Talk:Unix/Archive 3

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

Archive 2

Archive 3

Archive 4

Archive 5

→

Archive 9

I took out the link. Now will you quit deleting section. There is already an article on it. Unix electronics There should be a disambig at the top of the Unix article. Someone should make an article and a disambig notice. Deleting discussion sections is very rude. --Gbleem 03:23, 4 November 2006 (UTC)

Deleting the Unix Electronics article is also very rude. --Gbleem 03:33, 4 November 2006 (UTC)

Please read WP:CORP out guidelines on what makes a company notable enough to warrant a separate article on Wikipedia. It is not rude to apply the rules that the community has agreed on. We do not tolerate spamming here. Thanks, Gwernol 03:36, 4 November 2006 (UTC)

You are right it's not that big of a company. I only put the link on to show it was a real company. I was surprised when I arrived in Korean and saw Unix on hair driers and such. I'm surprised there hasn't been a lawsuit. I didn't get to read the article but I remember it was very short. --Gbleem 04:04, 4 November 2006 (UTC)

It cracks me up everytime I see the commercial that says, "Unix is magic." --Gbleem 15:44, 25 November 2006 (UTC)

The list of systems licensed to use the UNIX brand appears to be unsourced, and rather dated as well (e.g. A/UX has been effectively dead for over a decade). The Open Group website contains a register of licensees, if anyone has time to update the article. However, that may not be the end of the story, if derived systems are allowed to use the name as well. For example, Microsoft licensed the UNIX SVR5 source code from the SCO Group, for the UNIX subsystem offered in some versions of Windows, and subsequently started using the 'UNIX' name for the resulting subsystem, in lieu of the former 'Interix' name. Does that mean Microsoft's UNIX subsystem for Windows is a properly licensed UNIX? I don't know, but apparently an earlier version Interix had been officially certified as UNIX (http://www.opengroup.org/press/15_oct98.htm), so I suppose that may still be valid. Shalineth 21:47, 3 March 2007 (UTC)

I found a page on "Stat-Unix" while doing a search for unix programming topics (3/4/07) I feel strongly that the resources of Wikipedia should not be wasted on recreating or mirroring another "d_mned" unix manual on the web. There are ample resources for programming and operating systems related concerns on the web and so while the "history of" would be alright for Wiki. It is wholly inappropriate as a mirror or summary of so much material already available on the web (last look +1 million hits and counting)... Wiki editors you need to put limits which prevent wiki from becoming a "web to itself". Like a good poem...Wiki will benefit from the constraints of common sense. 71.156.174.2 21:28, 4 March 2007 (UTC)

While I agree with the intent of that, perhaps there was some specific reason a "Stat (Unix)" article was created, such as to elaborate on what "stat" meant in some context where the elaboration would have been distracting. -- DAGwyn 22:26, 6 March 2007 (UTC)

The "Impact" section says:

... Over time, text-based applications have also proven popular in application areas, such as printing languages (PostScript), and at the application layer of the Internet Protocols, e.g. Telnet, FTP, SSH, SMTP, HTTP and SIP.

FTP's command channel was text-oriented at least as far back as RFC 354, in July 1972. Telnet dates back at least as far as RFC 318, from April 1972. The quoted sentence seems to be at least suggesting that the popularity of text-based applications stems from UNIX's text orientation, which is definitely not true of FTP or Telnet, and might not be true of the other protocols, either (SMTP, HTTP, SIP, RTSP, etc. are somewhat modeled after FTP, and SSH is a remote login protocol like Telnet, and the concept of remote login long antedates UNIX). Even if that's not what the sentence is saying, it does seem to suggest that the popularity of text-based applications at the application layer of Internet protocols came after the popularity of UNIX, which is also not the case. Guy Harris 00:49, 22 July 2007 (UTC)

Unix certainly popularized the use of textual representations of data instead of binary formats. I don't know about those specific protocols. Are all of SSH's transmitted characters really confined to text characters? (SSH was a recent addition to that list, by the way.) There may have been some influence on PostScript. — DAGwyn 23:33, 23 July 2007 (UTC)

No, SSH's transmitted characters aren't confined to text characters; neither are RSH's. I'd say we should at least remove SSH from the list - and I'd be tempted to remove Telnet, too, as remote login commands tend to be text-based because humans typing at terminals tend to be text-based. :-) Guy Harris 00:15, 24 July 2007 (UTC)

yes we can justify this inference actually if you look at the history of 'tools' in unixworld. --Buridan 23:44, 23 July 2007 (UTC)

We can? To which inference, and to which, tools are you referring? (In particular, if you're referring to the suggestion that the popularity of text-based protocols can be credited to UNIX, to which 1972-era UNIX FTP command are you referring? :-)). Guy Harris 00:15, 24 July 2007 (UTC)

which tools? or which unixen? ftp was developed on what system? unless you are making a differentiation between unixen and bsd... i have ftp as 4.2bsd, what do you have it first written in? one other point here is the bsd stack author and the bits he stripped out of the bbn stack. text and stream tools specifically pipe tools of unix helped popularize early internet ala the software tools book. you can probably actually cite that line from one of the history books. the inference that i am arguing for is that the popularity of text tools in unix encouraged the development of an internet that was textual. i think that inference is valid and i'm pretty sure there might be a citation for it. --Buridan 22:33, 24 July 2007 (UTC)

I meant "which tools". FTP wasn't originally developed on any Unix system. If you look at RFC 354, you'll see references to PDP-10s (probably running either TENEX or TOPS-10), "360's" (meaning IBM System/360's, probably running one of various OSes, including OS/360), and Multics. You won't see any references to Unix, because, as of 1972, when that RFC was written, Unix had just been ported to the PDP-11 a year ago, had not been ported to the PDP-11/45 yet, and wasn't available outside AT&T, according to Dennis Ritchie's notes on a talk he gave in 1972. Nobody'd connected Unix systems to the ARPANET yet.

The Software Tools book had nothing to do with the Internet, or even the ARPANET. And as for the Internet being "textual", at most, one can say that some protocols were textual, e.g. FTP, but FTP with textual commands was created before Unix systems were on the ARPANET (and before the Internet existed). The data transferred by FTP could be textual or binary. The Simple Mail Transfer Protocol also used textual commands, but that was inherited from FTP. The mail message format was textual going back at least as far as RFC 733; there might have been Unix systems connected to the ARPANET at that time, but they hardly made up the majority of ARPANET systems at that time. Even in a world with no Unix systems, text had the advantage of being a simple format that multiple different operating systems could agree on as a transfer format, even if they stored the text files locally in different ways.

So I see no evidence to suggest that the inference that Unix somehow encouraged the use of text-oriented protocols and data formats on the Internet is valid; they were used before Unix was on the ARPANET, and were used when Unix was more than just one of many operating systems on the ARPANET and the Internet, and there were reasons other than "Unix uses text files" for using text-oriented protocols and text file formats. I'm pretty sure you either won't be able to find the citation you think you'll find, or will find a citation that's ignorant of ARPANET history. Guy Harris 23:23, 24 July 2007 (UTC)

It mentions application layer, not the bits at the network layer. SSH and the others were created for text communication with a command line interface on a text terminal. They were, like many UNIX tools, generalized so they have many uses. SSH is "Secure SHell", a text command line interface, a secure replacement for telnet. (SEWilco 02:53, 24 July 2007 (UTC))

And what about the notion of text communication with a command line interface on a text terminal originated in UNIX? (Hint: the correct answer is "absolutely nothing".) Guy Harris 06:50, 24 July 2007 (UTC)

The point the article was making, or should be making, is that Unix showed that using a universal human-readable, text-editable format for all kinds of file contents allowed a small set of simple tools (filters, etc.) to perform many needed file-content operations very well. Prior to Unix, it was rare for very many programs on a typical system to be able to process the contents of any particular file. Some OSes tried to address this by interposing a "record manager" to provide access to, and translation between, different known file formats. The Unix approach (make nearly every data file a text file, with only one record-level format) was very successful, and may have inspired some network protocols — but not all the ones listed in the article. As to the network protocols, if what is transmitted for the protocol itself (encapsulated in the IP packet) is all-text, such as HTTP is, that is "kind of" Unix-like, and allows use of TELNET to the port for testing, for example, but that model was in place (to some extent) in the ARPAnet before Unix was used there. So it's not correct, for some of those protocols, to attribute the design to inspiration from the Unix model. — DAGwyn 14:22, 24 July 2007 (UTC)

"i have ftp as 4.2bsd, what do you have it first written in? --Buridan 22:33, 24 July 2007 (UTC)"

Ahh, you kids and your fancy Unixes and tubes and trucks... . Read RFC 114:

The protocol described herein has been developed for immediate

implementation on two hosts at MIT, the GE645/Multics and the PDP-

10/DM/CG-ITS (and possibly Harvard's PDP-10).

--NapoliRoma 23:25, 24 July 2007 (UTC)

The green-on-black screen images are totally useless in my browser (Firefox on Solaris). Can they be redone, perhaps by converted to black-on-white via Photoshop? — DAGwyn 00:35, 17 February 2007 (UTC)

Define "totally useless". Are you using a monochrome monitor? I didn't know they still made those.

The images are an attempt to capture the look and feel of Unix as it might have been seen back when it was usually accessed through green-on-black or amber-on-black VT-100 terminals, in contrast with today, when it is accessed through graphical, full-colour VGA displays, with the shell being something you only use occasionally through the small window of the (typically black-on-white) xterm program. As the author of the images in question, I'd be more likely to replace them with photographs of Unix as seen on VT-100 terminals (if I could get access to such systems) or hard-copy terminals (if I could get access to those, and no, xterm does not look like a hard-copy terminal) than to edit them so as to look more xterm-like. 216.23.105.16 14:17, 1 May 2007 (UTC)

VT100 was never green on black nor amber on black. (DEI did make a green-phosphor VT640 that was a modification of a VT100.) Actually I was talking about the thumbnails, which look almost totally black. They are viewable enough when clicked on to show them separately. — DAGwyn 17:01, 1 May 2007 (UTC)

that sounds like a screen gamma issue with your terminal. the thumbnails look like that on windows but they look fine at home on linux69.125.110.223 (talk) 20:21, 18 December 2007 (UTC)

No, it is not a gamma issue! Most likely it has to do with how the image is sampled to downsize it to a thumbnail. And if you acknowledge that the thumbnails look as I described on Windows then certainly there is a problem. Note that I reported the problem for Firefox on Solaris, so there are at least two problem environments. — DAGwyn (talk) 05:05, 19 December 2007 (UTC)

Image:Solaris8-cde.png is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images lacking such an explanation can be deleted one week after being tagged, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

BetacommandBot (talk) 08:48, 21 January 2008 (UTC)

I see this all over Wikipedia, and this is what it says in the Overview section:

"...operating systems such as Linux..."

That sentence should be changed to a context where describing Linux as a kernel (which it is), would make sense. Or it should use the preferred term for the GNU operating system with the Linux kernel: GNU/Linux. —Preceding unsigned comment added by Afarnen (talk • contribs) 07:25, 18 January 2008 (UTC)

Except for pedants, everybody refers to the combined package simply as "Linux". There certainly is not a "GNU operating system," although Stallman originally intended to create one eventually. — DAGwyn (talk) 16:08, 18 January 2008 (UTC)

You can call any operating system with a GNU userland and a GNU-compatible kernel a GNU operating system. You can also call any operating system with a Linux kernel a "Linux operating system". This is sometimes expressed by referring to the combined product, which is the most common use of the Linux kernel, as "GNU Linux", but again that's open to accusations of pedantry. "Operating systems such as Linux", when referring primarily to attributes of the Kernel, is fine. --Tony Sidaway 16:17, 18 January 2008 (UTC)

Another point: When one buys "Debian Linux" or "Red Hat Linux", he doesn't receive just a kernel. — DAGwyn (talk) 16:05, 31 January 2008 (UTC)

It is "Debian GNU/Linux" and "Red Hat Enterprise Linux", just to be pedant (look at their home pages).--Per Abrahamsen (talk) 18:58, 31 January 2008 (UTC)

Isn't Linux the operating system in a GNU/Linux system? The GNU applications run under the Linux OS. -- SEWilco (talk) 16:13, 31 January 2008 (UTC)

Most people consider an OS to be more than a kernel, and most people who use the term "GNU/Linux" seem to want to tag the Linux bit as being "just the kernel", not an OS on its own.

Although I proudly let my pedant flag fly, in this case usage is way on the side of using the term "Linux" to describe the operating system, and I don't think there's any slighting of GNU intended in doing so. I think any reasonable Linux person (not an oxymoron, I hope) would be glad to agree that the GNU project is a major contributor to the thing called Linux. However, It's very clear that, just as one example, the number one commercial distro is "Red Hat Enterprise Linux," not "Red Hat Enterprise GNU/Linux." To attempt to impose the "GNU/Linux" styling within WP is not being encyclopedic, it's promoting a point of view.--NapoliRoma (talk) 17:34, 31 January 2008 (UTC)

what is awk command? —Preceding unsigned comment added by 210.212.251.163 (talk) 12:48, 9 September 2008 (UTC)

See Awk. Gwen Gale (talk) 13:43, 9 September 2008 (UTC)

What is the market share of the various proprietary and open source distributions, for e.g. the server, desktop, and other (high-end computing?) markets? -- Beland (talk) 15:44, 6 March 2009 (UTC)

I'm not exactly the tightest holder to these things, but why exactly is there a screenshot of Ubuntu on the Unix page? Is it necessarily related enough to be posted while still adding to the content—or at least not being a superfluous image.

Specifically, I'm referring to this: [[Image:Live USB ubuntu.png|thumb|left|240px| Linux has been adapted to boot from [[CD]]s, [[USB flash drive]]s and [[memory card]]s.]] 24.94.62.194 (talk) 07:25, 8 October 2008 (UTC)

I've rm'd the image. Gwen Gale (talk) 07:55, 8 October 2008 (UTC)

Well there is nothing to say its wrong. The OpenSolaris (UNIX) kernel has made to work a modified Ubuntu (GNU) userland. —Preceding unsigned comment added by 134.36.93.46 (talk) 04:55, 21 May 2009 (UTC)

I'm not exactly sure whether or not UNIX still has modern distributions available, but if it does I think this page would do well to have a section with a list of current versions being used. I'm not sure if I'm being clear, but something similar to the list of distributions which would exist on the GNU page. Mlo0352 (talk) 19:19, 30 July 2009 (UTC)

First one would have to decide what qualifies — trademark licensees? look-alikes? There is considerable information already in the article about UNIX systems, for example Sun's Solaris, which is available in both open-source and proprietary (supported) versions, and Apple's OS-X. Why is that not sufficient? — DAGwyn (talk) 07:34, 8 August 2009 (UTC)

Which trademarks does that phrase refer to in the introduction? One of the trademarks referred to is clearly "Unix," but what's the other? Is it just a variant spelling? —Preceding unsigned comment added by 76.121.176.186 (talk • contribs) 02:45, 23 September 2009 (UTC)

In my understanding it's the variant spelling: UNIX and Unix... both. man with one red shoe 03:09, 23 September 2009 (UTC)

Well-spotted. Unreferenced, confusing anon change reverted. ¦ Reisio (talk) 04:20, 23 September 2009 (UTC)

With no reference to the section immediately below I'd like to point out that the article's bit about the evolution of the name 'Unix' is a bit hazy at best. Kernighan did in fact find the name as a joke - on Multics. The decision to rewrite Unix in C was prefixed by an interest in first creating C. This part did NOT take three years.

Kernighan has himself told the story of how they all watched ken and dmr at their 'grand opening': ken logged in to the console, everything was fine; then dmr tried to get into a terminal and the system just stood still. BWK broke the silence with his now famous quip. That 'Unics' has been backronymed here is I think a bit silly. That ken immediately after this 'fiasco' decided he and dmr needed a new language is also part of the official version. And after dmr had C they set about with rewriting the kernel.

And I've not read this account anywhere. It was told to me by BWK himself, face to face. If someone wants my identity - fine; but BWK is an amenable chap and I'm sure someone who wrote to him and asked him would receive a cordial and informative reply.—Preceding unsigned comment added by 193.92.203.237 (talk • contribs)

So write him. ¦ Reisio 02:05, 14 January 2007 (UTC)

I too think that the etymology of 'Unix' is doubtful, and at least could be more complete, but I am not sure exactly how the article might be improved.

Back in 2002 I investigated a little and asked some of the people involved, as I had understood that there was the idea that Unix was a pun on 'castrated Multics'. Below are two responses on the subject:

From: "Peter G. Neumann"
Subject: Re: Origin of the name Unix
There is some confusion.
I indeed referred to an early version of the system as a "one-user castrated Multics" (UNICS). Ken's very first code did not support multiple users, although that changed rapidly.
As I recall, Peter Salus's book A Quarter-Century of Unix gives me credit for the name. But I think perhaps Brian Kernighan seems to deserve the credit for the system name (Unix) in which case I only gave it an interpretation. You might ask Brian what his recollection is.
Peter

So I did:

From: "Brian Kernighan" Date: Thu Feb 07, 2002 06:33:20 pm Europe/London
To: [email protected]
Subject: re: Origin of the name Unix

my memory (of the same vintage as peter's, of course, and perhaps even more suspect) is that i coined the name "unics" but as a vaguely latin pun: multics was many (multi) of everything, while unics was at most one (uni-) of anything. that's not a bad characterization for the time. i do not know who improved the spelling. i could readily believe that peter supplied the castrated multics interpretation -- he has more of a way with words than i do.
good luck with the scholarship, though it will probably be hard to get further on this specific aspect.
brian k

So I am not sure where that leaves us - the people who were involved are not sure, but it is certainly unlikely that the acronym in the article is correct.

The quip by bwk was after witnessing the first ever launch of Unix in ken and dmr's office at the CSRC. ken started the machine up at the console - so far so good - but then dmr was to log in to the system which was still written in interpretive B. The machine didn't have the power to run two terminals at once, prompting a chuckle (and bwk's famous wise crack). PN was at the labs until 1970, but his name is never mentioned in that context.

There was no slur about the system being castrated. According to bwk, he added: 'Unics - a multi-user system for at most one user'.

I would at least suggest deleting ", short for Uniplexed Information and Computing System, and could - eventually - support two simultaneous users", unless someone can show authority?

--Hugh.glaser 14:35, 6 February 2007 (UTC)

How's UNIX pronounced, anyway? Do you actually say "eunuchs"? SuperGerbil 15:16, 19 May 2007 (UTC)

Pretty much. "YEW-nicks"; I'll leave the IPA to others.--NapoliRoma 17:15, 19 May 2007 (UTC)

I don't know where the name came from, but I was at Murray Hill from 1981-1992 and didn't hear kind things about Multics. The story I heard was that UNIX was meant to be a cleaner leaner operating system. Ideas were borrowed from Multics, but UNIX was also a reaction against what the Bell Labs folks percieved as an overly complex and inefficient Multics system. DonPMitchell (talk) 04:03, 25 July 2008 (UTC)

ken got management to order two PDPs. They arrived - but their OS did not. He and dmr used to spend a lot of time speculating what the OS would be like, coming up with ideas of their own. Joe Osanna told them to stop talking about an OS and write one instead.

Include Section - I noticed the header/summary paragraph does not mention anything about the 'UNIX' naming. I think a short sentence or two would be nice to have at the top of the page (and not hidden in the History section). --Evoluzion (talk) 03:49, 30 September 2009 (UTC) ......... —Preceding unsigned comment added by 59.97.33.172 (talk) 10:10, 19 July 2010 (UTC)

The sentence `During the late 1970s and early 1980s, the influence of Unix in academic circles led to large-scale adoption of Unix (particularly of the BSD variant, originating from the University of California, Berkeley) by commercial startups, the most notable of which are Solaris, HP-UX and AIX.' in the overview is rather ungrammatical, leading one to the conclusion that AIX was a startup. A trivial fix is to replace `startups' with `companies', but the real problem is that `adoption' and `derivatives' are muddled in this sentence. I don't feel I have the right phrasing yet, but something like this might be better.

`During the late 1970s and early 1980s, the influence of Unix in academic circles led to large-scale commercial adoption of Unix, both in its original AT&T form and in variants, particularly the BSD variant, originating from the University of California, Berkeley. Subsequently, some computer vendors produced derivatives of the original AT&T code base; the most notable of these are Solaris, HP-UX and AIX.'

Vmanis (talk) 20:13, 22 February 2010 (UTC)

Unfortunately it is hard to get this right in a single sentence. Solaris, HP-UX, and perhaps AIX started out based on the BSD variant, migrating to AT&T System V later. Therefore "derivatives of the original AT&T code base" would be misleading. Sun was a start-up, but not HP nor IBM. There were a lot of "supermini" startups who chose some form of Unix for their primary OS. — DAGwyn (talk) 03:31, 24 July 2010 (UTC)

Given the wording of the text is indeed a bit fuzzy, I don't quite understand what you mean, since all the variants were drawn, one way or another, from the AT&T code base. Moreover, I don't see what startups have to do with it and don't know what you mean by "supermini" (or why it matters). Hence, I feel kinda thick-headed here. To save some time, could you go into this a bit more? Gwen Gale (talk) 13:56, 24 July 2010 (UTC)