[TUHS] Thompson at Berkeley on an 11/70

[TUHS] Thompson at Berkeley on an 11/70

[Doug McIlroy]

To tell whether Ken installed v6 or a copy of his home
system, look at /usr/dict/words. On the home system
that file is the wordlist from Webster's Collegiate
Dictionary, 7th edition, licensed for Bell Labs use
only. On distribution systems we substituted the wordlist
for "spell". The latter list contains many more proper
names, acronyms, etc than the dictionary did, in
particular names that appear in Unix documentation
such as Ritchie, Kernighan, and McIlroy. It also lacks
lots of trivially derivable words like "organizationally".

If you do have the Webster file rather than the spell
file, please don't propagate it.

Doug

[TUHS] Thompson at Berkeley on an 11/70

[Greg 'groggy' Lehey]

On Friday, 10 April 2015 at 14:37:25 -0400, Doug McIlroy wrote:
> To tell whether Ken installed v6 or a copy of his home
> system, look at /usr/dict/words. On the home system
> that file is the wordlist from Webster's Collegiate
> Dictionary, 7th edition, licensed for Bell Labs use
> only. On distribution systems we substituted the wordlist
> for "spell". The latter list contains many more proper
> names, acronyms, etc than the dictionary did, in
> particular names that appear in Unix documentation
> such as Ritchie, Kernighan, and McIlroy. It also lacks
> lots of trivially derivable words like "organizationally".

In FreeBSD there are two files /usr/share/dict/web2 and
/usr/share/dict/web2a, suggesting that they're Webster.  web2 sounds
like the words file, while web2a apparently consists of compounds.
web2 doesn't contain Ritchie or McIlroy, though it does contain
"organizationally".  The oldest entry in the svn history is:

r1638 | rgrimes | 1994-05-31 05:09:18 +1000 (Tue, 31 May 1994) | 2 lines

BSD 4.4 Lite Share Sources

I don't have sccs, so I can't check the origins of that file.

Comments?

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 MUA reports
problems, please read http://tinyurl.com/broken-mua
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20150411/b59178c3/attachment.sig>

[TUHS] Thompson at Berkeley on an 11/70

[Doug McIlroy]

"Greg 'groggy' Lehey" <grog at lemis.com> wrote:
In FreeBSD there are two files /usr/share/dict/web2 and
/usr/share/dict/web2a, suggesting that they're Webster.  web2 sounds
like the words file, while web2a apparently consists of compounds.
web2 doesn't contain Ritchie or McIlroy, though it does contain
"organizationally".  The oldest entry in the svn history is:

r1638 | rgrimes | 1994-05-31 05:09:18 +1000 (Tue, 31 May 1994) | 2 lines

BSD 4.4 Lite Share Sources


Those files are Webster's Unabridged. They were derived from
a tape produced by the army and distributed without restrictions.
The provenance is more fully described at puzzlers.org; burrow
down their file tree from "Solving Tools of the NPL" to "The NPL
Wordlists" and "Our Wordlists" to "Doug McIlroy's Wordlists"

Doug

[TUHS] Thompson at Berkeley on an 11/70

[Warner Losh]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1981 bytes --]


> On Apr 11, 2015, at 12:39 AM, Doug McIlroy <doug at cs.dartmouth.edu> wrote:
> 
> "Greg 'groggy' Lehey" <grog at lemis.com> wrote:
> In FreeBSD there are two files /usr/share/dict/web2 and
> /usr/share/dict/web2a, suggesting that they're Webster.  web2 sounds
> like the words file, while web2a apparently consists of compounds.
> web2 doesn't contain Ritchie or McIlroy, though it does contain
> "organizationally".  The oldest entry in the svn history is:
> 
> r1638 | rgrimes | 1994-05-31 05:09:18 +1000 (Tue, 31 May 1994) | 2 lines
> 
> BSD 4.4 Lite Share Sources
> 
> 
> Those files are Webster's Unabridged. They were derived from
> a tape produced by the army and distributed without restrictions.
> The provenance is more fully described at puzzlers.org; burrow
> down their file tree from "Solving Tools of the NPL" to "The NPL
> Wordlists" and "Our Wordlists" to "Doug McIlroy's Wordlists"

More importantly, it was from Websters 1913 edition, which had passed from
copyright protection at the time it was produced. There’s several words in it
that have shifted in accepted spelling since that edition which I noticed when
a spelling program flagged the now-correct spelling years ago…

There were other word lists floating around the Internet in the 80’s that one
could obtain with some effort that differed. I may have been on a system once
that had both lists. There’s about 30 words that are different, and about 200
words in that other file not in web2. The system I know that had this word list
on it has been turned off for 20 years now, so I can’t easily check these
numbers. I assume other file was the file talked about as being licensed to
Bell Labs…

Warner

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20150411/c113fcb2/attachment.sig>

[TUHS] Thompson at Berkeley on an 11/70

[Jeremy C. Reed]

On Sat, 11 Apr 2015, Greg 'groggy' Lehey wrote:

> In FreeBSD there are two files /usr/share/dict/web2 and
> /usr/share/dict/web2a, suggesting that they're Webster.  web2 sounds
> like the words file, while web2a apparently consists of compounds.
> web2 doesn't contain Ritchie or McIlroy, though it does contain
> "organizationally".  The oldest entry in the svn history is:
> 
> r1638 | rgrimes | 1994-05-31 05:09:18 +1000 (Tue, 31 May 1994) | 2 lines
> 
> BSD 4.4 Lite Share Sources
> 
> I don't have sccs, so I can't check the origins of that file.
> 
> Comments?

There is no existing SCCS history for those two files.
But you can see the background about it in the README from the SCCS.
http://svnweb.freebsd.org/csrg/share/dict/README?revision=49371&view=markup
(Also in the later FreeBSD README too.)

The web2 files were first introduced in 4.3BSD as documented in the 
changes documentation (between the 4.2BSD distribution of July 1983 and 
the 4.3 revision in 1986) 4.3BSD's usr/doc/smm/12.uchanges/7.t (but I 
don't see this in the SCCS history).

We only have the complete operating system as of 3BSD and later and they 
include "words" file from 32V. It doesn't have "Kernighan", "McIlroy", 
nor "organizationally". The version continued from 32V is 
http://svnweb.freebsd.org/csrg/share/dict/words?view=log

[TUHS] Thompson at Berkeley on an 11/70

[Jacob Goense]

On 2015-04-10 20:37, Doug McIlroy wrote:
> To tell whether Ken installed v6 or a copy of his home
> system, look at /usr/dict/words. On the home system
> that file is the wordlist from Webster's Collegiate
> Dictionary, 7th edition, licensed for Bell Labs use
> only. On distribution systems we substituted the wordlist
> for "spell". The latter list contains many more proper
> names, acronyms, etc than the dictionary did, in
> particular names that appear in Unix documentation
> such as Ritchie, Kernighan, and McIlroy. It also lacks
> lots of trivially derivable words like "organizationally".
> 
> If you do have the Webster file rather than the spell
> file, please don't propagate it.

Good to know, I doubt I'll ever run into a dump of a 1975 CORY though.

For now I assume Thompson erred on the safe side and brought his home
system as reference material rather than install media.

/Jacob

[TUHS] Thompson at Berkeley on an 11/70

[Jacob Goense]

I'm experimenting with adapting Unix history and lore using the new
EXPECT/SEND feature in simh. My favorite guinea pig is the story of Ken
Thompsons sabbatical at Berkeley where he brings up V6 on new 11/70 with
Bob Kridle and Jeff Schriebman. Any details not yet recorded in obvious
places[1] are of course more than welcome!

One of the things I'm trying to get right is what they actually brought
up there initially in 1975. This must have been standard V6 or the
Bell UNIX Ken brought with him, but I can't figure it out.

Salus has Schriebman, Haley and Joy installing the fixes on the 50 bugs
tape late summer 1976. This suggests it was stock V6 initially, but they
might have been playing on a different system or working from a fresh
install in 1976.

If it was stock V6 initially, what were they waiting for? Legal stuff?

If it was 1975 Bell UNIX, can I reconstruct this using the 54 patches
collected by Mike O'Brien[2], or is that going to be way off from what
Thompson left in Urbana-Champaign with Greg Chesson in 1975?

[1] http://www.tuhs.org/books.html minus the Bell journals for example
[2] Hidden in /usr/sys/v6unix/unix_changes in one of the Spencer tapes   
   http://www.tuhs.org/Archive/Applications/Spencer_Tapes/unsw3.tar.gz

[TUHS] Thompson at Berkeley on an 11/70

[Jeremy C. Reed]

On Fri, 10 Apr 2015, Jacob Goense wrote:

> If it was stock V6 initially, what were they waiting for? Legal stuff?
>
> If it was 1975 Bell UNIX, can I reconstruct this using the 54 patches
> collected by Mike O'Brien[2], or is that going to be way off from what
> Thompson left in Urbana-Champaign with Greg Chesson in 1975?

When Thompson arrived, the Math and Statistics department had a shared 
machine running RSTS-E and Unix part time. McKusick wrote that the 11/70 
arrived in Fall of 1975 and it was installed with Version 6. (Marshall 
Kirk McKusick. A BERKELEY ODYSSEY: Ten years of BSD history. Unix 
Review. January 1985. Volume 3. Pages 31-42.) But the Quarter Century 
book (page 155) says it was upgraded to V6 after Thompson left (he told 
me he left right after exams in June). Thompson told me he brought up 
the 11/70 for the EECS (but didn't tell me the version). (Jolitz and 
Allman told me the shared 11/45 also used for CS classes ran v6 (and 
RSTS-E) but I don't when it was upgraded to v6. The INGRES group also  
their own dedicated computer which ran 5th Edition and then v6.)

(Thompson told me that he wrote his Pascal between quarter 1 and 
quarter 2 and that Fabry used it for his instruction the next quarter. 
Fabry told me that Thompson's students completed their programming 
assignments using the Cory Hall PDP-11/70 running Unix.)

O'Brien suggested to me that the annotated fifty changes wouldn't apply 
to Berkeley's already hacked kernel.

Allman and Fabry told me that in their code reading sessions, Thompson 
had the John Lions' annotated code for the v6 kernel. I believe these 
(Lions') lecture notes were from May 1976 (so not much time before 
Thompson left).

> [2] Hidden in /usr/sys/v6unix/unix_changes in one of the Spencer tapes
> http://www.tuhs.org/Archive/Applications/Spencer_Tapes/unsw3.tar.gz

Thank you so much for pointing me to this. Too bad there aren't dates 
mentioned in the code or diff files. I am authoring a book that includes 
this same story (and I had basically the same questions).


  Jeremy C. Reed

echo 'EhZ[h ^jjf0%%h[[Zc[Z_W$d[j%Xeeai%ZW[ced#]dk#f[d]k_d%' | \
  tr            '#-~'            '\-.-{'