[TUHS] Re: Ken Thomson's email address?

[TUHS] Re: Ken Thomson's email address?

[Noel Chiappa]

    > From: Jackson Helie G

    > I would like to ask Ken Thompson if he has the first C compiler.

He wouldn't; he didn't do much work on C.

Dennis Ritchie (who created C) posted two very early ones here:

  https://www.nokia.com/bell-labs/about/dennis-m-ritchie/primevalC.html

	Noel

[TUHS] Re: Ken Thomson's email address?

[Pete]

On 2025年05月08日 03:14, Noel Chiappa wrote:
> I'm not up for looking through TUHS archives for it.

What I did when I subscribed (thanks, wkt) was download the archives
(which are in mbox format), cat them, and then copy the messages to the
directory where the TUHS messages are filtered. Took almost no time, but
my mail server is in my house.

This makes it pretty easy to search the archives locally: you can just
use grep or if you have a mail reader with good search features, it's
much easier than searching on the server. It looks like there was "A
Census of /etc and /sys Prior to V4" on 2023-05-15, and "Census of Early
(Pre-C Kernel) UNIX Documents" on 2024-02-14, both from segaloco. (The
only downside to importing the mboxes is that the email addresses are
tweaked to prevent spam.) That should make it easier to find the threads.

Pete

[TUHS] Re: Ken Thomson's email address?

[Marc Rochkind]

[-- Attachment #1: Type: text/plain, Size: 1953 bytes --]

Seems that the subject line of this thread should be changed.

Marc

On Thu, May 8, 2025 at 1:00 PM segaloco via TUHS <tuhs@tuhs.org> wrote:

> On Thursday, May 8th, 2025 at 7:23 AM, Clem Cole <clemc@ccc.com> wrote:
>
> > Thanks, Noel. I should have pointed to that paper also. I have always
> considered the demarcation point when structures were added, but I darned
> if I knew at what date or time Dennis chose to rename it. Clearly, one day
> (night, more likely), he realized it nb was really not B anymore and needed
> a new name.
> > ᐧ
> >
> > On Thu, May 8, 2025 at 10:04 AM Noel Chiappa <jnc@mercury.lcs.mit.edu>
> wrote:
> >
> > > > From: Clem Cole
> > >
> > > > The first "C" compiler wa an ephemeral state some time in the process
> > > > of its evolution. Dennis started with his B implementation and began
> to
> > > > add features he needed
> > >
> > > See:
> > >
> > > The Development of the C Language
> > > https://www.nokia.com/bell-labs/about/dennis-m-ritchie/chist.html
> > >
> > > for detail on the evolution.
> > >
> > > Noel
>
> For the record another source of old C compiler code would be in
> the s1/s2 tapes here:
> https://www.tuhs.org/Archive/Distributions/Research/1972_stuff/
>
> I'm not in a place to dig into them right now but one of the two
> tapes is a V2 root filesystem backup, the C compiler should be
> in a couple stages in /lib.  Granted, its binary, but with
> enough disassembly work and comparison with other sources, it
> may be possible to recreate a compelling facsimile of whatever
> revision that represents.
>
> Otherwise as others have stated, the historic C compiler
> situation has been quite well plundered by this and other
> groups, supplemented of course by Dennis Ritchie's foresight to
> keep and provide so many artifacts and anecdotes.
>
> - Matt G.
>


-- 
Subscribe to my Photo-of-the-Week emails at my website mrochkind.com.

[-- Attachment #2: Type: text/html, Size: 3009 bytes --]

[TUHS] Re: Ken Thomson's email address?

[segaloco via TUHS]

On Thursday, May 8th, 2025 at 7:23 AM, Clem Cole <clemc@ccc.com> wrote:

> Thanks, Noel. I should have pointed to that paper also. I have always considered the demarcation point when structures were added, but I darned if I knew at what date or time Dennis chose to rename it. Clearly, one day (night, more likely), he realized it nb was really not B anymore and needed a new name.
> ᐧ
> 
> On Thu, May 8, 2025 at 10:04 AM Noel Chiappa <jnc@mercury.lcs.mit.edu> wrote:
> 
> > > From: Clem Cole
> > 
> > > The first "C" compiler wa an ephemeral state some time in the process
> > > of its evolution. Dennis started with his B implementation and began to
> > > add features he needed
> > 
> > See:
> > 
> > The Development of the C Language
> > https://www.nokia.com/bell-labs/about/dennis-m-ritchie/chist.html
> > 
> > for detail on the evolution.
> > 
> > Noel

For the record another source of old C compiler code would be in
the s1/s2 tapes here:  https://www.tuhs.org/Archive/Distributions/Research/1972_stuff/

I'm not in a place to dig into them right now but one of the two
tapes is a V2 root filesystem backup, the C compiler should be
in a couple stages in /lib.  Granted, its binary, but with
enough disassembly work and comparison with other sources, it
may be possible to recreate a compelling facsimile of whatever
revision that represents.

Otherwise as others have stated, the historic C compiler
situation has been quite well plundered by this and other
groups, supplemented of course by Dennis Ritchie's foresight to
keep and provide so many artifacts and anecdotes.

- Matt G.

[TUHS] Re: Ken Thomson's email address?

[Clem Cole]

[-- Attachment #1: Type: text/plain, Size: 807 bytes --]

Thanks, Noel. I should have pointed to that paper also. I have always
considered the demarcation point when structures were added, but I darned
if I knew at what date or time Dennis chose to rename it. Clearly, one day
(night, more likely), he realized it nb was really not B anymore and needed
a new name.
ᐧ

On Thu, May 8, 2025 at 10:04 AM Noel Chiappa <jnc@mercury.lcs.mit.edu>
wrote:

>     > From: Clem Cole
>
>     > The first "C" compiler wa an ephemeral state some time in the process
>     > of its evolution. Dennis started with his B implementation and began
> to
>     > add features he needed
>
> See:
>
>   The Development of the C Language
>   https://www.nokia.com/bell-labs/about/dennis-m-ritchie/chist.html
>
> for detail on the evolution.
>
>         Noel
>

[-- Attachment #2: Type: text/html, Size: 1653 bytes --]

[TUHS] Re: Ken Thomson's email address?

[Noel Chiappa]

    > From: Clem Cole

    > The first "C" compiler wa an ephemeral state some time in the process
    > of its evolution. Dennis started with his B implementation and began to
    > add features he needed

See:

  The Development of the C Language
  https://www.nokia.com/bell-labs/about/dennis-m-ritchie/chist.html

for detail on the evolution.

	Noel

[TUHS] Re: Ken Thomson's email address?

[Clem Cole]

[-- Attachment #1: Type: text/plain, Size: 911 bytes --]

On Thu, May 8, 2025 at 6:59 AM Jackson Helie G <yuyi15968@gmail.com> wrote:

> If this is true, it's a pity. Otherwise, I would like to ask Ken Thompson
> if he has the first C compiler.
>
The first "C" compiler wa an ephemeral state some time in the process of
its evolution.  Dennis started with his B implementation and began to add
features he needed to support their new byte addressed PDP-11.  He
originalled called it nb or "new B."   The compiler accepted the B syntax
and new extensions for the new features.  Eventually, the new feature
started to make the language different enough the language became something
he called C.   At what time what was, I'm not sure I've ever heard anyone
ever declare and definitive time or date.

Look in https://www.tuhs.org/Archive/Applications/Dennis_Tapes/ And start
poking if you want to see early versions of the compiler.
Have fun
Clem
ᐧ

[-- Attachment #2: Type: text/html, Size: 2646 bytes --]

[TUHS] Re: Ken Thomson's email address?

[Jackson Helie G]

[-- Attachment #1: Type: text/plain, Size: 106 bytes --]

If this is true, it's a pity. Otherwise, I would like to ask Ken Thompson
if he has the first C compiler.

[-- Attachment #2: Type: text/html, Size: 132 bytes --]

[TUHS] Re: Ken Thomson's email address?

[Noel Chiappa]

    > From: Jackson Helie G

    > I was wondering if anyone knows Ken Thomson's email address? I was
    > wondering if he has any more archives of Unix from 1972 and before.

He does read this ist, and very occasionally posts to it.

But there's no point bothering him, to ask; anything he had, he turned over
many years ago.

(To the point that people have recently been poring through the 'trash' bits
in the "s1-bits" and s2-bit" tapes, IIRC, for lack of anything else to look
at. Look for "A Census of /etc and /sys Prior to V4" in the TUHS archive to
see some discussion of some of this work, by Matt Gilmore. I think somene else
was working on it too, but I couldn't find it; I'm not up for looking through
TUHS archives for it.)

	Noel