Talk:Unix/Archive 4

From Wikipedia, the free encyclopedia

This is an archive of past discussions about Unix. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

→

Multics

The History section contains the sentence `Multics was highly innovative, braving many new computing frontiers for the first time, including the ability to serve several users from the same computing machine all at one time. It had many problems, but eventually Multics became a functional commercial product.' This statement has a couple of problematic issues.

1. Multics was definitely not the first system that that served `several users from the same computing machine'. The article on Time Sharing says that CTSS was demonstrated in 1961. (CTSS ran on a specially modified IBM 7094, and individual jobs were swapped in and out, so one could say that `at one time' in the Multics statement is intended to refer to multiple processes in memory, I suppose. I can provide a source for that, if anyone cares.) Perhaps the intention is to refer to virtual memory, in which case `first' probably refers to Atlas 2, at Cambridge, also in the mid-1960s. Other somewhat contemporaneous systems include TSS/360, the Michigan Terminal System, and CP/CMS, which was released in 1968, and whose successors are still current today. Probably what is meant in this statement is `serve many users', which was part of the `central computer utility' notion that was prevalent at the time. If we move away from general time-sharing systems (implied but not required by the original phrasing), then SABRE, demonstrated in 1960, certainly supported many users concurrently.

2. Multics didn't have many problems, or at least many more than other systems of the time. (IBM's TSS/360, in 1967, turned out to be too slow for supporting more than one user concurrently, and of course OS/360 was plagued with bugs and performance problems). There is a common myth that Multics `failed', but in fact the system was first described in 1965, released in the early 1970s, and lasted until 2000 (see the Multics article, which also quotes Peter Salus as saying that it `failed miserably'). However, the lifespan, in particular the 13 years after development ceased in which installations continued to use it, doesn't suggest failure. It's certainly true that AT&T management decided that the project wasn't relevant to them, and that's sufficient for Unix history.

I would prefer to say that Multics was an early innovative system, but that the developers had underestimated the resources needed to bring it to fruition.

Vmanis (talk) 20:54, 22 February 2010 (UTC)

an example of a processing software —Preceding unsigned comment added by 110.55.7.125 (talk) 10:30, 1 August 2010 (UTC)

Logo

D*n infobox has a "logo" field, so if we can't find a real logo, why, we make our own Wikiversion. Could we get an authenticated use of a Unix logo by The Open Group or whoever holds the trademark? --Wtshymanski (talk) 20:01, 21 September 2010 (UTC)

I've always liked this one. Gwen Gale (talk) 20:59, 21 September 2010 (UTC)

Trademark

Right, the article's pretty inconsistent about Unix and UNIX already, but if we're going to use UNIX anywhere it should be accompanied by the trademark. Can we get some solid ground rules for when and where the capitalisation (and trademark) should be used? Should pages which currently use the capitalisation be moved to include the trademark in the title? (is that even possible?) Chris Cunningham 12:41, 10 January 2007 (UTC)

Unless explicitly otherwise formatted, it is "Unix"
Names of specific things (like those of books [The UNIX Programmer's Manual], websites [The UNIX Forums], formal names [UNIX System III], etc.) retain their original formatting.
UNIX® and UNIX™ only if you're like being wicked explicit and referring specifically to the trademark alone
- "UNIX" for casual reference to the trademark

¦ Reisio 04:17, 11 January 2007 (UTC)

So the system itself shouldn't ever have the trademark next to it? Is that MoS policy? It seems a bit odd to bother with a trademark symbol at all if it's never going to be attached to the thing it labels... Chris Cunningham 10:14, 11 January 2007 (UTC)

Checkout http://www.unix.org/ & Trademark#Terminology_and_symbols. ¦ Reisio 21:30, 11 January 2007 (UTC)

Cheers. Chris Cunningham 01:59, 12 January 2007 (UTC)

I notice that Programmable logic controller has chosen to use asterisks to denote that PLC is a registered trademark of Allen-Bradley, and Unix has chosen to use the extended ASCII registered trademark character. I'll let you guys duke it out, but we really shouldn't have two standards.--Superluser 00:35, 19 April 2007 (UTC)

Actually, the User:Fd0man/Templates/sc article is currently using the UTF-8 registered trademark symbol. It could use the XHTML entity reference, a footnote, or something else entirely, and be just fine. Given the power and ease of use of Unicode, however, I would be inclined to retain the UTF-8 symbol directly. —Mike Trausch ^{Fd0man•Talk to me} 17:52, 21 May 2007 (UTC)

This UNIX -v- Unix business really annoys me.

As an Australian technical writer for the last 20 years, I try to weed out 'Unix' wherever I find it and have been doing so since the mid 90s because I had to look for the correct capitalisation at the time and back then, found the ONLY usage was 'UNIX' (all capitals). All the installation media cited UNIX as did all the literature and on-line references by 'official' sources. Anything else was considered a mistake by those not knowing any better or who didn't care how they capitalised (developers, sys admin, PMs, business staff—you get the point I'm sure). Around 2002, I had to set up a new Trademarks section for another company and went through the references/TM process again and still only found 'legitimate' references to UNIX, but I DO remember being annoyed when I found the use of 'Unix' WAS now being shown, but only for 'Unix-like' operating systems (I'm sure that came from Wikipedia at the time). This disappointed me because I felt it pandered to the great unwashed who don't care how they spell or capitalise. So anyway, here I am again, looking for a definitive link to educated someone and I find the inmates are now running the asylum; so my question is: WHO authorised the change in Wikiperia from 'UNIX' to 'Unix'. What are their literary and grammar credentials. What does the Open Group say about it and if UNIX is going to be demeaned to Unix, then let's all throw the baby out with the bath water and simply call it unix. <sigh> Ask yourselves why you rarely see a professional author using the term Unix. We use UNIX (or sadly, I should say that we did). 220.237.96.213 (bja) 00:38, 7 September 2010 (UTC)

You used it wrong. Read up. ¦ 24.250.139.61 (talk) 10:42, 21 January 2011 (UTC)

USG-Unix?

In the Unix-FAQs (e.g.: FAQ 3.16) there are references to USG-Unix. Since I was not a Unix user at that time, I added what appears to me are the correct facts, but some brave soul, who remembers that terminology should check that. Treutwein (talk) 12:47, 13 October 2011 (UTC)

Who is Steven Gladstone?

The name Steven Gladstone was added to the list of folks from AT&T that developed Unix, but without a reference (but no one on the list is supported by references). Who is Steven Gladstone and did he work on Unix at Bell Labs/ATT? Sorry, if this is something that I should know. Jeff Ogden (W163) (talk) 00:49, 7 November 2011 (UTC)

I think this may have been added as a "test edit". I've removed it for now as the name was added by an IP editor with no previous (or subsequent) edits recorded, and I can't see any coroborative evidence on wikipedia or elsewhere of his involvement with any of the others, although they all have interlinking references to each other. Regards, Lynbarn (talk) 01:11, 7 November 2011 (UTC)

Clarification

This sentence (or fragment I extracted) is very confusing:
"in the 1980s non-blocking I/O and the set of inter-process communication mechanisms was augmented"
It either means that both things were augmented in which it should say:
"in the 1980s the non-blocking I/O and the set of inter-process communication mechanisms were augmented"
Or it means that the second thing was augmented within the first thing in which it should say:
"in the 1980s non-blocking I/O the set of inter-process communication mechanisms was augmented"
Someone who knows the history and what this actually means should make the change. Bugefun (talk) 22:06, 10 November 2011 (UTC)

Unix programmer's prefer prototyping over the Waterfall model?

I wrote:

"Unix programmer's prefer Software prototyping, Rapid application development, Iterative and incremental development over the Waterfall model. Prototype before polishing. Get it working before you optimize it.^[1]"

The Unix philosophy states that:

Rule of Optimization: Prototype before polishing. Get it working before you optimize it.

Build a prototype as soon as possible.

A user undid my version: "too general a statement - and not true. I am a Unix programmer, and I prefer the waterfall model"

--Christopher Forster (talk) 23:13, 25 October 2011 (UTC)

You gave a reference/source. Does the source say specifically that Unix programmers prefer something or other? If it does, I think you would be OK with your addition. If not, not. I'm guessing that you might want to reword what you wrote to be a little less all encompassing. Perhaps something along these lines:

The Unix philosophy encourages Software prototyping, Rapid application development, Iterative and incremental development over the Waterfall model; prototype before polishing; get it working before you optimize it.

With some wording like that, you aren't claiming that all Unix programmers prefer something or other. There are enough strong willed Unix programmers around that you probably can't say that all of them prefer any one thing. Whatever you write, make sure it is backed up by the source you reference (possibly even quote it here on the talk page).Jeff Ogden (W163) (talk) 23:36, 25 October 2011 (UTC)

About the waterfall model. "Experience, and a strong tradition of collaborative development, had already taught them that prototyping and repeated cycles of test and evolution are a better way."

The art of Unix programming By Eric S. Raymond

--Christopher Forster (talk) 00:37, 26 October 2011 (UTC)

I am the Unix programmer that reverted your edit! :) It is quite some time since I read that book, but whilst your quote above, IIRC, is accurate, it is the rather generalised opinion of one author, and as published in the article, suggests that all Unix programmers PREFER it. I take my medication rather than drink alcohol, not because I prefer to, but because it is the best way to survive:) Also, preferences for RAD etc aren't exclusive to Unix programmers. Which model is used also depends on the commercial environment in which the programmer is working, regardless of the platform, language etc. As worded, it didn't seem to me to be very encyclopedic, or add much to the article. regards, Lynbarn (talk) 08:50, 26 October 2011 (UTC)

Throwaway prototyping model works better than the waterfall model if you are short of time. See also Software prototyping#Throwaway prototyping

--Christopher Forster (talk) 19:23, 26 October 2011 (UTC)

I agree with Lynbarn. Just because somebody published an opinion doesn't mean that it needs to be noted in the article. Unix provided good facilities for a wide variety of software development. — DAGwyn (talk) 06:33, 21 December 2011 (UTC)

No Mention that Apple has by far the largest distribution of Unix.

No Mention that Apple has by far the largest distribution of Unix. It is used on both Macintosh Mac OS X (75 Million?) and iOS (300 Million). These both feature BSD Unix. Which has passed all the UNIX approvals. — Preceding unsigned comment added by 139.149.31.231 (talk) 13:53, 15 February 2012 (UTC)

Request for change of page title to UNIX

In common with some others, I feel this should be referred to as UNIX, and not Unix. Arguments for this are:

1) It was originally called UNIX by the developers (my comment on the main page about Kerningham and Ritchie calling it UNIX in their C programming book has been removed, despite this clearly indicating it was called that originally. Someone has changed that to indicate upper case was a convention at the time, I accept is true. Either way, it was called UNIX when developed.

2) The operating system is trademarked at UNIX and not Unix. See the web site of the owners of the trademark.

http://www.unix.org/

See for example the trademark page.

http://www.unix.org/trademark.html

To quote from there:

The correct attribution is:

"UNIX is a registered trademark of The Open Group"

Given it was originally called UNIX and the current trademark owners refer to it as UNIX, any references to Unix or unix, which have occured in other publications should be treated as incorrect usage, and not as any authorative standard.

Although I'm not suggesting the following is in itself a good argument for the use of UNIX rather than Unix, a look at the manual pages (man pages) on a modern UNIX system (Sun's Solaris), clearly shows UNIX in the title of all the pages, except one, for 'brltty' which is a braille display driver - hardly very authoritave, since braile drivers are neither part of the UNIX specification nor developed by the original developers. In other words, that one is in error - all the sytem pages, produced by Sun, use UNIX and not Unix or unix.

Here are the UNIX related man pages on a modern UNIX system

teal /export/home/drkirkby % man -k unix
authunix_create rpc_soc (3nsl)  - obsolete library routines for RPC
authunix_create_default         rpc_soc (3nsl)  - obsolete library routines for RPC
crypt_unix      crypt_unix (5)  - traditional UNIX crypt algorithm
cu              cu (1c)         - call another UNIX system
dos2unix        dos2unix (1)    - convert text file from DOS format to ISO format
kernel          kernel (1m)     - UNIX system executable file containing basic operating system services
pam_unix_account                pam_unix_account (5)    - PAM account management module for UNIX
pam_unix_auth   pam_unix_auth (5)   - PAM authentication module for UNIX
pam_unix_cred   pam_unix_cred (5)   - PAM user credential authentication module for UNIX
pam_unix_session                pam_unix_session (5)    - session management PAM module for UNIX
un              un.h (3head)    - definitions for UNIX-domain sockets
un.h            un.h (3head)    - definitions for UNIX-domain sockets
unix2dos        unix2dos (1)    - convert text file from ISO format to DOS format
uucp            uucp (1c)       - UNIX-to-UNIX system copy
uuglist         uuglist (1c)    - print the list of service grades that are available on this UNIX system
uulog           uucp (1c)       - UNIX-to-UNIX system copy
uuname          uucp (1c)       - UNIX-to-UNIX system copy
uupick          uuto (1c)       - public UNIX-to-UNIX system file copy
uuto            uuto (1c)       - public UNIX-to-UNIX system file copy
uux             uux (1c)        - UNIX-to-UNIX system command execution
xdr_authunix_parms              rpc_soc (3nsl)  - obsolete library routines for RPC
brltty          brltty (1)      - refreshable braille display driver for Linux/Unix

I've not edited the above list in any way, so some are not too relavant. Neither have I removed the single page on the braile display driver which does use Unix.

Overall, despite the fact UNIX, Unix and unix can all be found in publications, the fact remains it was originally called UNIX and the current trademark owner calls it UNIX.

PS, I guess I should have signed, this, so will do now Drkirkby 21:06, 24 October 2006 (UTC)

Nope, you're wrong.
Sure, but we don't name things based on trademarks; even if we did, "Unix" still would've existed before its trademark (and even longer before its all-caps trademark).

¦ Reisio 02:28, 25 October 2006 (UTC)

Well, given the first publications outside of Bell used UNIX, and its currently offically that, I find the attitude a little odd. I guess its like religion - you are most unlikly to convince anyone they are wrong and you are right!! I don't know if there are legal implications of not following a trademark, but that is for the Open Group to take up. It's good to see the Japanese, Russian and French Wikipedias use UNIX and not Unix. Perhaps those that feel it should be Unix rather than UNIX should learn Japanese, Russian and French and try to get those pages changed!! Drkirkby 12:40, 25 October 2006 (UTC)

I agree -- Unix jaick 03:25, 25 October 2006 (UTC)

I assume you agree not to change it, given your user name, although that is not 100% clear. Drkirkby 12:40, 25 October 2006 (UTC)

I agree - this is a history piece, not an article about a product. 'Unix' is prevalent today. The article may point out this dichotomy of course.

Citing the case in a book published four years after the year Ritchie's been quoted as saying it was accidentally capitalized doesn't mean much, IMO. ¦ Reisio 21:53, 2 November 2006 (UTC)

Oppose. This article doesn't deal exclusively with systems which have a right to use the UNIX trademark. Fix that first. Chris Cunningham 11:14, 25 October 2006 (UTC)

True, although even those pages that deal exclusively with UNIX certified systems are not consistant. I just checked the Solaris Operating System page, which is certified to use the trademark UNIX. The Wikipedia page says: It is certified as a version of Unix. It does seem wrong to me to use Unix and certified in the same sentence. (BTW Chris, did you work at MOC in Essex? If so, we have probably met.) Dr. David R. Kirkby Ph.D Drkirkby 12:40, 25 October 2006 (UTC)

I didn't, but I do work for Sun :) I still think the uppercase version has rather slipped out of the vernacular by now, and the Wikipedia style guidelines indicate that the most common name be chosen for articles. It's not like it isn't mentioned or anything. Chris Cunningham 19:38, 25 October 2006 (UTC)

Strongly Oppose. The article explains the distinction. The reason you see "UNIX" in the published documentation was that the corporate lawyers insisted on it for brand-name protection, to prevent the trademark from becoming a generic term. In fact many if not most of the original developers and users spelled it "Unix" in informal communications. (The Solaris reference should use "UNIX" since there is no certification for "a version of Unix".) — DAGwyn 05:55, 31 October 2006 (UTC)

Agree. "Overall, despite the fact UNIX, Unix and unix can all be found in publications, the fact remains it was originally called UNIX and the current trademark owner calls it UNIX." Indeed. S shadow 07:40, 31 October 2006 (UTC)

It was originally "Unics", then most likely "Unix"; this article is about an operating system originally popularized as "Unix", not the trademark "UNIX". ¦ Reisio 21:53, 2 November 2006 (UTC)

Strongly insist. The fact is that the name is UNIX, the fact is that not only dinosaurs write it that way ;) My POV is that many people tend to forget that their POV is not the only POV. If I have to write a filename, username or hostname, it would be all-lowercase. But I personally am not writing Dec, Hp, Ibm or Ms-Dos, and for some weird reason am still writing it UNIX. Actually the introduction should also change to something like "UNIX (informaly written by many as 'Unix', or even 'unix') ...". -- Goldie (tell me) 20:44, 1 November 2006 (UTC)
You're kidding, right? DEC, HP, IBM and MS-DOS are all initialisms. Chris Cunningham 21:05, 1 November 2006 (UTC)
acronyms.
IBM and HP are not acronyms. Jeez. Chris Cunningham 09:01, 2 November 2006 (UTC)
They certainly originated as acronyms, for International Business Machines and Hewlett-Packard. DAGwyn 20:13, 2 November 2006 (UTC)
As the acronym and initialism page says, it's "a contentious point" whether an initialism that's not pronounced as a word (e.g., "IBM" pronounced as "eye bee em" as opposed to "ib 'em") is an acronym or not. I think this subdiscussion is an example of that contention. Guy Harris 20:24, 2 November 2006 (UTC)
Actually, seeing as DAGwyn actually corrected me for using "initialisms", I'm not really sure what it was. Chris Cunningham 09:45, 3 November 2006 (UTC)

Wrong! No, I am not kidding, joking or whatsoever. Thank you for asking anyway. You're teasing, right?

For those who have not understood what I have written, I shall repeat: My POV is that many people tend to forget that their POV is not the only POV. If one desires so, (s)he can read Wikipedia:Neutral_point_of_view/FAQ#Anglo-American_focus - people outside U.S.A. do not bother what is behind the scenes, and tend to take some things as they are. A native English speaker might instantly recognize what is acronum and what is not. However I doubt that every native speaker will do so at first glance and am rather certain that almost all non-native speaker will not. After some time have passed the initially used form is carved in stone (or engraved on the one's forehead) and regardless whether it is right or wrong, the person uses that form. -- Goldie (tell me) 07:30, 2 November 2006 (UTC)
Errr, except this hasn't actually happened in Unix's case. The title-case form is at least as common as the upper-case in the vernacular. You haven't yet bothered to explain why the article should be upper-case rather than simply insisting on it. Chris Cunningham 09:01, 2 November 2006 (UTC)

And what exactly is that unknown "that" which haven't happened? Is at least as common where? Have ever bothered to read what is going on beyond the East Coast and/or West Coast (of U.S.A.).

In my part of the world beginners are learning from official (name them formal) materials. Go to Google or Yahoo and search "site:ibm.com unix" or "site:hp.com unix", you might be surprised to see the complete absense of any other form than all-caps. And I've tried to explain that after people get accustomed to that writing they tend to stick to it.

If you enjoy it, interpret it that way - it is your word ("I know it my way") against mine ("I know it my way"). Are you trying to suggest that I've been stupid and illiterate last decade or two, or what? -- Goldie (tell me) 16:14, 2 November 2006 (UTC)

The article for PINE uses "Pine", the article for EMACS uses Emacs. Historically, these were in uppercase, too. --69.173.175.94 17:51, 2 November 2006 (UTC)

There is no, "too". :p Again, there are actual sourced claims at /Archive 1#UNICS_vs_UNIX. ¦ Reisio 21:53, 2 November 2006 (UTC)

Agree: Since this article is focused primarily on UNIX rather than Unix-like, I agree it should be retitled UNIX. -- Steven Fisher 22:19, 12 November 2006 (UTC)

Oppose: The lawyers called it UNIX, the developers and users called it Unix, except when publishing in a context controlled by the lawyers.--Per Abrahamsen 09:16, 23 November 2006 (UTC)

Compromise: What about this? A template can be used to render the name Unix as User:Fd0man/Templates/sc (which, on browsers that support the CSS standard, will render small-caps, while non-standards compliant browsers will merely render “Unix”). Using small caps means that the word “Unix” shows up in the source as a title-cased word, and browsers that support small-caps will see UNIX with NIX being smaller than the “U”. A special-case template (say {{unix}}) could be used, or a general template (e.g., {{smc|Unix}}) could be used. Presently, I have a template at User:Fd0man/Templates/sc that can be used to write User:Fd0man/Templates/sc in a way that approaches a compromise on this issue. —Mike Trausch ^{Fd0man•Talk to me} 17:46, 21 May 2007 (UTC)

But the common usage is "Unix", not small caps. The current article makes the distinction well enough, and agrees with general usage. No change is necessary. — DAGwyn 19:33, 21 May 2007 (UTC)

Common usage is either “UNIX” or “Unix”—one can use Google to find more than enough references for both methods of writing it in all sorts of fields and formats, and one can consult the local library or the bookshelf of someone with many books on the subject for even more (I know that I have tons of books on User:Fd0man/Templates/sc-like subject and they refer to it in a mixed fashion, as well). Thus, User:Fd0man/Templates/sc is the logical compromise, and would very much so make it so that consistency were actually possible throughout the entire Wikipedia. The small caps is, as I said, a compromise, because it addresses both having smaller letters for “nix” as in typical English usage and the all-caps “NIX” to accomodate those who use it that way (including vendors, writers of published papers, and the founders of the system themselves). Also, the small-caps version “degrades” to plain “Unix” in browsers that do not support it. I would love to hear an argument against it, though.

The major arguments for Unix are that it is written that way now, by most people involved in things close to User:Fd0man/Templates/sc-like development today. As much as I like these people and the work that they do, and might personally be biased towards it because it is just easier, it still is not a compromise like User:Fd0man/Templates/sc is. As I recall, Wikipedia is not a place where popularity is the key factor, either, so what is popular is not always relevant. My point, all in all, is that there are two positions here: one for Unix and one for UNIX. User:Fd0man/Templates/sc meets up somewhere in the middle of those two positions. Further commentary is of course welcome. —Mike Trausch ^{Fd0man•Talk to me} 15:34, 22 May 2007 (UTC)

My point remains: the small-caps form is not common usage, and is not the approved form for the trademark (which is the main motivation for the upper-case form). It's a compromise that nobody would like (except perhaps those with deficient browsers). — DAGwyn 22:25, 22 May 2007 (UTC)

Okay, a lesson in basic grammar is apparently due here. Whenever a word is written in all-caps, the implication is that it's an acronym. In other words, if you write "UNIX" in an encyclopedic article, readers will be left wondering what on earth it stands for ... is it "Universal Network Intralinked Xylophone" or what. Since Unix doesn't stand for anything at all, then it's properly written as "Unix"; that's just basic grammar. The fact that the people behind the software choose to write it as "UNIX" changes nothing; that's a common marketing ploy, intended to make the name standout, not unlike the marketers behind "GLOCK" and "REALTOR". Nolefan32 00:54, 21 June 2007 (UTC)

Maybe UNIXES? Or is that UNICES?Shjacks45 (talk) 18:55, 16 March 2012 (UTC)

[1]
Raymond, Eric S. (2003). The Art of Unix Programming (Addison-Wesley Professional Computing). Addison-Wesley. p. 406. ISBN 0131429019. Retrieved 2011-10-25. {{cite book}}: More than one of |pages= and |page= specified (help)

This UNIX -v- Unix business really annoys me.

Related Articles