Talk:Berkeley Software Distribution

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search


It would seem, in my opinion at least, that this does not have many citations or reliable sources mention in the article. I would add them myself, but I'm really not much of an BSD guy and would not know where to look. Just delete this message if it is unneeded or useless. Zen Clark (talk) 00:24, 3 November 2008 (UTC)

Older discussion[edit]

Am not sure why this was entered under "Berkeley_System_Distribution" but the NetBSD documentation 1 has the S standing for Software, and that seems to be the predominate usage (even in the article itself, leading to the odd situation in which the article and article title don't match). --JoeAnderson

In the early days it was called "Berkeley Source Distribution". This was certainly the case still in 1984. It has changed at some stage since then, but i don't know why or when.

the /usr/share/doc/en_US.ISO8859-1/articles/bsdl-gpl/unix-license.html file ("Unix from a BSD Licensing Perspective") on FreeBSD 6.3 (on the 6.3-RELEASE-i386-docs.iso ISO at [1]) says "Berkeley Standard Distribution":

Unix author Ken Thompson returned to his alma mater, University of California Berkeley (UCB), in 1975 and taught the kernel line-by-line. This ultimately resulted in an evolving system known as BSD (Berkeley Standard Distribution).

It goes on to mention enhancements such as the TCP/IP stack and then licensing issues with AT&T. The file is also available under

This book and this book both have it as "Berkeley Software Distribution" and don't mention other variations. Letdorf (talk) 12:17, 13 February 2008 (UTC).
Well, I'd always heard/said Software, but I think it's worth mentioning variations in the article. Ramorum (talk) 17:24, 13 February 2008 (UTC)

—Preceding unsigned comment added by Ramorum (talkcontribs) 06:52, 13 February 2008 (UTC)

Trivial BSD's shouldn't be listed at the same precedence as the 3 major ones. Relevance. I'd be suprised if more than a few hundred (or in some cases, dozen) people run the smaller ones.

I've changed to the structure used at Unix-like - better? DragonFly is still a bit minor to get its own listing IMO - David Gerard 13:26, 14 Aug 2004 (UTC)
I don't mind how they are listed (though I would prefer more information over less), but the last revert removed TrustedBSD. Should it be put back?
Tim Ivorson 06:54, 15 Aug 2004 (UTC)
I suspect Darrien was reverting blindly. I've directed him to this talk page - David Gerard 11:51, 15 Aug 2004 (UTC)
Although his new layout does work well enough :-) - David Gerard 22:47, 16 Aug 2004 (UTC)
DragonFly is still a "bit player" to be sure, but unlike the other trivial BSDs, it is backed by longtime BSD developers who are doing significant, low-level work, and it seems fitting that they have a place with the big three. 22:49, 15 July 2005 (UTC)


I've changed the BSD page from a redirect to a disambig, including the Birsa Seva Dal expansion recently mistakenly added to this page. yes —Preceding unsigned comment added by (talk) 07:17, 13 March 2008 (UTC)

FOSS licenses[edit]

Isn't there a FOSS Licenses called BSD?

Of course. There is a link to in this article even, but it is a bit buried in the history section. See BSD license. - Taxman 09:30, Jan 22, 2005 (UTC)


The current history section is extremely skimpy and mostly ignores the actual historic bits (1BSD, 2BSD, 3BSD, 4BSD, 4.1BSD, 4.2BSD, and 4.3BSD). A capsule summary of the history is available in the intro to McKusick et al, The Design and Implementation of (insert flavor here). I may write this some day, but anyone else out there should feel free to preempt me. 03:44, 26 Feb 2005 (UTC)

standards compliance is untruthful[edit]

User:Taxman writes, "if you have something that shows standards compliance is untruthful bring it to talk", so...

Just how many years have BSD systems been intentionally failing to support the "ps -ef" command? This is not merely a bug. This is willful violation of the POSIX and UNIX standards.

Fair enough. I don't have any knowledge to agree or disagree. Do you have a reference to back up that it is "willful violation of POSIX and UNIX standards"? - Taxman 18:52, Mar 29, 2005 (UTC)
  • That it is standard: just read the UNIX standard
  • That is is willful: well, just send in a patch to fix the behavior and you'll see. (this flaw is very well known)
AlbertCahalan 02:26, 5 Apr 2005 (UTC)
There's also "ls -lg" and "ls -lo" not working right.
Is that a standard? Again references. - Taxman 18:52, Mar 29, 2005 (UTC)
Same place, same standard AlbertCahalan 02:26, 5 Apr 2005 (UTC)

Yet another willful violation: "ps -u root"

There are many more... those are just the most glaring ones.

It's not right to claim that "the BSD operating systems are notable for their standards conformance" when they are notable for their lack of standards conformance.

Again, have a source to support that? - Taxman 18:52, Mar 29, 2005 (UTC)
The above qualifies I believe. AlbertCahalan 02:26, 5 Apr 2005 (UTC)
-e and -f are clearly marked as XSI in the 1003.1-2001. These options did not come from POSIX, they came from XPG3, and in any case I don't believe any of the BSD's have ever made it a goal to implement the XSI option (which mostly memorializes mistakes from System V that Berkeley had the good sense not to copy). These options were not included in the mandatory part of the standard specifically because there was and is no consensus on making them mandatory. All of the command-line options about which you complain are part of XSI, except for "-o" which is implemented to specification.
In any case, the 4.4BSD library and most of the userland tools were quite definitely written to the 1003.1-1990 standard and to drafts of the (then still in development) 1003.2 effort. 06:33, 17 Apr 2005 (UTC)

The article could say "the BSD operating systems might someday be notable for their standards conformance", but it seemed more polite to just quietly delete that section. If you want to air the dirty laundry though, be my guest.

who wrote this???


I wanted to read about tha famous BSD-style ports and was amazed that I haven't found any, only a short paragraph under FreeBSD. Would someone please create an article? Helix84 18:20, 29 Mar 2005 (UTC)

Good idea, they seem notable enough for their own article. The trick is to pick the right article name. ports (BSD) perhaps? - Taxman 18:52, Mar 29, 2005 (UTC)

history confusion about 4.1BSD[edit]

"(The release was not called 5BSD to avoid confusion with AT&T's UNIX System V release.)"

This is bunk. SysV wasn't on the roadmap in 1981. Who is claiming this?

I've never heard that about 4.1, but I have heard that was a reason with 4.2 (however, I can't track down a solid ref). However, Don Libes' "Life with UNIX" says on pg. 18 says 4.2 was originally slated to be 5 BSD, but they would have had to relicense it with AT&T according to university rules.--Agarvin 23:25, 16 October 2006 (UTC)

System V was concurrent with BSD 4.1 and 4.2. We ran BSD in our department at Bell Labs in Murray Hill in the mid 1980s, and we were asked several times why we were not running System V. It should be noted that John Reiser's version of VAX UNIX in Bell Labs did have virtual memory, well before Berkely's system did. DonPMitchell (talk) 19:15, 14 April 2009 (UTC)

"Linux comparison section should be reworked"[edit]

The prose is awful and I feel the content favors a loyalty perspective to informative facts. Furthermore, it completely re-lists the bulleted summaries of the major goals of each distro.The preceding unsigned comment was added by (talk • contribs) 13:05, Jun 20, 2005.

I agree, the summaries are redundant and should probably just be removed. As for the rest, why not just fix what you see? The article FreeBSD and Linux has some more material, but it's not that well written either. I gave up because I couldn't think of a way to write it NPOV enough. - Taxman Talk 17:22, Jun 20, 2005 (UTC)
I feel we should just link to FreeBSD and Linux instead. Huge parts of the section are completely laughable. Ninuor 00:49, 22 Jun 2005 (UTC)
I tried rewriting it. It was still shit. Replaced it with a link to the above-mentioned article.Ninuor 01:00, 22 Jun 2005 (UTC)

Timeline mistake?[edit]

There is a mistake in timeline. FreeBSD was forked after NetBSD. First NetBSD release was in 1993 April, FreeBSD was in 1993 December. Probably need to fix it?

And why does BSD1 appear on the chart? IIRC, wasn't BSD1 just a tape containing Berkely's Pascal compiler and ex? ( 00:20, 1 November 2006 (UTC))

Why is Mac OS X not included in the timeline? It definitely is a Unix system. —Preceding unsigned comment added by Mijter (talkcontribs) 20:16, 11 November 2007 (UTC)


Devilette is a keen glance brunette woman dressed like BSD Daemon:

Somebody please contact copyright owners and upload these pictures to WikiCommons :-) Vugluskr 15:23, 14 January 2006 (UTC)

We only need one of these linked from the article, surely. I've cut it down to Ceren since she was the original, as far as I am aware. NicM 14:59, 14 January 2006 (UTC).
AFAIK all'em were made in one exhibition by different people. I'll not upload them to WikiCommons, I just want to inform you that I've found four pictures. They're beautyful. Much better than penguins, flags or rhombs. Upload them to WikiCommons and put a square like "Wikimedia has pictures about devilettes". Vugluskr 15:23, 14 January 2006 (UTC)

On Portal:Free software, BSD is currently the featured article[edit]

Just to let you know. The purpose of featuring an article is both to point readers to the article and to highlight it to potential contributors. It will remain the feature for a week or so. The previous feature was MySQL. Gronky 10:31, 24 February 2006 (UTC)


it goes




This error is now corrected. Jean-Baptiste CAMPESATO / camje_lemon 15:33, 13 May 2006 (UTC).

EuroBSDCon is not mentioned[edit]

This conference seems worth mentioning, but I don't know much about it. Gronky 13:10, 31 August 2006 (UTC)

add Prof. Bob Fabry to BSD History[edit]

Prof. Bob Fabry should be added to the BSD history on this page. Peter Salus has some info on him in A Quarter Century of UNIX. Fabry secured the grants that allowed Unix development to occur at Berkeley, and supervised it. He had the vision that UCB could make a big difference in Unix. He was responsible for getting Bill Joy involved. Bill Joy lucked into the environment that Prof. Fabry set up. Without Fabry therw ould have been no BSD Unix. - Lentower 21:38, 6 October 2006 (UTC)

Origin of BSD's virtual memory code[edit]

At the time, in the 1980s, when BSD with virual memory came out, the story we heard was that this code was developed elsewhere (Ohio I think) and incorporated by Berkeley without proper attribution. Does anyone have some evidence for this or know where the VM code came from?

I do know for a fact that UC Berkeley did plagiarize some code. The program vroff (troff for the Versetac printer) was written at the University of Toronto. The BSD version of this program was exactly the same source code, but the names of the authors at Toronto was removed and a UCB copyright boilerplate put in its place. DonPMitchell (talk) 19:20, 14 April 2009 (UTC)

Origin of name 4.3BSD-Reno[edit]

The article currently states:

It was an interim release during the early development of 4.4BSD, and its use was considered a "gamble", hence the naming after the gambling center of Reno, Nevada.

However I think there is more to the name Reno than that. Please see [2] which currently states:

First there was 4.3. Then there was 4.3 Tahoe. And then there was 4.3 Reno.

Tahoe was so named because one of the things it added was support for the CCI machines which were code named "Tahoe". Reno was called Reno as a joke based on its predecessor Tahoe.

See also [1] -- (talk) 12:05, 28 July 2009 (UTC)

Özalp Babaoğlu[edit]

His name is mentioned only in the "See Also" section -- if he's important to the development, his part should be mentioned inline somewhere, and if not, he shouldn't be in See Also. (talk) 21:38, 3 September 2009 (UTC)

Multics family[edit]

Moved to Talk:Mac OS X: Subsection is Talk:Mac OS X#Multics Family --Tothwolf (talk) 19:23, 16 September 2009 (UTC)

I propose that we re-classify this article under 'OS family: Multics' in the info box, for the reason that Unix is based on Multics. MFNickster (talk) 03:17, 16 September 2009 (UTC)

Please keep the discussion in the one place. It started in Talk:Mac OS X. AlistairMcMillan (talk) 18:41, 16 September 2009 (UTC)

When did csh appear?[edit]

This article says it was in 2BSD in 1978. But according to a published email, 22 Apr 2009 by Kirk McKusick, csh was already in 1BSD in 1977. Can anyone supply anything more authoritative? (I'm working on the csh article and I'd like to get the date right.) Msnicki (talk) 02:39, 6 December 2009 (UTC)

1BSD - 1977 or 1978-03-09?[edit]

According to 1BSD came out on March 9 1978. What is the source for the 1977 date in the article? ★NealMcB★ (talk) 01:13, 25 September 2010 (UTC)

The 1978 date also seems to be used in [QCU] Salus, Peter H. A quarter century of UNIX.
See the previous section - MKM says Bill Joy started work on 1BSD in 1977. Letdorf (talk) 11:25, 27 September 2010 (UTC).

License of Net/1[edit]

The article claims:

This led to Networking Release 1 (Net/1), which was made available to non-licensees of AT&T code and was freely redistributable under the terms of the BSD license. It was released in June 1989.

The license used in net/1 reads (randomly chosen from ping.c):

 * Copyright (c) 1987 Regents of the University of California.
 * All rights reserved.
 * Redistribution and use in source and binary forms are permitted
 * provided that the above copyright notice and this paragraph are
 * duplicated in all such forms and that any documentation,
 * advertising materials, and other materials related to such
 * distribution and use acknowledge that the software was developed
 * by the University of California, Berkeley.  The name of the
 * University may not be used to endorse or promote products derived
 * from this software without specific prior written permission.

(years used in the copyright statement include 1980, 1981, 1983, 1984, 1985, 1986, 1987, 1988) While clearly ancestral to the BSD license, the language is different; notably it does not use the 4-clause format. ~ 10nitro (talk) 01:32, 3 October 2010 (UTC)

BSD for IA-32 in original context[edit]

BSD from Berkeley in '77 to '95 did support IA-32's x86, and few other platforms, but modern BSD, like FreeBSD, supports many other architectures. — Preceding unsigned comment added by (talk) 20:48, 14 May 2013 (UTC)

Significant BSD Descendants?[edit]

Perhaps someone should cleanup & clarify what 'Significant' means here. It seems more than a few listed don't qualify as significant, or if they do it should be explained because its not clear. For example: MicroBSD? That died in 2002 with no fanfare; EkkoBSD died the year it was born and OPNSense is barely a year old fork pfsense with 82 downloads on sourceforge (not the best metric, i know)... --Strangerpete (talk) 12:05, 1 February 2016 (UTC)

Unclear sentence about Windows and MacOS[edit]

The final sentence of the introduction currently reads: “These, in turn, have been incorporated into proprietary operating systems including Microsoft Windows[1] and Apple's macOS and iOS.[2]”

I think the sentence is confusing, especially when both operating systems are put side by side. It looks like it asserts that both Windows and MacOS are derived from BSD, which is not true. While MacOS truly derived from BSD, Windows was created independently, and only later, it reused some BSD code for its network stack. Therefore, I suppose that it is important to emphasize this difference. What about the following reformulation?

“These, in turn, have been used by proprietary operating systems, including Apple's macOS and iOS[2] which derived from them, and Microsoft Windows[1], which used a part of its TCP/IP code.” — Preceding unsigned comment added by MainMa (talkcontribs) 07:17, 13 October 2018 (UTC)

  1. ^ Encyclopaedia of Operating System By Sudhir Kumar, p 838