Talk:OpenBSD

Request for BSD/OpenBSD artwork (for Portal:Free software)

Hi. Wikipedia rules do not allow "fair use" images on non-article pages, so this means that Portal:Free software cannot use logos, mascots, etc. without explicit permission. I have some usable GNU art, and some usable Tux, but I don't want to add those until I have something else to balance out the GNU+Linux tilt that that art would give to the portal. Portal:Free_software will be applying for Featured portal status soon, and one thing is lacks is art.

I saw this page: http://openbsd.org/art4.html but the actual licence/permission terms aren't specified, and looking at Image:Paintedpuffy1000X907px.gif, it seems that commercial use is not allowed, which is an unacceptable restriction for use in non-articles on Wikipedia.

AFAICT, the only acceptable terms are:

• public domain
• revised-BSD-style permissive licence
• GNU FDL

So, can anyone point me to some BSD or OpenBSD pictures which are under one of those licences? Thanks. Gronky 00:45, 10 March 2007 (UTC)

You can use this image:

It isn't an official picture of Puffy, but it has a sufficiently acceptable license for Commons. : )

— Armed Blowfish (talk|mail) 01:28, 10 March 2007 (UTC)

The above picture is from http://art.gnux.info/, which has unofficial pictures of both Puffy and Beastie. — Armed Blowfish (talk|mail) 01:50, 10 March 2007 (UTC)

Great, thanks. Creative Commons by 1.0 is indeed acceptable IIRC. Gronky 02:24, 10 March 2007 (UTC)

Market share percentages add to over 100%

Article says: "The nascent BSD Certification project performed a usage survey which revealed that 32.8% of BSD users (1420 of 4330 respondents) were using OpenBSD,[10] placing it second of the four major BSD variants, behind FreeBSD with 77.0% and ahead of NetBSD with 16.3%.[11]"

32.8 + 77 + 16.3 > 100 —The preceding unsigned comment was added by 67.101.68.207 (talk • contribs).

People were allowed to pick multiple options. NicM 19:56, 24 March 2007 (UTC).

You can read a more detailed explanation at Image:Bsdusage.gif#Explanation_of_chart.  : ) Armed Blowfish (talk|mail) 20:35, 24 March 2007 (UTC)

Is this right

NetBSD (the oldest of the three most popular BSD-based operating systems still active today, with FreeBSD being the third)

I thought FreeBSD was the oldest and then NetBSD and then OpenBSD —The preceding unsigned comment was added by 72.135.112.43 (talk) 03:05, 27 March 2007 (UTC).

NetBSD's first release in April 93 was a few months before FreeBSD's in December. NicM 08:46, 27 March 2007 (UTC).

FreeBSD started out as a bunch of patches, if I remember correctly, which makes this a bit more complex. — Armed Blowfish (talk|mail) 11:40, 27 March 2007 (UTC)

Both of them came from patchsets to 386BSD. NetBSD was the first to make a formal release, which is good enough for me, but it may be best just to strip out "oldest," this article isn't the place for that kind of fact. Nor is it really the place for lists of the most popular BSDs (people occasionally try to add DFLY), although its nice to mention FreeBSD and NetBSD in the first paragraph. Maybe we should just get rid of the whole section in parentheses. NicM 15:58, 27 March 2007 (UTC).

Sounds good to me. History and popularity can be discussed in more detail elsewhere. For readers, the parenthetical version of the lead can be found here. — Armed Blowfish (mail) 01:58, 10 April 2007 (UTC)

A lot of crap

Just sorta added to it, but I think perhaps there should be a sub-page made, like the archives and todo which holds the assessment stuff, just having a link to it instead of the many templates repeating the same thing. All the stuff that mentions FA should probably be put there if it were to be done, but that ould probably mess up those templates' inner workings, how they add to various lists and whatnot, so I won't do any such thing myself. But I will leave this comment here, to fester in the minds of any who read it. 74.13.60.113 07:08, 23 April 2007 (UTC)

At least the small ones are now out of the way a bit, and if we do something special we have extra maintenance and people adding them back in the wrong place. NicM 08:05, 23 April 2007 (UTC).

"At present"

It's probably best not to use phrases like that, in case there isn't someone maintaining the page, if it does fall out of step with what presently is happening, would it not be best to have it state a date? Not that the people working on the article will disappear, but if they did, it would probably be for the best if everything was linked to a time, no? 74.13.57.76 23:23, 1 September 2007 (UTC)

Absolutely right. The preferred form is "As of (year)". I've fixed this here. Feezo (Talk) 00:27, 2 September 2007 (UTC)

I disagree, it just invites people to change it every year rather than just fixing it when something actually changes. But since it is policy, who am I to argue ;-). NicM 11:25, 2 September 2007 (UTC).

I think the idea behind the guideline is that, if left unchanged, (if people forget, or if it's printed out, or republished) the text will continue to be correct in years to come, even if it starts to sound a bit dated. The ideal solution IMHO would be to find a reference (posting by Theo or someone) that would confirm the date. We could then use "as of 20XX" with less risk of people robotically incrementing the year. Feezo (Talk) 20:04, 2 September 2007 (UTC)

How about just removing the sentence entirely? Even without it it is pretty clear the removal hasn't been undone. NicM 11:29, 3 September 2007 (UTC).

UNIX versus Unix-like

Technically OpenBSD is not a UNIX, it's a BSD, and while BSDs and UNIXes share code, they are not the same. In both conventions and in the actual implementation of many tools and systems, the UNIXes are different from OpenBSD. While OpenBSD has imported a UNIX compiler, awk and many other tools, that doesn't make them a UNIX. To do that at this point would require a big pile of money, as Apple has thrown around in order to get Mac OS X's most recent release certified. I have because of this reasoning, reverted the change from Unix-like to Unix. 74.13.43.249 00:52, 2 November 2007 (UTC)

Agreed. That's how the article stated the situation until just a few hours before your edit. An anonymous user made this change without explaining why it should be so. (A similar change was made to the FreeBSD article.) There's also the small matter of the court battle in the 1990s, which settlement resulted in the BSDs not being able to claim to be UNIX, and which is how UNIX-like came about. Mindmatrix 14:12, 2 November 2007 (UTC)

Timeline/releases for OpenBSD

Think we can compile a list of OpenBSD releases similar to this table?

Colour	Meaning
Red	Old release; not supported
Yellow	Old release; still supported
Green	Current release
Blue	Future release

Version	Code name	Testing name	Release date	Supported until	Features and Changes
4.10	Warty Warthog	Sounder	2004-10-20	2006-04-30	Initial release; ShipIt
5.04	Hoary Hedgehog	Array	2005-04-08	2006-10-31	Update Manager; Upgrade Notifier; readahead; grepmap; laptop suspend, hibernate and standby; dynamic frequency scaling; Ubuntu hardware database; Kickstart; installation from USB devices; UTF-8 by default; APT authentication
5.10	Breezy Badger	Colony	2005-10-13	2007-04-13	Usplash (graphical boot sequence); "Add/Remove..." application tool; easy language selector; logical volume management support; full Hewlett-Packard printer support; OEM installer support; Launchpad integration
6.06 LTS	Dapper Drake	Flight	2006-06-01	2009-06 (desktops)	Long Term Support (LTS) release; Live CD and Install CD merged onto one disc; Ubiquity graphical installer on Live CD; Usplash on shutdowns; Network Manager for easy switching of multiple wired and wireless connections; 'Humanlooks' theme implemented using Tango guidelines, based on Clearlooks and featuring orange colours instead of brown; LAMP installation option; installation to USB devices; GDebi graphical installer for package files
				2011-06 (servers)
6.10	Edgy Eft	Knot	2006-10-26	2008-04	Ubuntu 'Human' theme heavily modified; Upstart init daemon; automated crash reports (Apport); Tomboy notetaking application; F-spot photo manager; EasyUbuntu merges into Ubuntu via meta-package installs and features
7.04	Feisty Fawn	Herd	2007-04-19	2008-10	Migration assistant; Kernel-based Virtual Machine support; easy codec and restricted drivers installation; Compiz desktop effects; Wi-Fi Protected Access support; PowerPC support dropped; Sudoku and chess games added; disk usage analyser (baobab) added; GNOME Control Center; Zeroconf for many devices
7.10	Gutsy Gibbon	Tribe	2007-10-18	2009-04	Compiz Fusion by default; AppArmor security framework; fast desktop search; fast user switching; some plug-ins for Mozilla Firefox now handled by APT (Ubufox); graphical configuration tool for X.org; a revamped printing system with PDF printing by default; full NTFS support (read/write) via NTFS-3G
8.04 LTS	Hardy Heron	Alpha	2008-04-24	2011-04 (desktops)	Long Term Support (LTS) release;; Better Tango compliance; compiz usability improvements; tracker integration;; Brasero disk burner, Transmission BitTorrent client and Vinagre VNC client by default; PulseAudio by default
				2013-04 (servers)
8.10	Intrepid Ibex	Alpha	2008-10-30	2010-04	Complete interface redesign; improvements to mobile computing and desktop scalability; increased flexibility for Internet connectivity

Altonbr (talk) 03:31, 25 February 2008 (UTC)

There is an entire article of that, along with other historical notes regarding OpenBSD. I don't see what this idea has which shows superiority over the already existing article. 74.13.29.9 (talk) 17:19, 25 February 2008 (UTC)

If you want to improve the existing OpenBSD timeline article, that would be really great, but including such a table in the main article would be a mistake. NicM (talk) 07:20, 27 February 2008 (UTC).

Which bootloader ?

Hi. In the page Comparison of boot loaders, the column for OpenBSD (the only BSD of the table) is nearly empty. Could some people fill it with bootloaders able to boot OpenBSD ? JeromeJerome (talk) 13:51, 13 June 2008 (UTC)

Perfection

While I'll not agree with the subject's title, it seems pretty stable. I am just sort of wondering, "out loud," I think, but the article here seems to tinge a little too close to OpenBSD's perspective to be the top of the line, doesn't it? Could this not be made a little more distanced from the project's perspective? Are there no other information sources which can be used? Hell, I am not saying that the article needs a review, but after reading it, it seemed to be something that would be in an OpenBSD book or the official website. 74.13.48.202 (talk) 11:45, 12 July 2008 (UTC)

Please be more specific, the article is about OpenBSD so it cannot cover viewpoints which are not connected with the project, but I don't see any areas where other viewpoints are connected but are ignored. We could perhaps include a paragraph on Linus' views (and the OpenBSD responses) in the security section, but as the article was written before his comments, it is unfair to criticise it for not including them. I'm not sure exactly where to fit this. Perhaps something like (in the last paragraph of the security section):

The project has a policy of continually auditing code for problems, work developer Marc Espie has described as "never finished ... more a question of process than of a specific bug being hunted." Linux kernel creator Linux Torvalds has expressed the view that a security problem is no more or less important than any bug and criticised the OpenBSD community's policy: "[T]hey make such a big deal about concentrating on security that they pretty much admit that nothing else matters to them." OpenBSD developers dismissed these comments as the words of a "troll"(Marco Peereboom msg-id <20080716190300.GK5324@peereboom.us>) with Marc Espie commenting: "It's a totally misinformed opinion ... [Fix normal bugs is] exactly what the OpenBSD project do, all the time." (Marc Espie <20080716231959.GB6918@lain.home>).

Thoughts? Obviously cites are to be added. NicM (talk) 18:06, 20 July 2008 (UTC).

I'm going to add this anyway and we'll go from there. NicM (talk) 18:30, 20 July 2008 (UTC).

Alright, done, with some modifications (better quotes mainly), not sure about the positioning. It might be better a bit higher in the section, but maybe better to get the discussion of the actual attitude to security and the features done before mentioning criticism of it. Hmm. NicM (talk) 19:12, 20 July 2008 (UTC).

Agree. The section "Open Source and Open Documentation" sounds like a teenager's account of the early history of BSD designed to sexually gratify those involved with the project. This article needs to be more subjective.

Suggestion for removing the "biased" words/phrases like "chastised" and "castigated" and "pointed out". The first two hint that the actions taken were unfair (with no backing), while the last one suggests that the subject of the statement is in fact correct in their argumentation (when they may not be). --198.151.13.8 (talk) 19:32, 14 July 2008 (UTC)

I don't agree, I think you are seeing bias where none exists. These words are neutral and used neutrally - castigated and chastised merely mean criticized, there is no overtone of correctness or incorrectness that I am aware of and the dictionary confirms this. The section carefully includes the opposing view (describing it as "disagreements", with no judgement) and I do not see where it implies the OpenBSD viewpoint on binary drivers is more correct. Making crude sexual references is not helpful. NicM (talk) 18:06, 20 July 2008 (UTC).

CORRECTION: "Chastised" does not appear in that section. I had a mental misfire there. —Preceding unsigned comment added by 198.151.13.8 (talk) 19:36, 14 July 2008 (UTC)

Hang on a tick, isn't the masterbating monkeys comment exactly what this article needs, in order to balance the perspective more? 74.13.28.55 (talk) 17:49, 19 July 2008 (UTC)

No, that comment is not criticism it is just abuse; however, some mention of Linus' view may be worth making. NicM (talk) 18:12, 20 July 2008 (UTC).

The Masturbating Monkeys

"Linus Torvalds has expressed the view that a security problem is no more or less important than any bug" - Actually, this is not exactly what he said. He only said security for Linux isn't less important than normal bugs, which is obviously the opposite of what he really meant in his rage against OpenBSD. Still, I'm not sure how we should encrypt Linus's stand on security bugs vs. normal bugs, which implies that security bugs aren't normal but somehow abnormal... --84.250.188.136 (talk) 14:22, 22 July 2008 (UTC)

Actually, what he said was that normal bugs are more important than security bugs ("In fact, all the boring normal bugs are _way_ more important, just because there's a lot more of them.") We should probably fix that. NicM (talk) 17:35, 22 July 2008 (UTC).

Proposal to remove date-autoformatting

Dear fellow contributors

MOSNUM no longer encourages date autoformatting, having evolved over the past year or so from the mandatory to the optional after much discussion there and elsewhere of the disadvantages of the system. Related to this, MOSNUM prescribes rules for the raw formatting, irrespective of whether or not dates are autoformatted. MOSLINK and CONTEXT are consistent with this.

There are at least six disadvantages in using date-autoformatting, which I've capped here:

Disadvantages of date-autoformatting

• (1) In-house only

• (a) It works only for the WP "elite".
• (b) To our readers out there, it displays all-too-common inconsistencies in raw formatting in bright-blue underlined text, yet conceals them from WPians who are logged in and have chosen preferences.
• (c) It causes visitors to query why dates are bright-blue and underlined.

• (2) Avoids what are merely trivial differences

• (a) It is trivial whether the order is day–month or month–day. It is more trivial than color/colour and realise/realize, yet our consistency-within-article policy on spelling (WP:ENGVAR) has worked very well. English-speakers readily recognise both date formats; all dates after our signatures are international, and no one objects.

• (3) Colour-clutter: the bright-blue underlining of all dates

• (a) It dilutes the impact of high-value links.
• (b) It makes the text slightly harder to read.
• (c) It doesn't improve the appearance of the page.

• (4) Typos and misunderstood coding

• (a) There's a disappointing error-rate in keying in the auto-function; not bracketing the year, and enclosing the whole date in one set of brackets, are examples.
• (b) Once autoformatting is removed, mixtures of US and international formats are revealed in display mode, where they are much easier for WPians to pick up than in edit mode; so is the use of the wrong format in country-related articles.
• (c) Many WPians don't understand date-autoformatting—in particular, how if differs from ordinary linking; often it's applied simply because it's part of the furniture.

• (5) Edit-mode clutter

• (a) It's more work to enter an autoformatted date, and it doesn't make the edit-mode text any easier to read for subsequent editors.

• (6) Limited application

• (a) It's incompatible with date ranges ("January 3–9, 1998", or "3–9 January 1998", and "February–April 2006") and slashed dates ("the night of May 21/22", or "... 21/22 May").
• (b) By policy, we avoid date autoformatting in such places as quotations; the removal of autoformatting avoids this inconsistency.

Removal has generally been met with positive responses by editors. Does anyone object if I remove it from the main text (using a script) in a few days’ time on a trial basis? The original input formatting would be seen by all WPians, not just the huge number of visitors; it would be plain, unobtrusive text, which would give greater prominence to the high-value links. Tony (talk) 06:42, 24 July 2008 (UTC)

Will this unlink the years too? I'm not happy if it turns 1 July 2005 into 1 July 2005. Also, we should lose stuff like October 1995 as well. If it will unlink too this is probably a good step; I'd probably use US style for this article since it is already used in the references. NicM (talk) 20:15, 24 July 2008 (UTC).

Bad implication: "security problems are less important than other bugs"

The last paragraph of the "Security and code auditing" section starts off as:

"Linux kernel creator Linus Torvalds has expressed the view that—due to their smaller numbers—security problems are less important than other bugs ("all the boring normal bugs are _way_ more important, just because there's a lot more of them"[40])..."

That quote of Torvalds does not imply the view that "security problems are less important than other bugs". There are only two ways to interpret Torvalds's quote. First, he could mean that every one of the normal bugs is way more important, which is ridiculous, of course. That interpretation, however, is the only one that supports the implication given. The other interpretation, that the class of normal bugs is more important than the class of security bugs, is likelier what Torvalds meant. Of course, this does not imply the view that security problems are less important than other bugs, particularly since he explains that there are _more_ normal ones. That is, it does not follow from this likelier interpretation that Torvalds thinks a security bug matters less than some other bug.

The paragraph should be changed to something logical. Freed42 (talk) 16:29, 31 July 2008 (UTC)

Huh? 74.13.56.168 (talk) 20:25, 31 July 2008 (UTC)

The phrase, "security problems are less important than other bugs", misrepresents the Torvalds quote. A correct phrase would be, "the set of all security problems is less important than the set of all other bugs". See the difference? - Freed42 (talk) 21:48, 31 July 2008 (UTC)

Nope. 74.13.56.168 (talk) 13:17, 1 August 2008 (UTC)

OK, just consider it another way. If my "view" is, "Security problems are less important than other bugs", then someone could logically deduce that I mean any security bug is less important than any other bug. E.g., a serious security bug matters less than the most trivial typo. If you do not understand this, then could you at least explain using something other than one-word sentences how "Security problems are less important than other bugs" does not imply "Any security bug is less important than any other bug"? - Freed42 (talk) 13:41, 1 August 2008 (UTC)

I must say that I agree with your interpretation of Linus' quote, insofar as we can assert anything (as you say, this interpretation is just more likely what Linus meant). I realise the quote from him is poor, but thus far I haven't found a better one.

Saying that, the article seems clear enough to me, the plural "security problems" means the set (or class, if you like) of security problems and likewise "normal bugs" means the set of normal bugs; if it said "any security problem" then I would agree it is incorrect. However, if you think it is unclear feel free to tweak the language. If you could find a different quote from Linus showing his view more clearly it would be nice, but please bear in find firstly that this article is not about Linus' view, it is about his criticism of OpenBSD (so we should not dwell on the substance of his view on security problems vs normal bugs, just express it succinctly and with a short illustrative quote); and secondly that discussion of Linus' point of view should be balanced with discussion of response from the OpenBSD developers. NicM (talk) 20:44, 1 August 2008 (UTC).

OK, I think his quote is good enough, especially since he refers to quantity of bugs. Therefore, I will add "class of" to the ambiguous inference preceding the quote. - Freed42 (talk) 22:01, 1 August 2008 (UTC)