thompson-announces-hes-switching-from-mac-to-linux.88451/>.

[toc] | [prev] | [next] | [standalone]

#76819

On 10/28/25 18:59, Lawrence D’Oliveiro wrote:
> <https://thenewstack.io/ken-thompson-recalls-unixs-rowdy-lock-picking-origins/>
> 
> MULTICS didn’t die. Bell Labs pulled out, but the project continued,
> and did finally reach production, obviously too late to satisfy some
> parties. GE sold its hardware business (along with their rights to
> MULTICS) to Honeywell who didn’t know what to do with that weird OS.
> They charged an arm and a leg for it, but that didn’t seem to deter a
> small cohort of loyal customers.
> 
> There are some documents at Bitsavers -- look at the brochure here
> <https://bitsavers.trailing-edge.com/pdf/honeywell/large_systems/multics/>,
> for example, and see how advanced the capabilities were that it
> offered, compared to any other early-1980s large-system OS.
> 
> By the way, Thompson doesn’t use Unix any more. He gave up on Apple a
> couple of years ago, and uses Linux now
> <https://forums.freebsd.org/threads/unix-pioneer-ken-thompson-announces-hes-switching-from-mac-to-linux.88451/>.

   I remember just a little about MULTICS ... basically
   a 'high-concept' project that became a total cluster-fuck
   when implementation was attempted.

   UNIX was much better, more 'organic'. Get enough of
   the right people together and great things happen.

   Nothing really wrong with modern UNIX implementations,
   but Linux IS easier all-around.

[toc] | [prev] | [next] | [standalone]

#76820

On Thu, 30 Oct 2025 01:29:14 -0400, c186282 wrote:

>   I remember just a little about MULTICS ... basically
>    a 'high-concept' project that became a total cluster-fuck when
>    implementation was attempted.

I was never exposed to MULTICS. I first ran into *nix on a PDP-11. I 
forget what it was called but it was in the era when Bell couldn't really 
sell Unix because of the monopoly considerations. Being Boston Unix like 
software had a way of falling off the back of a truck. 

>    Nothing really wrong with modern UNIX implementations, but Linux IS
>    easier all-around.

AIX wasn't bad. Its problem was it ran on RS6000 systems. They weren't bad 
either as long as you had deep pockets. Our source tree compiled on either 
with a few flags for AIX or Linux. The biggest problem was the db_Vista 
geodata had to have its ends swapped. 

[toc] | [prev] | [next] | [standalone]

#76822

On 30/10/2025 05:29, c186282 wrote:
> I remember just a little about MULTICS ... basically
>    a 'high-concept' project that became a total cluster-fuck
>    when implementation was attempted.
> 
>    UNIX was much better, more 'organic'. Get enough of
>    the right people together and great things happen.
> 
Computer scientists add in all the features anyone could need whilst 
engineers build in only the ones that *are* needed...

>    Nothing really wrong with modern UNIX implementations,
>    but Linux IS easier all-around.

Just more people know it and document it...

-- 
     “I know that most men, including those at ease with problems of the 
greatest complexity, can seldom accept even the simplest and most 
obvious truth if it be such as would oblige them to admit the falsity of 
conclusions which they have delighted in explaining to colleagues, which 
they have proudly taught to others, and which they have woven, thread by 
thread, into the fabric of their lives.”

     ― Leo Tolstoy

[toc] | [prev] | [next] | [standalone]

#76801

By the way, “Unix” is supposed to be a trademark. That means you use it as 
an adjective, not a noun.

[toc] | [prev] | [next] | [standalone]

#76807

On Wed, 29 Oct 2025 02:54:59 -0000 (UTC), Lawrence D’Oliveiro wrote:

> By the way, “Unix” is supposed to be a trademark. That means you use it
> as an adjective, not a noun.

Trying to be more pedantic than thou, UNIX is the trademark, not Unix.

https://unix.org/trademark.html

Unlike most of the Microsoft world, case matters.

[toc] | [prev] | [next] | [standalone]

Page 2 of 3 — ← Prev page 1 [2] 3  Next page →

Back to top | Article view | comp.os.linux.misc

csiph-web