logan@vsedev.VSE.COM (James Logan III) (11/29/88)
In article <9003@smoke.BRL.MIL> gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>) writes: >That's pretty silly, since they have no indication of which releases of >what news software each of us is using, nor what differences in the GCOS >field usage Does anyone know what GCOS really stands for and where it came from? Just send me email so we don't waste everyone else's time... If there is interest, send me email and I'll forward the responses to you. If there's a LOT of interest, I'll post to the net. -Jim -- Jim Logan logan@vsedev.vse.com (703) 892-0002 uucp: ..!uunet!vsedev!logan inet: logan%vsedev.vse.com@uunet.uu.net
jpr@dasys1.UUCP (Jean-Pierre Radley) (11/30/88)
In article <1257@vsedev.VSE.COM> logan@vsedev.VSE.COM (James Logan III) writes: >Does anyone know what GCOS really stands for and where it came from? In my flavor of unix, that's described as the "Comment" field. In volume 1 of the Unix Programmer's Manual, the old green books, that field was listed as GCOS job number, box number, optional GCOS user-id and the manual page went on to state: The GCOS field is used only when communicating with that system, and in other installations can contain any desired information. Elsewhere in the old green books, I get the implicit, not explicit, information that GCOS referred to a typesetting system at Murray Hill at the time that Osanna & Co. were inventing troff. -- Jean-Pierre Radley Honi soit jpr@dasys1.UUCP New York, New York qui mal ...!hombre!jpradley!jpr CIS: 76120,1341 y pense ...!hombre!trigere!jpr
pim@mrevox.UUCP (11/30/88)
Someone just the other day saw the GCOS(GECOS) field being prompted for and remarked that he hadn't seen that name for a while. He said that it stood for: General Electric Computer Operating System or General Electric Comprehensive Operating System It seems far fetched, but interesting. Paul MacMurdo pim@mrevox.bellcore.com ...!bcr!mrevox!pim
gwyn@smoke.BRL.MIL (Doug Gwyn ) (12/01/88)
In article <7996@dasys1.UUCP> jpr@dasys1.UUCP (Jean-Pierre Radley) writes: >In article <1257@vsedev.VSE.COM> logan@vsedev.VSE.COM (James Logan III) writes: >>Does anyone know what GCOS really stands for and where it came from? >In my flavor of unix, that's described as the "Comment" field. That's what it has evolved to, since most current UNIX systems don't spool printer output etc. to Honeywell (formerly GE) systems.
guy@auspex.UUCP (Guy Harris) (12/01/88)
>In my flavor of unix, that's described as the "Comment" field. Well, sort of. Actually, in the 4.BSD "struct passwd", there are both "pw_comment" *and* "pw_gecos" fields; the former is set to point to a null string, and the latter is set to point to the GCOS/comment/full name/whatever field. The S5R3.1 one also has both those fields in "struct passwd"; it sets them both to point to the GCOS/comment/full name/whatever field. >Elsewhere in the old green books, I get the implicit, not explicit, >information that GCOS referred to a typesetting system at Murray Hill at >the time that Osanna & Co. were inventing troff. Actually, GCOS was the operating system on the machine on which the typesetting stuff ran. I think it was originally GECOS, for GE Csomething Operating System, for the GE 6*5 series ("*" replaceable by 3, and perhaps by other digits, except 4 - if you replace it with 4, it runs Multics instead). When Honeywell bought out GE's computer operation, they probably dropped the "E".
prc@ERBE.SE (Robert Claeson) (12/01/88)
In article <1257@vsedev.VSE.COM>, logan@vsedev.VSE.COM (James Logan III) writes: > Does anyone know what GCOS really stands for and where it came from? Or what the BSD and AT&T GCOS field formats are? -- Robert Claeson ERBE DATA AB rclaeson@ERBE.SE
mrj@stratus.UUCP (Michael R. Jordan) (12/02/88)
My recollection is GEneral Comprehensive Operating System. When Honeywell
bought out GE's LSD (Large Systems Division) they dropped the Es in all
of the products that started out with a GE.
--
Stratus Computer, Inc. | Internet: mrj@stratus.com
Western Development Center | UUCP: {amdahl,oliveb,uunet}!stratus!mrj
12980 Saratoga Ave. |
Saratoga, CA 95070 | Phone: (408) 725-4220charette@edsews.EDS.COM (Mark A. Charette) (12/02/88)
In article <4400001@mrevox.UUCP>, pim@mrevox.UUCP writes: > Someone just the other day saw the GCOS(GECOS) field > being prompted for and remarked that he hadn't seen > that name for a while. He said that it stood for: > > General Electric Computer Operating System > > or > > General Electric Comprehensive Operating System > > It seems far fetched, but interesting. > > Paul MacMurdo > > pim@mrevox.bellcore.com > ...!bcr!mrevox!pim Not so farfetched. GCOS did stand for General Electric Comprehensive Operating System, but changed to General Comprehensive Operating System after Honeywell bought the GE computer division. Multics ran on the Honeywells, and we all know (or can guess) at the evolution after that. BTW, I first started my programming career on a Honeywell 6060 Level 6 running GCOS 3. Since then, having worked on a few mainframes in between, seeing a file system done right was refreshing when I started working with Unix. One little question - anyone ever remember seeing the 'B' language on Honeywell mainframes? I worked with it without a manual for a while when I was playing on the Honeywell. I think it was from the U. of Waterloo (as were a fair percentage of the other utilities on the Funnywell). -- Mark Charette "People only like me when I'm dumb!", he said. Electronic Data Systems "I like you a lot." was the reply. 750 Tower Drive Voice: (313)265-7006 FAX: (313)265-5770 Troy, MI 48007-7019 charette@edsews.eds.com uunet!edsews!charette
dmr@alice.UUCP (12/02/88)
A couple of brief historical notes-- GCOS used to be GECOS and was GE's, and is now Honeywell's system. Xerox wasn't involved. The field in the password file originally contained a G[E]COS login name and account number and was used when jobs were sent to that system from Unix. Mostly these jobs printed things on a line printer, definitely not a typesetter. We were the ones who had that (and didn't have a fast printer). B was done on the PDP-7 in an interpretive implementation. On the -7, I wrote a cross-compiler for it to the Honeywell-- actually still GE at that time. This was a peculiar sort of tour-de-force, since it worked in 4K words of memory. The interpreter had a software paging mechanism, so the virtual space was larger. Later, in the early '70s, Steve Johnson spent a year at Waterloo and took the B compiler with him. It became popular there, and even had some offshoots-- Eh and Zed. Incidentally, Steve brought back some interesting perceptions about change. Previously, Waterloo had become well-known for WATFOR, a quick, student-oriented version of Fortran that ran on IBM systems. A little later, Morven Gentleman moved (from BTL) to Waterloo and brought in Honeywell equipment, mainly because of its superior (among then available commercial systems) interactive computing. This was a sort of revolution that rousted the batch WATFOR hegemony, and B flourished. Not too much later, Unix came in, and there was another revolution. Dennis Ritchie
soley@ontenv.UUCP (Norman S. Soley) (12/02/88)
In article <4400001@mrevox.UUCP>, pim@mrevox.UUCP writes: > Someone just the other day saw the GCOS(GECOS) field > being prompted for and remarked that he hadn't seen > that name for a while. He said that it stood for: > > General Electric Computer Operating System > or > General Electric Comprehensive Operating System > > It seems far fetched, but interesting. Considering that Thompson had been working on Multics (at the time a joint project between MIT, Bell Labs and G.E.) the roots of which was the GCOS O/S (which by the way is still used on Honeywell DPS type boxes) before UNIX was even a glimmer in his eye, it's not so far fetched at all. Now the question still remains, what is the connection between the GCOS O/S and the GCOS field in the UNIX passwd file (if any). -- Norman Soley - Data Communications Analyst - Ontario Ministry of the Environment UUCP: uunet!attcan!lsuc!ncrcan!ontenv!soley VOICE: +1 416 323 2623 OR: soley@ontenv.UUCP " Stay smart, go cool, be happy, it's the only way to get what you want"
dmt@ptsfa.PacBell.COM (Dave Turner) (12/03/88)
In article <5544@edsews.EDS.COM> charette@edsews.EDS.COM (Mark A. Charette) writes: > >Not so farfetched. GCOS did stand for General Electric Comprehensive >Operating System, but changed to General Comprehensive Operating System >after Honeywell bought the GE computer division. Multics ran on the Almost right. I have a Honeywell ad from about 10 years ago that says that GCOS means: General Comprehensive Operating Supervisor -- Dave Turner 415/542-1299 {att,bellcore,sun,ames,pyramid}!pacbell!dmt
lazear@gateway.mitre.org (12/03/88)
GCOS stands for GE Comprehensive Operating System, which changed to General Compre. OS when Honeywell bought out GE's old 600 computer line long ago. GCOS remains the Honeywell OS of (their) choice. The GCOS mainframes required some user ID stuff when remote jobs were submitted and the GCOS field in the passwd file was a repository for that info. The "Programmer's Workbench" version of Unix made extensive use of this for submitting jobs to the Murray Hill facility, I believe. Walt (Lazear@mitre.org)
davidsen@steinmetz.ge.com (William E. Davidsen Jr) (12/03/88)
In article <552@auspex.UUCP> guy@auspex.UUCP (Guy Harris) writes: | >In my flavor of unix, that's described as the "Comment" field. | | Well, sort of. Actually, in the 4.BSD "struct passwd", there are both | "pw_comment" *and* "pw_gecos" fields; the former is set to point to a | null string, and the latter is set to point to the GCOS/comment/full | name/whatever field. The S5R3.1 one also has both those fields in | "struct passwd"; it sets them both to point to the GCOS/comment/full | name/whatever field. I'm not sure what you mean by "sets them both to point..." but if you try putpwent on Ultrix, Xenix, or V/386 it seems to use the GCOS field and ignore the comment field. That is, if you set the comment structure member to point to "comment" and the GCOS member to point to "gcos" the output says "gcos" and the output does not include the comment. I assume that your posting meant the the result of getpwent has pointers to the gcos field in both members, but they are not interchangable when modifying an entry and writing it out. -- bill davidsen (wedu@ge-crd.arpa) {uunet | philabs}!steinmetz!crdos1!davidsen "Stupidity, like virtue, is its own reward" -me
gwyn@smoke.BRL.MIL (Doug Gwyn ) (12/03/88)
In article <411@maxim.ERBE.SE> prc@ERBE.SE (Robert Claeson) writes: >In article <1257@vsedev.VSE.COM>, logan@vsedev.VSE.COM (James Logan III) writes: >> Does anyone know what GCOS really stands for and where it came from? >Or what the BSD and AT&T GCOS field formats are? Anything not containing a newline or colon. Unfortunately, some mail software thinks the field needs to contain specific user ID information. What this is expected to look like would depend on your mail system.
guy@auspex.UUCP (Guy Harris) (12/04/88)
>> Does anyone know what GCOS really stands for and where it came from? > >Or what the BSD and AT&T GCOS field formats are? Yes. They are both strings of text (preferably printable characters and blanks) containing no colons. Few, if any, programs provided by AT&T as part of its releases assume a particular format for that field; some people appear to think that the format in the file distributed with S5 releaes - which, as far as I know, just happens to be a format used by administrators within AT&T for their own convenience - is some sort of standard. The only AT&T convention I know of is the "pri=nnn" stuff at the front, which "login" uses to set the "nice" value when you log in. I know of no software provided with S5 that obliges you to have a GCOS field in the form nnnn-Name(nnnn) (in fact, several of the accounts used, I guess, for "sysadm" in the S5R3 "passwd" file *don't* use that format). Nevertheless, I see a lot postings from non-AT&T sites where they've blindly adopted that format. The BSD format is described in ADDUSER(8); this, unfortunately, is actually used by some programs - said programs even make assumptions about phone numbers and office names! Basically, leave out commas, and you can probably put anything there. The BSD format includes a convention that not all programs necessarily honor, which means you may not want to use that convention: if the "&" character appears in the "gecos" field, those programs replace it with the login name with the first letter capitalized. This means you probably want to leave out "&" as well.
guy@auspex.UUCP (Guy Harris) (12/04/88)
>I'm not sure what you mean by "sets them both to point..." but if you >try putpwent on Ultrix, Xenix, or V/386 it seems to use the GCOS field >and ignore the comment field. It can't choose both, so it has to ignore one; the code that sets the "comment" field may be a recent innovation antedating "putpwent" (I lack access to sources of sufficiently old releases to check this), so for historical reasons "putpwent" would use the "gecos" field. It is probably not a *good* innovation; it might be nice if the field had been named "comment", but it's probably not worth fixing now (it may be too late to fix it anyway). If somebody wants to fix names, perhaps they should think about changing "creat" to "create" :-) :-) :-) :-). (I seem to remember a comment along the lines of either "if I did UNIX again, the only thing I'd change is that I'd call 'creat' 'create" or "if I did UNIX again, I'd have called 'creat' 'create'" being ascribed to Ken Thomson; did he actually say something like either of those, and what did he actually say?) Basically, programmers should pretend the "pw_comment" field is a filler field, not containing anything and not used. By "sets them both to point" I meant "sets them both to point"; "getpwent" implementations generally read a line from the password file into a buffer, stuff '\0's into the buffer to cut it into fields, and set the "char *" members of the "struct passwd" structure to point into that buffer; the S5R3 implementation sets both "pw_comment" and "pw_gecos" to point to the same place in the buffer, namely the beginning of the GCOS field.
zap@front.se (Svante Lindahl) (12/07/88)
[comp.mail.sendmail added, followups directed back to comp.unix.questions.] [ Previous discussions about characters not too use in the GCOS field: ':' - obviously, this terminates the field ] In article <572@auspex.UUCP>, guy@auspex.UUCP (Guy Harris) writes: > The BSD format includes a convention that not all programs necessarily > honor, which means you may not want to use that convention: if the "&" > character appears in the "gecos" field, those programs replace it with > the login name with the first letter capitalized. This means you > probably want to leave out "&" as well. Another character to avoid is '\' if you are using sendmail (ohh noo, not sendmail again :-). If you have that in what the mail user agents considers to be the full name they will pass it on to sendmail, and bad things will happen (non-replyable adresses are bad things!). This is unfortunate, since \ is in one of the six positions that the ANSI-almost-equivalent-to-ASCII reserves for national characters. In the Swedish version of said ANSI-std capital o-umlaut is in this postion, a perfectly valid character in full names around these parts of the world. Unsuspecting sysadms are likely to enter full names with \-s in them. They, or their poor users (with names like \rjan \gvist), are in for a surprise. Svante (...am I glad all my names use regular ASCII letters only :-)