rdavenport@gtewis.arpa (07/05/88)
I've recently acquired a copy of the PWB/UNIX User's Manual in two volumes,
(one section looks like the current User's Manual and the other like the
Programmer's Manual) printed in May 1977. Can anyone tell me where PWB/UNIX
fits in to the UNIX history? I've never heard of it before.
thanks,
rob
_-----------------------------------------------------------------_
| Robertson G. Davenport |
| GTE Govt. Systems Corp. |
| Billerica, MA C the world.... |
| BELL : (617) 671-5180 |
| ARPA : rdavenport@gtewis.arpa |
| CIS : 73407,3716 |
| UUCP : ...!{harvard,rutgers,uunet,ulowell}!gtewis.arpa!rdavenport |
\___________________________________________________________________/
"If you want to eat hippopautamus, you've got to pay the freight.";
-- some IBM guy
-----ron@topaz.rutgers.edu (Ron Natalie) (07/06/88)
Off the top of my head, which is probably wrong, PWB UNIX (Programmers Workbench) saw the light of day (out side of ATT) around 1978 and was developed by the group that eventually evolved into the people who were responsible for PWBII, System 3, and System V. It's main raison d'etre was the enhanced ability to keep track of programs to be down loaded to other computer environments (like IBM RJE). SCCS has it's roots here. Originally it was a slightly bug fixed V6 kernel with a couple of enhancements (like the empty() system call) and featured a few extra niceties like the PWB shell, which was distinguished from the existing shell by having things like shell variables and a user setable path. -Ron
alanm@dvlmarv.UUCP (Alan Matsuoka) (07/07/88)
In article <16382@brl-adm.ARPA> rdavenport@gtewis.arpa writes: > > Can anyone tell me where PWB/UNIX > fits in to the UNIX history? I've never heard of it before. > [ somebody please correct me if I am wrong anywhere... This goes back inmy memory a ways]. I worked on a PWB system about 9 years ago. As I can recall, it was a sort of Version 6.5 ( At least the C compiler was ). Version 6.5 was released to a number of universities ( I think... I came from a Version 6 site [ or 6th Edition as it was called ]). PWB was a sort of beefed up version that came with SCCS. It had the 6th Edition I node format ( with some modifications [I forget] ). We didn't have such nice things like fsck to fix things up when things crashed, just fsdb , a 'file system debugger' that was considered a great help considering that all we had with V6 was rm, clri, dcheck, icheck and ncheck to fix up the file system. A crash took a while to recover from. There were the old restrictions on file sizes like in V6. There were restrictions in the C language at the time, things like no enums, voids, arrays of unions. unions, typedefs, and unsigned math were new back then. (How many of you can remember that unsigned arithmetic was done using char * in V6?). Last but not least, I believe that it ran only on PDP11/70's with Floating Point. There was some bizarre reason why you needed the floating point unit but you could live without it if you had a source license (everybody did back then). Oh, yes.. Then there was the Mashey shell. This was a shell that appeared only with PWB. Now most people, (mainly institutions) waited until Version 7 came out since it was a lot better. After V7, Western Electric changed the naming scheme and came out with System III. A sort of PWD, V7 and other stuff put together.... -------------------------------------------------------------------------------- Alan Matsuoka, Develcon Electronics, 515 Consumers Road, Suite 500 Willowdale, ON Canada M2J 4Z2 uucp: {decvax,ihnp4,utai,uunet}!watmath!dvlmarv!alanm phone: (416) 495-8666 {utai,mnetor}!lsuc!dvlmarv!alanm sask!zaphod!dvlmarv!alanm
ron@topaz.rutgers.edu (Ron Natalie) (07/08/88)
It is not true that PWB UNIX required an 11/70 and floating point. It didn't even require split-I/D so you could run it on 11/34's and such provided they had memory management. What I believe you are referring to as 6.5 is the phototypesetter C compiler. When you bought troff, you got a fixed up, pre-version 7 compiler that had things like the equals-op operators the right way around. There was also a kernel diff listing circulating around referred to as "fifty fixes." There was actually a PWB II release which had a V7 kernel as a base, if I recall. -Ron