Talk:MacOS/Archive 3

Opening sentence
The opening sentence currently reads as follows:
 * "Mac OS X is the latest version of the Macintosh operating system, and is designed and developed by Apple Computer to run on their Macintosh line of personal computers."

I believe this is ambiguous and possibly (though not intentially) misleading because it implies that Mac OS X was developed from previous versions of the Mac OS. OS X certainly is in the Mac OS family (and has replaced OS 9 as the operating system that Apple ships on its Macs), but one could just as easily say that it is a flavor of UNIX, or version of BSD, or a successor to NeXTStep. It has become very uncommon for Apple &mdash; or anyone, for that matter &mdash; to call the OS "Mac OS" with out the "X" part. I think something like:
 * "Mac OS X is an operating system designed and developed by Apple Computer to run on modern PowerPC- and Intel-based Macintosh computers."

would be better (as, for one, it provides better context for its role as an operating system by linking to the appropriate article). However, it seems like the current wording has been used for quite a while, so I decided not to change it yet and wanted to elicit comments about it. btm 10:49, 30 October 2005 (UTC)


 * I think your suggestion for the lead paragraph is logical, but your rewrite is not good either. This would be better:
 * "Mac OS X is an operating system designed and developed by Apple Computer for their current line of PowerPC Macintosh personal computers...."

It would be wrong to say Intel just yet, because thats only for developers, and is not for the public. So it would be wrong to put it in an encyclopedia. — Wackymacs 11:22, 30 October 2005 (UTC)


 * I agree with the Intel objection, and I don't think I would have included it in the main article because the wording implies that Apple is shipping Intel-based Macs, but this property of running on these two architectures is an important property of the system and deserves to be more prominently featured in the article (especially because this transition is such an important part of Apple's and the OS's future). Perhaps not in the opening sentence, but somewhere in the first or second paragraph of the article. btm 21:36, 30 October 2005 (UTC)
 * Replaced old line with: Mac OS X is an operating system designed and developed by Apple Computer for use on their current line of Macintosh computers. Please see diff for changes. TDS (talk • contribs) 04:54, 1 November 2005 (UTC)

Removed revision
Regarding the text:
 * A further complication is that, many critics also question the frequency of new versions in comparisons to Microsoft Windows. While this may seem like an advantage, some critics suggest some Mac OS X updates are more like Windows service packs, which are free [....]

I believe this was a Rob Enderle complaint, and shouldn't be hard to find a source. There HAS been grumbling about what IS and ISN'T an upgrade worth paying for. 10.1 was the only "freebie," but I think that was mainly because 10.0 AND 10.1 were released in 2001. 10.2 was August 2002, 10.3 was late 2003, and 10.4 was April 2005 - much more frequent, in some respects, to Windows, BUT there have been Media Center editions 2003/2004/2005 - anyone know what you have to pay if you want to update THAT edition of XP?

The way Apple has changed things since OS X is that "point" releases (10.2, 10.3, 10.4, and later 10.5) are NEW versions, that have to be paid for (like 8 was for System 7 users, OS 9 was for OS 8 users, etc.) and the "point.point" (10.4.1, 10.4.2, 10.4.3) releases are the "service packs" - compare this to 7.5, 7.5.3, 7.5.5, 7.6, 7.6.1....

--JohnDBuell | Talk 03:30, 5 November 2005 (UTC)


 * Hey there! I would have to say that Mac OS X releases are very different from service backs being that many, many things are changed in relation to the core operating system and many new major features. Service packs are more 'housecleaning' so to speak i.e they are fixing bugs, and upgrading security. Mac OS X releases change fundamentals things such as adding metadata, and adding the Quartz Compositor. And as well the frequency of updates is getting slower as more-or-less Apple was playing catchup. The 10.x can be seen as merely a marketing scheme, as Apple does not want to lose the Mac OS X designation =) Thanks, TDS (talk • contribs) 16:00, 5 November 2005 (UTC)


 * I think I said all of this, just using more words. I personally didn't think there was that much of a change from 10.2 to 10.3 (except for the brushed metal look everywhere), but the changes from 10.1 to 10.2 and 10.3 to 10.4 were major and obvious. The point was to describe where the ideas in the removed text came from, and the criticism wasn't mine, it was a "technology analyst" (or whatever Enderle's calling himself these days). --JohnDBuell | Talk 06:50, 7 November 2005 (UTC)


 * I personally treat Mac OS X releases as not 10.4.4. I don;t think that there will be an OS 11. I think the os is more like Mac OS X Version 4.4. The dot update confuses windows users into thinking that they are buying service packs, since it seems like one-tenth of a new os.

Ccool2ax 14:36, 9 February 2006 (UTC)

Pricing structure
Dedicating over 200 words to a debate over the fairness of the upgrade price in an encyclopedia article seems crazy. I understand that some people think Apple charges too much for OS upgrades (I agree!), but does anyone really think this belongs in an encyclopedia entry? Can anyone provide a compelling argument to leave that paragraph in there?
 * I agree, this does not really belong in an encyclopedia. Ramallite (talk) 04:36, 8 November 2005 (UTC)
 * I disagree. It is a valid criticism that is held by many people, and I see no reason as to why it is not encylopedic. While I personally do not agree with the criticism, I see no reason to remove it. Earlier on the talk page there is a cited expert who states that exact claim. TDS (talk • contribs) 04:58, 8 November 2005 (UTC)
 * I agree that it's valid criticism, but I also agree that it should be removed. I musses up what's otherwise a decent, concise article about an operating system with a petty point. If some people feel really strongly it should stay, then it should at least be shortened to one or two sentences. (Some people say it's too expensive. Others disagree.) 66.117.145.108 05:10, 8 November 2005 (UTC)
 * Placing criticisms of an operating system in an article about said operating system is not 'petty.' It is showing all sides of the manner, which an enclyopedia ought to do. TDS (talk • contribs) 05:56, 8 November 2005 (UTC)
 * It is fair to keep it in, but it is comparing it to Microsoft Windows, yet it doesn't mention that the initial retail price of XP Pro is a lot more expensive than Mac OS X is at retail price. Neither does it mention the cost of Microsoft Windows upgrades. — Wackymacs 08:17, 8 November 2005 (UTC)
 * =D In all fairness I agree with Wackymacs, we need more on how much Windows is, and how much Vista releases will be. TDS (talk • contribs) 15:12, 8 November 2005 (UTC)
 * It's likely you won't see anything on Vista pricing until MUCH closer to its release (whenever THAT turns out to be). But it's pretty commonly known that you can get XP Home (for computers without Windows), XP Home (upgrading prior version of Windows to current), XP Pro (for computers without Windows, higher than Home), XP Pro (upgrading versions, including Home, to Pro). The other editions (Tablet PC, Media Center edition, etc) are OEM only and not available to end users. --JohnDBuell | Talk 18:42, 11 November 2005 (UTC)
 * These complaints about the price (and the opposite arguments about Windows pricing) are original research: just the opinions of a Wikipedia contributor. Where is the source for the opinion that the OS is too expensive? See No_original_research An individual's opinions do not belong in the article. (There are lots of things I dislike about OS X, but I understand that this article is not the place for such discussions.) 68.125.62.126 00:02, 28 November 2005 (UTC)
 * No, it isn't. These are opinions that have been put forward by "technology analysts," NOT original opinions of Wikipedians (and similar grumbling has taken place on many Apple-related newsgroups, including some that Apple hosts). --JohnDBuell 01:27, 30 November 2005 (UTC)
 * Do you have a citation? 70.162.56.52 16:49, 23 April 2006 (UTC)
 * From the Apple’s EULA at : “Updates: If an Apple Software update completely replaces (full install) a previously licensed version of the Apple Software, you may not use both versions of the Apple Software at the same time nor may you transfer them separately.” That suggests that it is indeed a upgrade license and not a new license. The pricing argument, as written, is a bit sophomoric—and I can't rewrite it because I think it doesn't belong here. The pricing argument appears to be a holdover from a section chock-full of subjective criticisms of Mac OS X; the other arguments were dealt with accordingly. I propose renaming the "Pricing debate" section to "Pricing" and deleting the second paragraph. This discussion appears to have previously ended with only some superficial edits; barring a citation or substantial improvement to the copy, I'll be back with the cleaver. 70.162.56.52 16:49, 23 April 2006 (UTC)
 * I agree. Leave the facts in, but take the opinions out. The entire second paragraph relating OS X pricing to Microsoft's pricing is unnecessary, and quite frankly, doesn't make much sense. Cleave away! EricNau 00:57, 5 June 2006 (UTC)

Mac OSX as a distribution of Darwin OS
Mac OSX cannot be considered an operating sistem on its own. Darwin has a command-line interface, it handles all the hardware management, interprocess management, and protocol management of Mac OS X. This means Darwin is an OS. And, by contrast, OSX is just a GUI, based on proprietary libraries, plus a collection of application. In other words OSX is just a commercial distribution of the Darwin OS, as it occurs(E.G) for Red Hat and linux. Luxiake 18:52, 25 November 2005 (UTC)
 * In a way, that is true - but there are a lot of components of Mac OS X which can be considered part of the OS as a whole. I wouldn't consider Quartz, Cocoa, Carbon, Classic, Application Services, Quicktime etc. to be parts of Darwin, but they are a large part of what makes Mac OS X what it is, and make it different from other BSDs. These components are not essential, but they are not applications, utilities or tools either. Saying that it's a "commercial distro of Darwin" is true, but it's only part of the picture, which IMHO would make it a misleading claim. MFNickster 18:59, 25 November 2005 (UTC)

especially since darwin is really a stripped down version of mac os x... both are OSes, one is just a superset of the other...

--Yoasif 19:42, 25 November 2005 (UTC)

All the libraries and application of OSX are nonessential component of the Darwin OS. If you are not persuaded of Darwin being an OS please refer to the Darwin discussion page. To consider OSX as a distribution of Darwin OS is a natural consequence of consider Darwin an OS. The proof of that is that all the free software avaliable for Darwin OS can work both on GNU-Darwin and OSX Luxiake 20:15, 25 November 2005 (UTC)
 * Darwin is an OS, and Mac OS X is also an OS based on Darwin. As I pointed out above, though, Mac OS X is much more than a commercial distribution of Darwin. You can't run Mac applications on Darwin alone, and for that reason Mac OS X is a full-fledged OS at the very least in terms of being a target platform for development. There is gray area here, but seeing as how tightly bound (in technical and marketing terms) Mac OS X is with Macintosh computers and only Macintosh computers, I would say that calling it a "Darwin distribution" is not just misleading, it is factually incorrect. Specifically, Mac OS X is not only an operating system in its own right, it is the Macintosh operating system. MFNickster 21:18, 25 November 2005 (UTC)

What is incorrect is claiming OSX to be an operating system on its own. That's what Apple wants we to know. Actually, OSX without Darwin is no OS at all. On applying OSX(proprietary) on Darwin(opensource), Apple is try to prevent full compatibility with the Darwin OS, but isn't creating a separate OS. As there are other distributions of Darwin and they are growing day by day, I think we will see a Linux-like scenario in few years. I agree with you that calling OSX a distribution today sounds weird because Apple has a monopoly over the Darwin distributions scene. But the sole existence of the GNU-Darwin distribution today is per se a valid reason to call OSX a distribution of Darwin. OSX is not the only Apple computers operating system, there are a lot of distro out there, feel free to try UBUNTU, GNU-Darwin, Gentoo, Free-BSD, Open-BSD, or whatever Luxiake 22:07, 25 November 2005 (UTC)
 * You are technically correct, sort of, but this is an encyclopedia. The aim of the article is to give facts about Mac OS X that everyone at all levels of knowledge can easily understand, and the truth is that it is harder to understand if we say "a Darwin distribution". It is commercially advertised, promoted and described as an operating system, and is mostly an operating system. Darwin is only one part of Mac OS X, as described earlier by another user, there are many many parts of OS X that make it up to become an operating system. Just because the OS is based on something else doesn't mean it is a distribution of it. The majority of people understand it generally as an operating system, and it is more that than anything else. We don't need to confuse people. — Wackymacs 22:25, 25 November 2005 (UTC)
 * Well, I don't want to split hairs over what constitutes an OS, as has been done repeatedly on the GNU/Linux issue. But if you're going to limit the definition in that way, why not say that Darwin is actually an "XNU distribution"? MFNickster 22:30, 25 November 2005 (UTC)

Being wikipedia an encyclopedia and not simply an advertisement medium of apple computer as you mac-maniacs are triyng to do, it's goal is to give people a correct and unbiased point of view over the issue. And you have to admit that the XNU=Kernel, Darwin=OS, and OSX=distro, is the sole point of view on this matter. The goal of wikipedia is to show the truth, and not to repeat the advertisement and promotion slogans of apple computer. If common people aren't understanding the meaning of the word "distro" we are here to explain it, and not to tell them the opinion of Apple computer about this issue. The Mac OSX article has being a monopoly of mac maniacs over the years, but now it's time to state here a neutral point of view. @MFNikster: if you believe so, then go to the Darwin page and start a discussion about the issue.Luxiake 22:55, 25 November 2005 (UTC)
 * Far from being NPOV, you have just explicitly stated your own POV. I'm sorry to hear you're not interested in compromise. MFNickster 23:01, 25 November 2005 (UTC)
 * Mac maniacs aren't like George Bush: we don't repeat whatever we say until evryone believes it. We understand that Mac OS is Unix-based, but it is more than a distribution. And thats not just me trying to claim that OSX is advanced. OSX is more than a GUI or distro. Its a hybrid of UNIX, NeXT, and Apple technology. Ccool2ax 14:41, 9 February 2006 (UTC)

I'm certainly not an expert on software theory, but I think a good practical test of this would be whether or not most application software written for Mac OS X can run on free distrubutions like GNU-Darwin. If/when I'm able to install GNU-Darwin on my computer and get it to run Final Cut Pro I'll vote for OS X to be considered a commercial distribution of the open source Darwin OS. But if those applications cannot be run on free distributions of Darwin then I'd say that for practicle purposes at least, OS X is a separate OS. Blackcats 11:58, 27 November 2005 (UTC)

If final cut pro uses Apple's proprietary libraries then the software is not likely to run. This issue has to be added in the criticism section. By contrast, all the packages compiled for Darwin, that are mainly free-software, works both on GNU-Darwin and Mac OSX. GNU-Darwin has made pressure over the year on opensourcing, Cocoa, Carbon etc.. If this will occur, then Mac OSX will be 100% free, that means no profit for Apple. I don't think it's gonna happen in the next months. Luxiake 12:11, 27 November 2005 (UTC)


 * I wouldn't say that would mean no profit for Apple. I think a lot of people would still want the look and feel of the OSX, and a lot of people would also want to keep buying Apple's own hardware (even when it starts using the same x86 proccessors as everyone else).  But you're right that it would reduce sales of OSX if everyone could get a fully compatible OS for free (or for the nomial charge of paying for someone to burn you some CDs and ship them to you).


 * Assuming that Apple doesn't opensource their libraries, then it will up to the developers of GNU-Darwin (or any other free-software developers who step in) to develop compatible libraries. This would be somewhat analogous to ReactOS, which is working to develop and OS that's completely compatible with Windows.  But, though I'm not an expert, I would think that the job of the GNU-Darwin people would be a whole lot easier than that of the RactOS people - since the ReactOS team has to design not just libraries but a whole new kernel, whereas the GNU-D team would just have to re-write a couple libraries to acheive their goal, since the kernel code is open source.  There'd also be the issue of file sytem compatability, but that probably wouldn't be too hard to work around, since I think OSX also supports an open source Unix FS or two, and proprietery distributions of the HFS are already sold to Windows users by companies like MacDrive, so if nothing else, they could probably be convinced to sell an adaptation of that software to GNU-Darwin users.


 * I speculate though that there may be some political reasons why the GNU-Darwin team hasn't been working towards full compatability (assuming they haven't yet been). One might be that they'd be afraid of pissing off Apple to the point where they might turn away from using an open source kernel.  The other might be a fear that if they did successfully develop and distribute a free-software OS distribution that was completely compatible with OSX then this might lead to a loss in profitability to the point where Apple would no longer developed and sold a MacOS, and this would give Microsoft a more complete monopoly in the commercial OS market.  But of course since the GNU-Darwin OS is completely open source, any qualified and motivated team can fork off that project and work towards complete compatibility....   Blackcats 01:25, 28 November 2005 (UTC)


 * I'm not sure how GNU-Darwin feels about X Window applications (which would be the free equivalent of Aqua apps), but as far as the kernel goes, chances are pretty remote that Apple would switch to a closed-source kernel. The main reason Darwin is open-source in the first place is because it's derived from open-source underpinnings (BSD and Mach). I'm sure Apple would love to have a fully proprietary platform, but they would have to write a new kernel/server/driver base and give up all the things Darwin has made possible, along with backward compatibility for BSD and Mac applications. MFNickster 03:18, 28 November 2005 (UTC)


 * For some part I agree with Blackats. But you have to keep in mind that writing graphic or audio libraries from zero, it is a huge work, but at least it's possible. But writing libraries compatible with closed source libraries is nearly impossible, or at least requires many years of testing, debugging, it is the main difficulty that the WINE project is facing. But aside from this, I found in some interviews, that even the same creator of GNU-Darwin, while critizing this policy, implicitly acknowledge a right of Apple in retaing the closed source libraries. Plus, I find that the free-software universe (meaning Linux users, BSD etc..) it's not so unhappy with Apple, and I suppose that this is because they think the work of Apple, while trying to dethrone Microsoft, will ultimately help the cause of the free-software movement. I think this is a huge mistake. In replacing Microsoft with Apple, the free-software movement, will face a much stronger enemy, that offers a reliable OS, a very appealing brand-name, ultimately it offers an alternative philosophy, that is no free-software, no closed source philosophy, it's mac philosophy, that is based on digital consumerism, digital un-awareness, digital egoism (I-tunes,I-book, I,I,I) as opposed to digital solidarity among users, digital awareness(everyone is pushed to give some sort help of to the cause) that rely at the base of the free-software movement, and even Wikipedia.

Luxiake 09:53, 28 November 2005 (UTC)


 * You may be right about that, and making the libraries and such compatible could well be more work than re-writing the kernel. I don't know since I haven't done either or spent much time researching this.  I think another thing that may have slowed development of a fully compatible OS was that until next year, all the OSX applications that have been distributed have been written for PowerPC, so open source developers would likely think "what's the point in developing a fully compatible OS if people have to run it on hardware that they buy from Apple and which all come bundled with the OSX?"  Of course you could probably put together some generic computer with a PPC processor, but they're certainly much less readily available than the x86 ones.  But now that the Mac OS and software are gonna be running on the same processor, I'd think there'd be more demand for a compatible OS since it's a lot easier to but the generic x86 hardware with no OS installed.  Blackcats 02:09, 29 November 2005 (UTC)


 * All I have to say is - has this kind of debate raged on the NeXTStep/OpenStep pages? :P You could NOT separate the GUI from the underlying kernel in NeXTStep, but OpenStep was the GUI sold WITHOUT the kernel (and I forget how it was sold, but it wasn't as an OS). The Darwin kernel can NOW be distributed without the GUI, but it isn't SOLD without the GUI. Mac OS X is a commercial package of both, and in my opinion, is a standalone OS. Sure, you can slap X11 on top of Darwin, but you still can't have nearly the functionality of OS X. I'd also imagine that an installation of Tiger, run as a CLI only, would differ quite a bit from an installation of Darwin, having all of Apple's proprietary non-GUI components removed. Remember that the Open Sourced Darwin STILL has a lot of pieces missing, because of copyrights and patents....


 * Further, yes Apple is obviously a commercial enterprise, and has to live on its revenues to survive. That's why it'll never "give away" everything, and why things like FairPlay licensing or even OS X x86 licensing will never happen. I don't disagree that replacing MS with another closed source system is a bad idea. But you have to agree that it's a step in the right direction to have companies willing to offer part of their systems (Darwin Streaming Server, Darwin itself, Bonjour source for Apple, OpenOffice.org and OpenSolaris for Sun, among other projects). Plus you're missing a major point on the philosophies of avowed Linux geeks and Mac users. Mac users WANT their computers to "just work." They don't want to have to tinker, they don't want to have to recompile, they don't want to have to add/change/modify source code. They want it to do the tasks that they throw at it, even if that means using the programs Apple includes (and you're always welcome to remove Apple's default installed programs and throw your own in, even the free/open source ones). I gave up having a Linux box because I was always having to fine tune the **** thing. There will always be tinkerers, just like there will always be guys trying to fine tune the heck out of their cars, because they can. And I frankly feel that REAL true "user friendliness" is years away, and in directions other than the tired old (25 year plus!) mouse+keyboard+menus GUI, but that's a discussion for somewhere else. --JohnDBuell 01:41, 30 November 2005 (UTC)

Just so we are all clear. Mac OS X and Darwin are operating systems. OpenDarwin is an operating system that is based on Darwin. And GNU-Darwin is just a bunch of applications that are packaged to run on Darwin. GNU-Darwin is not an operating system. AlistairMcMillan 02:31, 30 November 2005 (UTC)

@AlistarMcMillan: How did you dare! I tried to discuss, and you changed the whole things, you erased the links to GNU-Darwin. I'm not even like Michael Love, that is a peaceul and discreet man, I'm evil like hell. Now, you restore the whole things, or you will see the Armageddon. Restore everything about GNU-Darwin first. Then I'll discuss about what is an OS and what is not an OS. Luxiake 18:21, 30 November 2005 (UTC)


 * I'll take (B) Armageddon please. AlistairMcMillan 18:26, 30 November 2005 (UTC)

i tend to agree with AlistairMcMillan, so go ahead with the armageddon... i'll be sure to correct and revert as needed. --Yoasif 18:32, 30 November 2005 (UTC)

OSX w/o Darwin is no OS at all and everybody knows that. FreeBsd was opensource, Mach was opensource too. Apple did some cut&paste of both and here we have Darwin. Then applied the aqua GUI on Darwin and said to the dull massess of mac-maniacs, like you: Here's MAC OSX! And you started clapping with big enthusiasm. But I'm very tired of monkey talks like this, and to make effort for the truth to come out. My Armageddon is my silence, keep your MacOSX like it is and be happy with it, and please tell the other dull people of the world how much Steve Jobs is clever. I will retire to my previous ignorance of the whole matter. Bye bye Luxiake 22:03, 30 November 2005 (UTC)


 * considering that mac os x is based on NeXTSTEP and not just some amalgamtion of "darwin" and apple libraries, plus the fact that the people that worked on mach (like avie tevanian) moved on to NeXT, mac os x is not simply a new apple created os. it's got a long hostory as NeXTSTEP. is nextstep also not an OS? and was nextstep just a "next" branded freebsd and mach? --Yoasif 23:48, 30 November 2005 (UTC)

Boils down to
OK, I've just read most of this debate (I skimmed the middle because it's late, and my eyes hurt) and I think it boils down to a very simple question with a very complex answer: "What is an operating system?" To quote from Talk:Operating system:
 * I think it would be worthwhile to discuss more the differences between "operating systems" as computer scientists see them (kernels -- scheduling, virtual memory, real-time, and so on) and "operating systems" as marketers see them (libraries, GUIs, bundled applications). There is a real terminology gap there. --FOo 04:13, 22 November 2005 (UTC)

Or from our current Operating system article:
 * The exact delineation between the operating system and application software is not precise, however, and is occasionally subject to controversy.

That paragraph goes on to mention the debate over whether IE is part of an OS, which makes debates over something as fundamental as Aqua pale into comparison.

Of course, if you accept that the availability of Aqua makes for a distinct OS, then you'd have to identify "Linux+GNU tools+X11+GNOME" as a distinct OS to "Linux+GNU tools+X11+KDE", and probably that any "distribution" is a distinct OS. But, as the term is generally used, that's probably more-or-less true, in that there's no very fixed line between a distinction that amounts to a new "distribution", and one that makes for a new OS. Measures might include compatibility, bespoke components (e.g. Debian is "more unique" than Ubuntu), adoption, and maybe marketting - on all of which Mac OS X would score very highly as a distinct OS, not a derivative or distribution of something else.

Now, let's turn this debate into something productive and discuss this issue in an article somewhere, perhaps by extending the intro of operating system... - IMSoP 01:11, 1 December 2005 (UTC)


 * Hear hear! --JohnDBuell 03:33, 1 December 2005 (UTC)

Why doesn't the main body of text does not discuss FreeBSD? Is is because a large number (well atleast the vocal ones) of open source fans are linux zealots who are ashamed that their belowe OS was not selected as a basis (well partly) for a mainstream _desktop_ OS?

-sdf

BSD itself (not one of the versions) as mentioned early on as a major basis of OS X. If I remember right, FreeBSD is used as a 'reference platform' but functions/features of the other *BSDs have also been incorporated into Darwin. --JohnDBuell 18:06, 29 December 2005 (UTC)

The users AlistairMcMillan and Yoasif are Acting like punks
They are messing up the Gnu-Darwin article with no criterion at all, expecially that moron of Alistair that put a screenshot from the GNU-Darwin site in the Darwin article.100%IDIOT. If they don't stop then I'll ask for arbitration quite soon. Luxiake


 * Please see our policy on No personal attacks. Thank you. AlistairMcMillan 23:48, 1 December 2005 (UTC)


 * Sorry, but the way I see it, and reading up on what's currently posted at gnu-darwin.org, they're both correct, and making accurate changes to that article. And I hate to say this, but that sort of immature namecalling will likely NOT get very far in an article dispute. --JohnDBuell 23:49, 1 December 2005 (UTC)


 * Oh and for the record it was Yoasif who added the screenshot.  And you can find the Arbitration page at: Requests for arbitration.  Thanks again. AlistairMcMillan 23:51, 1 December 2005 (UTC)

Fine,then.luxiake
 * Luxaike, don't you think you are taking this a little bit too far? You need to chill, if you don't mind me saying. Your arguments over both OS X being a distribution of Darwin and Gnu-Darwin have both been incorrect. — Wackymacs 07:43, 2 December 2005 (UTC)

Are you a member of the gang or just a common mac-maniac? I know that these guys are administrators, but I don't care, because I got truth by my side and moreover their behaviour was simply vandalic. luxiake


 * Please look at the history of this guy's edits before taking him seriously. He really doesn't know what he is talking about. AlistairMcMillan 09:19, 2 December 2005 (UTC)

Naming
Now, isn't Mac OS X POSIX-compliant? Does that justify its Unix-naming? 20:30, 5 December 2005 (UTC)


 * According to the article Mac OS X "is not compliant with the Single UNIX Specification" (it has compatibility features, but then so does Windows NT) and Apple "has not had the OS officially certified" with the holders of the "Unix" trademark. So, no and no. Whether or not they are right in considering the trademark to have already become "genericized" (and thus unenforcable) is for the courts to decide, if The Open Group bother to defend it. - IMSoP 23:35, 5 December 2005 (UTC)

Cross Pollenating?
I would be interested in an article on the feasibility/implications of installing Windows on an originally Mac computer, or the other way around.


 * This isn't really possible, or at least won't be until Macs switch to Intel CPUs. The only ways to do it on a PowerPC Mac are through emulation software. Mostly because there was never really a version of Windows for this particular model of PowerPC (yes, I think MS worked on a version of NT 3.51 for PPC, but I doubt you'd get it to work on a G3, G4 or G5 :). And as far as putting the developer versions of Mac OS X for Intel onto non-Apple Intel boxes, Apple considers that to be illegal and a violation of their intellectual property, and will send their lawyers after any site posting that information, which would likely include Wikipedia. They have said that anybody will be able to take their Intel-based machines and install Windows on them (on another drive or partition) but that kind of installation will be unsupported. --JohnDBuell 21:46, 16 December 2005 (UTC)


 * Boot Camp Beta released by Apple Inc. on 4th April 2006. See and download here how to run WinXP officially on girlie computers (size 82MB): http://www.apple.com/macosx/bootcamp/publicbeta.html


 * If you weant to run Mac OSX on a cheap grey box common PC, a russian hacker-god known as Maxxuss has the tools you ned to do that, and Apple Inc. is having a futile crusade to have his wesites closed. 195.70.32.136 08:27, 6 April 2006 (UTC)

Open Group vs. Apple
I have commented out this unsourced section:


 * The Open Group has criticized Apple for use of the term "Unix" in advertisements for Mac OS X as Apple has not had the OS officially certified, and their use of the term could constitute a violation of trademark. Apple claims that they use the term as a genericized trademark and that the cost of certification would make the OS prohibitively expensive, although The Open Group has stated that there is a 110,000 USD upper limit on the cost of certification for one company. Though Mac OS X is contains code from BSD (which was once based on Unix, though modern BSDs avoid the term for legal reasons), it is not compliant with the Single UNIX Specification. The reason for Apple not seeking "official" Unix branding may simply be that compliance is not a near- or medium-term goal for Apple instead of the potentially misleading cost claim.

According to CNET, the Open Group sued apple over the use of the term UNIX. They probably reached an agreement since Apple is now listed here under "Platform Vendors Supporting the Single UNIX Specification", but I couldn't find a source for this. Mushroom 05:11, 11 January 2006 (UTC)

I don't think there HAS been resolution, but I'm still checking. The suit was filed in 2003, a year later there was some commentary on it, AND a notation that Apple was on that list, but that they were being sued anyway. http://www.osviews.com/modules.php?op=modload&name=News&file=article&sid=1151 There's another note about it at http://www.bellevuelinux.org/unix-like.html (which claims the osviews article is from 2005; it's not, it's from 2004). --JohnDBuell 07:08, 12 January 2006 (UTC)

See also this Groklaw article. It's from August 2004 and it says: "UPDATE: The parties have signed a stipulation to stay proceedings because they are working out a settlement". Mushroom 07:16, 12 January 2006 (UTC)


 * Well I'd say we've got sources for the root of the dispute in the main article now :) --JohnDBuell 22:03, 12 January 2006 (UTC)

"Based on C" trimmed from intro
I deleted this statement from the intro paragraph: "Unlike previous versions of the Macintosh operating system, such as Mac OS 9, which were based upon the C Programming Language, Mac OS X is based on a type of UNIX kernel (see History below)." because the intro should be as concise and straightforward as possible - it should only say what Mac OS X is, and what it's for. It's an operating system which is distributed by Apple computer for use on (some of) its computers.

The statement was added by User:71.125.172.173. I originally removed it because it is:
 * confusing - readers who are unfamiliar with C and/or Unix won't understand or care until they read further. What's more, it confuses the language used to program the system with the architectural base of the system.
 * technically inaccurate - the original Mac OS was written in Pascal, except for some of the Toolbox and OS functions in ROM which were assembly language. Apple didn't start using C until much later. Also, XNU is not simply a "Unix kernel", as the Mach/BSD discussions illustrate.
 * unnecessary - as I wrote above, the intro doesn't need to state this information, and it is contained further down in the article (and in supporting articles such as Darwin (OS) and Architecture of Mac OS X.

I won't keep reverting the statement indefinitely, but I wanted to get a feel for whether you guys (the regular editors on this article) agree with the removal, or would you rather see a corrected version of the same information added to the intro. Thanks! MFNickster 23:26, 14 January 2006 (UTC)


 * Agreed. huwr 23:52, 14 January 2006 (UTC)


 * ...and there's more to Darwin than just the xnu kernel, in any case (other libraries, commands, daemons, etc.). Guy Harris 00:08, 15 January 2006 (UTC)


 * I agree. Mushroom 00:45, 15 January 2006 (UTC)

"all shipping Apple Macintosh computers"
The intro claims that Mac OS X is included in "all shipping Apple Macintosh computers", however if I recall correctly Yellow Dog Linux is included in some machines by Terra Soft Solutions. Which has a special agreement with Apple Computer. —Ævar Arnfjörð Bjarmason 09:57, 29 January 2006 (UTC)
 * Yes, Terra Soft is an Apple reseller. Whether or not they get the systems from Apple with an OS already on them is anybody's guess, but they add Linux as a VAR service [ side note: is putting Linux on a Mac adding or subtracting value? ]. In any case, I take the statement to mean that the OS is included with all models of Mac shipped by Apple, which I believe is accurate. - MFNickster 14:09, 29 January 2006 (UTC)

POV?
Some parts of this article, in particular the "Pricing Criticism" section, and parts of the "History" and "Description" sections, seem rather POV in a pro-Apple sense. Has anyone else noticed this? 68.238.93.104 19:35, 20 February 2006 (UTC)
 * Agreed. Some of the recent additions by User:207.135.154.96 are short on facts and big on gushing. Most of them should probably be reverted. MFNickster 19:38, 20 February 2006 (UTC)

Mac viruses
Anything about the recent virus attacks on mac os x? --hello, i'm a member  |  talk to me!  23:03, 20 February 2006 (UTC)


 * Easy, they weren't viruses, they were "Proof of Concept" code, and Trojan Horses at best, not real self-replicating viruses that required NO human intervention. See the EXCELLENT discussion/dissection of them at Ambrosia Software's forums: . --JohnDBuell

xnu isn't a microkernel
It's not a microkernel in the way the microkernel article describes microkernels:


 * A microkernel is a type of kernel which consists of a simple abstraction over the hardware, with a set of primitives or system calls to implement minimal OS services such as address space management, thread management, and inter-process communication. All other services, those normally provided by the kernel such as networking, are implemented in user-space programs referred to as servers.

In OS X, networking and file systems, and device drivers for networking and storage devices, as well as some other device drivers, run in kernel space, just as they do in, for example, Windows NT. Both NT and OS X are described on their pages as having a Hybrid kernel, which is a more accurate description of their architecture.

In addition, there's nothing about being a hybrid kernel that makes moving to another platform easier; the way the low-level platform support in xnu works is similar to the way it works in portable monolithic kernels such as the SunOS 4.x and 5.x kernels, the various BSD kernels, and the Linux kernel. For example, the VM system in SunOS 4.x and 5.x has a "HAT layer" (Hardware Address Translation layer), which offers low-level MMU services to the platform-independent VM system and code using the VM system; the differences between the MMUs are largely encapsulated in the HAT layer (with some exceptions in, for example, startup code). Similarly, the Mach VM system has a "pmap layer" offering low-level MMU services - but the Mach VM system was adopted by BSD without BSD becoming a microkernel. Various I/O bus services, etc. are also offered by low-level code, with different implementations on different platforms. Guy Harris 23:45, 25 February 2006 (UTC)


 * Okay the point I'm trying to make is that it was easier for Apple to port Rhapsody/Mac OS X to another platform than it would have been to port the old Mac OS. If you have a better way of wording that, then please feel free to make any changes as you see fit.  AlistairMcMillan 00:03, 26 February 2006 (UTC)


 * It appears you've already done that. Thanks! :-) Guy Harris 00:38, 26 February 2006 (UTC)

Using OS X on non-Apple hardware
24.6.64.34 changed 'several people have been using a development version' to 'some people have been illegally using a development version'. Which weasel word is best, 'some' or 'several'? Or should we do something else there, such as cite statistics?

Also, what is the legality of Mac OS X in various countries? Should we comment on this? 4.242.147.106 00:40, 8 March 2006 (UTC)


 * Use of OS X is legal worldwide, to the best of my knowledge. It's usually encryption algorithms that get restricted for sale outside of certain nations. But, IANAL. --JohnDBuell 04:19, 27 March 2006 (UTC)

The flames of the holy spirit did not descend on Mac OS X.
What language versions exists of MacOSX? You can have Windows XP versions in 40+ native tongues like tamil, suaheli or hungarian. I heard Mac OS only exists in american, german and french. This sounds like cultural imperialism, while Windows supports preservation of ethnic diversity.

I give credit to Bill Gates for promoting understanding between different cultures, e.g. the arabic and hebrew software versions of MSDN are always given on the same single CD, forcing the enemies to handle and accept the other side's "unholy" material. this kind of activism is missing from Apple. 195.70.32.136 08:13, 6 April 2006 (UTC)


 * Language editions for OS X are, as per the OS X technical specifications, "English, Japanese, French, German, Spanish, Italian, Dutch, Swedish, Danish, Norwegian, Finnish, Traditional Chinese, Simplified Chinese, Korean, Brazilian Portuguese". Guy Harris 08:47, 6 April 2006 (UTC)
 * 195.70.32.136, Mac OS X, in addition to being available for those languages, lets you switch between languages with the same installation. --BDWill Talk Contribs 16:16, 6 April 2006 (UTC)


 * Those are the languages Apple lists there, but I see a many more in my System Preferences. If you have a (Tiger) Mac, go into System Preferences, click on International, and in the Language tab, click the Edit List button. In there I see all kinds of languages... Schweizer Deutsch, Tagalog, Malti, Polski, and many things in fonts I can't begin to read. What are all these if not supported languages? Epastore 02:14, 7 April 2006 (UTC)


 * Try adding, say, "Català" to the list with "Edit List", close the "Edit List" box, and drag "Català" to the top of the list. Then log out and log in again; how many of the apps that come with OS X show menus, etc. in Catalan?  If the answer is "none", then Catalan is not a supported language, at least for those applications.  Guy Harris 02:39, 7 April 2006 (UTC)


 * The Finder is an application too, and it's up to the application developer to localize the app. The Finder is localized to 15 languages, but there is font and input support for more. I must say, though - I think criticizing Apple for not supporting enough languages is fine, but criticizing them for not promoting a particular international political or ethnic agenda is (forgive me) silly. MFNickster 03:57, 7 April 2006 (UTC)


 * Indeed. ~ Ross (ElCharismo) 19:47, 13 April 2006 (UTC)

Proprietary or not
Doesn't Mac OS X qualify as proprietary? Apple owns it. They restrict how people can use it. They control how it evolves. Granted part of it is released as free software and people can do, within the restrictions of the license, whatever they want with that. But that doesn't change the fact that Apple owns and controls the development and use of Mac OS X. AlistairMcMillan 18:19, 22 April 2006 (UTC)


 * It depends on how you define 'proprietary,' but I think the most important point is that it is developed, published, and sold by Apple. You can only get it from Apple, even though you can get Darwin and some other parts of the OS from other sources. That said, I think it's safe to say that Mac OS X as a whole is proprietary - but I don't think it's necessary to put that in the intro, because it's 1) fairly obvious and 2) discussed in detail in the Description section. MFNickster 21:48, 22 April 2006 (UTC)


 * "fairly obvious" and "discussed below". Couldn't someone make the same argument for mentioning that Mac OS X is an "operating system" and produced by "Apple Computer"?  Why do we mention them in the intro?


 * The intro should be as concise as possible, stating what Mac OS X is (an operating system), and what it's for - which I think is reasonable, but don't let me stop you from holding it up to rhetorical scrutiny. MFNickster 23:16, 22 April 2006 (UTC)


 * Given that Microsoft Windows mentions Windows proprietyary-ness in the intro and Linux & FreeBSD etc mention their un-proprietary-ness in their respective intros, I think we should probably include the fact that Mac OS X is proprietary front and centre. AlistairMcMillan 22:43, 22 April 2006 (UTC)


 * In that case it wouldn't really be out of place, but it's hardly necessary. Given the ambiguity of its overall proprietariness (shown by the simple fact that there's an argument about it). I don't really mind it being there as long as it's by consensus, but I do think it has been stuck in there by open-source advocates simply to label it "non-free." Large parts of the package are free: the kernel, BSD userland, firewall, Web server, shells, utilities, MySQL, Perl, Python, blah blah blah. It's been argued to death already. MFNickster 23:16, 22 April 2006 (UTC)


 * I don't really think there is any debate. Apple has sole ownership and control over it, so that makes it proprietary.  They may release parts as free software, but only Apple controls how that affects Mac OS X.  And there is a massive difference between using Darwin as your operating system and using Mac OS X.


 * Well, at the risk of having another round of argument over a very minor edit... the terms in question are not crystal clear. Is "Mac OS X" just the OS (kernel and userland tools) or is it everything included on the disc you buy? Read the file "Acknowledgments.rtf" in your Library/Documentation folder. Mach is owned by Carnegie Mellon University. The Regents of the University of California have ownership and control over BSD. Is CUPS part of the operating system? Apple doesn't own CUPS. They don't own GNU Emacs or Bash or Apache or ipfirewall or Samba either, but those are included with the OS when you buy it. I'd say it's false that Apple has sole ownership and control over the entire Mac OS X package. MFNickster 03:26, 23 April 2006 (UTC)


 * I Agree with alistair - it is definately proprietary. I suppose one could debate the merits of it as per above, but the fact is that OS X itself (Aqua, Quartz, Cocoa, Carbon, QuickTime, etc. etc.) is locked up tight in a traditional MegaCorporation manner. I was sort of excited when wxWidgets was included as part of the OS (I was a developer on that project for a year), but it didn't make it any less proprietary - that's for sure. You know, I personally kind of wish they would release the old classic API source one day so that people can hack around with that... Just another star in the night T 05:47, 23 April 2006 (UTC)


 * I get your point, but I notice that you're begging the question by picking the parts of the OS which are undeniably proprietary, and calling those "OS X itself." MFNickster 07:09, 23 April 2006 (UTC)


 * Windows includes tons of code licensed or purchased from other developers, code that did not originate at Microsoft, does that mean Windows is not proprietary either? And the parts of Mac OS X that Apple release under a free license don't come anywhere close to giving you the same experience.  Apple does have sole control over Mac OS X as a whole.  Who else has any control over what you get on the Mac OS X CD/DVD?  Who else controls what ships as part of the operating system? AlistairMcMillan 21:20, 23 April 2006 (UTC)


 * I've already granted that the Mac OS X package as a whole is proprietary, for that very reason - Apple is the sole vendor and they control what gets shipped as "Mac OS X." Again, I think we're using the terms in an unclear way - by "control" I don't mean just choosing whether to license a tool or technology, I mean copyright control - which is to say that Apple cannot prevent you from licensing, modifying or redistributing Mach, BSD, Apache, Bash, or any of the other non-Apple items I mentioned above. The point is that some software shipped as integral parts of Mac OS X (e.g. Bash, ipfirewall) is emphatically not proprietary, and cannot be made proprietary just by licensing it. MFNickster 22:27, 23 April 2006 (UTC)


 * Do you accept that if you put together all the free or non-Apple parts that what you have in no way resembles Mac OS X? AlistairMcMillan 23:04, 23 April 2006 (UTC)


 * How is that relevant? Nobody said Mac OS X was 100% free - I was only trying to show why there's an argument over it being "100% proprietary." MFNickster 23:07, 23 April 2006 (UTC)


 * How is it relevant that some parts of Mac OS X are available elsewhere? The only way to get Mac OS X is to buy it from Apple.  The only people that control what makes Mac OS X are Apple.  AlistairMcMillan 23:19, 23 April 2006 (UTC)


 * You guys already beat this to death on the debate over Darwin and whether Mac OS X is a "distro" of Darwin. I agree that Mac OS X is proprietary even though it is built on a non-proprietary foundation, because the licensing on the free parts allows it. Others may disagree, and I think they have good reasons to disagree. Apple could have written an all-new kernel and userland for the OS, but they didn't. That's why Mac OS X is considered a Unix-like OS. If you put together all of Apple's proprietary code (that which they have all rights to), what you have "in no way resembles Mac OS X." Is that relevant? MFNickster 00:00, 24 April 2006 (UTC)


 * Correct me if I'm wrong but I don't think Apple had to release anything, did they? They bought the XNU kernel from Next, the BSD parts are under the BSD license obviously, they wrote the I/O Kit themselves, Apache is under the Apache license which allows closed source, if I remember correctly they used another shell up to v10.3 instead of Bash... as far as I can see the only bit you've mentioned above that requires them to release source changes is the GPL licensed Samba.


 * It doesn't matter whether they need to release it or not. If they don't retain all the rights to it, they don't control it - which I would call a requirement for it being proprietary. CUPS is GPL, by the way. MFNickster 01:37, 24 April 2006 (UTC)


 * But if the license doesn't require them to do anything except not remove the license (kernel, Apache, etc) then they do control it. They can, and do, make whatever changes they want.  Granted they have to release any improvements they make to Samba (and CUPS) and any other GPL licensed components, but that doesn't stop them from taking their versions in whatever direction they want. AlistairMcMillan 01:50, 24 April 2006 (UTC)


 * See above point about the term "control." Also: that strikes me as an argument in favor of those parts being free rather than proprietary. MFNickster 03:37, 24 April 2006 (UTC)


 * And if you "agree that Mac OS X is proprietary" why are we even discussing this? AlistairMcMillan 00:25, 24 April 2006 (UTC)


 * Because you brought it up after I reverted what I saw as a superfluous edit. Even though I may agree with you, I'm trying to show you that it's not a simple "yes or no" answer, because it depends on how you define the terms. MFNickster 01:37, 24 April 2006 (UTC)


 * Also the fact that open-source advocates are trying to insert this into the intro doesn't change the fact that it is true. Besides, as anyone who uses one of those open source operating systems for any length of time quickly learns, to paraphrase JWZ, it is only free if your time is worth nothing. AlistairMcMillan 02:31, 23 April 2006 (UTC)


 * No matter how much time you devote to it, it's still "free as in speech, not as in beer." :) MFNickster 03:26, 23 April 2006 (UTC)


 * "How did you dare! I tried to discuss, and you changed it back to 'proprietary!' Now, you restore the whole things, or you will see the Armageddon!"
 * (...sorry, couldn't resist!) MFNickster 02:34, 24 April 2006 (UTC)


 * Ah the good old days. AlistairMcMillan 02:45, 24 April 2006 (UTC)


 * YOU, MAC MANIACS! Free Software will kill the bad apples! Luxiake 15:30, 2 May 2006 (UTC)

New image inappropriate for this article
I have removed the screenshot image from the article, for the following reasons: Can someone create a new image like the old one, showing a default install? --TreyHarris 17:56, 23 April 2006 (UTC)
 * The old image showed a default install of OS X 10.4 Tiger. The new image now displays several pieces of software which are not part of OS X, and so the image is less relevant to the article.
 * These other pieces of software (Starcraft, Firefox, Adium, X-Chat, etc.) are not the subject of this article, so the screenshot does not qualify as fair use.  While some of them, being GPL software, are still allowable images, display of proprietary software like Starcraft isn't allowable in this article.


 * Fixed. Just reverted Image:TigerDesk.png to an earlier version.  AlistairMcMillan 18:08, 23 April 2006 (UTC)

256x256
Mac OS X icons were limited to 128x128 pixels until v10.4. Apple upgraded Tiger to allow 256x256 pixels. 

However the Finder wasn't upgraded to use 256x256 icons. So the only way you can see icons of that size is to use a third party application like Path Finder.

Do I have something wrong here? We can't just say 256x256 icons, because version 10.4, 10.3, 10.2, 10.1 and 10.0 didn't have support for that, and I'm assuming given that this is Mac OS X and not Mac OS X v10.4 everything stated on this page should apply to all versions unless stated otherwise. AlistairMcMillan 18:51, 1 May 2006 (UTC)


 * If the wording I've chosen is awkward, or could cause confusion, please by all means change it. AlistairMcMillan 18:52, 1 May 2006 (UTC)


 * If you want to put the 256x256 feature under "Versions -> Mac OS X v10.4 (Tiger)" that would be one solution, otherwise I think the "Notable Features" section implies features of the current release. If there is no agreement on how to include the information, it's best to just leave it out. MFNickster 19:17, 1 May 2006 (UTC)


 * Spotlight, Dashboard, Expose etc, when listed under Notable features, all clearly state when they were introduced. AlistairMcMillan 19:28, 1 May 2006 (UTC)


 * True enough. How about "Full-color continuously scalable icons up to 256x256 pixels (introduced in version 10.4)"? MFNickster


 * Firstly, why is "full-color" notable when full-color icons have been available since Mac OS 8.5? Why can't we mention 128x128 since all versions of Mac OS X support 128x128 icons?  Why focus only on the half-implemented 256x256 feature of Tiger? AlistairMcMillan 18:01, 3 May 2006 (UTC)


 * "Full-color" is factual; it is no less a feature of the OS just because previous versions also had that feature. As for the size, it's trickier because the "half-implemented" status puts us in a position of not being able to say much about it. 128x128 and 256x256 are both partially correct. If full-color isn't notable, then scalability probably isn't either, and the entire line about icons should probably be stricken. MFNickster


 * And, just out of curiosity, is there anywhere else you can make use of 256x256 icons aside from Path Finder? AlistairMcMillan 19:29, 1 May 2006 (UTC)


 * I don't know. Is there anything Path Finder does to implement 256x256 icons that isn't already provided by Tiger's Icon Services? MFNickster 01:30, 2 May 2006 (UTC)


 * I haven't read their code, but my guess is they just use whatever API Icon Services provides. I just think if we are going to mention 256x256 icons, we should probably mention the only place (as far as I know) that you can actually make use of them.  Given that Apple didn't update their Finder to use them. AlistairMcMillan 18:01, 3 May 2006 (UTC)


 * It may be the only way to exploit the feature at the moment, but Path Finder is hardly a feature of the OS. I don't think it's relevant to the article. MFNickster 04:17, 4 May 2006 (UTC)


 * As a third party observer, I thought I'd add my two cents. I think the current level of detail ("full color, scalable") is probably right for this article. The bit about 256x256 being added for future resolution independence seems more relevant to the Tiger article, so I've added it there. --Steven Fisher 20:21, 3 May 2006 (UTC)

Windows Vista / Aero
The folks at Windows Vista and Windows Aero are trying to delete or bury mention of the similarities between Windows Vista/Aero and Aqua. Any help would be appreciated. -- Gnetwerker 20:28, 8 May 2006 (UTC)

Nothing mentioned regarding security
Searching for 'security' and 'secur*' returns 0 results. There should be at least a little info regarding MacOS security.

having little experience with MacOS I would really appreciate a security section. ---Fractal3 20:03, 9 May 2006 (UTC)


 * Second that - I came to the article to find out about why OS X is supposed to be better than Windows and that kind of thing. I appreciate it might be difficult to police with all the fanboys on either side, but it's certainly needed. Tyrhinis 19:36, 18 May 2006 (UTC)


 * I agree that a security discussion is important for any OS. I am thinking about mentioning the numerous old vulnerabilities found in OS X --> http://www.zdnet.com.au/news/security/print.htm?TYPE=story&AT=39234678-2000061744t-10000005c and http://www.sans.org/top20/2005/spring_2006_update.php . To balance, Apples recent publication of http://developer.apple.com/documentation/Security/Conceptual/SecureCodingGuide/SecureCodingGuide.pdf, containing a section titled "No Platform is Immune", eventually accompanied by this review: http://blogs.23.nu/ilja/stories/11976/ . I'd be grateful for comments and suggestions how to best integrate the security discussion in a friendly, consensual way into the article. --Ministry of Truth 12:33, 8 June 2006 (UTC)

Leopard "garbage collection" claim
I removed this text from the Leopard section:


 * It appears Leopard will support garbage collection, a form of automatic memory management, in the Objective-C runtime. Garbage collection allows developers to create applications that are less prone to crash and in less time. You can see the evidence for this in 10.4 with the latest developer tools installed by typing 'man gcc' in the Terminal (Applications > Utilities > Terminal) and searching for "garbage", which will reveal:
 * Enable garbage collection (GC) for Objective-C objects. The resulting binary can only be used on Mac OS X 10.5 (Leopard) and later systems, due to additional functionality needed in the (NeXT) Objective-C runtime.

This is original research, which is against Wikipedia policy. Wikipedia is also not a crystal ball, but if someone can find a reasonably good source for this information, that'd be good. Warrens 00:27, 11 May 2006 (UTC)


 * Reasonably good source? How is the man page not a good source? You've discredited the official Apple documentation included in every Mac OS X developer tools installation (hardly a crystal ball) but happily accept an internet link that duplicates it*. Have you checked out the man page yourself? Internet links are just one type of reference. I think the section in the v10.5 article is better than what's here now. *(no offence to DMacks - it's a good reference) Matt Peacock 11:02, 11 May 2006 (UTC)


 * I appreciate that you're offended because you wrote the original research into the article, but please understand that not everybody in the world has a copy of OS X 10.4 with the developer tools installed handy to verify a claim. The way you'd written it sounded like you came up with this observation yourself; while this is fine for blogs and news web sites, it's not appropriate for a Wikipedia article.  Verifiability is key, and that's why sources are required.  Another thing to keep in mind for future reference: use of the second person is strongly discouraged.   Warrens 11:18, 11 May 2006 (UTC)


 * I've no problem if someone wants to re-word what I've written - especially if it's not 'wikified'. I re-wrote DNA mismatch repair the other day due to it's appalling state, and hopefully no-one is offended. However, firstly it's not "original research" as you keep asserting and secondly your justification is off.
 * I assume you missed the fact that this has been widely recognised (try Google and ars). Regardless, "Verifiability is key, and that's why sources are required" - please explain why the gcc man page is not a "reasonably good source" - it's official Apple documentation. By all means provide a more accessible source, as DMacks did, or re-write it to conform to Wikipedia standards, but don't remove it simply because you yourself cannot confirm it, or indeed did not know how to confirm it.
 * Not everyone "in the world" has access to the journals or books which are cited as references, and indeed are typically more accurate than internet (many publications refuse internet references), should we exclude those on the same grounds that not everyone can verify them? TNF-alpha has negative paracrine effects on insulin synthesizing pathways by increasing serine phosphorylation of IRS-1 and down-regulating GLUT4 expression. True or false? Arner, P., (2005). Insulin resistance in type 2 diabetes - role of the adipokines. Current Molecular Medicine. 5(3):333-9.
 * "How-to"s aren't the kind of verifiable source that Wikipedia is looking for. You wouldn't, for example, prove that the sky is blue by instructing the reader to look at the sky.  Even that piece of information has an external source.  Remember, Wikipedia is a tertiary source -- not primary, not secondary.  Anything we want to write about needs to be verifiable by reliable external sources, and we need to cite those sources.  Providing a citation to Apple's developer documentation is sufficient for this case... but it is required, especially when stating something that can be categorised as being forward-looking or otherwise crystal-bally.  I'll delete anything that sounds crystal bally or of original research because such things don't belong on Wikipedia, period.  No original research is quite clear about this.  If reliable sources can later be provided, then the text can be re-added -- no harm done.  Warrens 15:45, 11 May 2006 (UTC)
 * Firstly your original research argument is tired (and incorrect). If I read and quote "A Christmas Carol" is it original research? No. Equally then reading and quoting the man pages (i.e. manual) isn't original research, so let go of that one. I didn't suggest that people looked to the sky - I told them the developer documentation included the information. Granted, it was poorly worded - perhaps "the latest gcc man pages state" would have been better - but that's easily corrected by someone more familiar with Wikipedia writing style, like you. Secondly, a more accessible source is better and once again that's easily amended. If you were aware of what the man pages were, you would have recognised what was being said and reworded the as to conform with Wikipedia style. Instead you lacked this knowledge, and so immediately assumed the "claim" was speculation or "crystal bally" when in reality it was fact, a fact more easily verified than most references provided on any article.


 * I asked you why the man pages are not a good source, and you (indirectly) reply that they are. However you don't accept the original (i.e shipped) man pages, you only accept a transcript of them provided on a website. According to you the former is completely dismissible whereas the latter is "excellent" and grounds for re-instating the entry, which is odd, considering that the the content is identical, from the same source . The issue is that rather than removing my contribution because it didn't provide a source, you removed it because you didn't recognise what the source was.Matt Peacock 17:32, 11 May 2006 (UTC)

Is the compiler manpage not sufficient? Or was the problem just that we need an external link to its actual text instead of a "he said that it said the following..."? There is even some detail about how it will work at the compiler level in there. DMacks 00:38, 11 May 2006 (UTC)


 * There is evidence for this, but I don't think it's necessary to cite it here. All we need to say is "It appears Leopard will support garbage collection, a form of automatic memory management, in the Objective-C runtime". Perhaps backed up with a link to one of the many sites that have picked up on this recently. WP is not a developer manual. Graham 00:45, 11 May 2006 (UTC)


 * That link is excellent! It doesn't support the claim of "Garbage collection allows developers to create applications that are less prone to crash and in less time"; a wikilink to Garbage collection (computer science) should be sufficient.  I'll add the information back into the article.  Warrens 00:46, 11 May 2006 (UTC)

To those people who are taking a snippet in a man page as hard evidence, keep in mind that many technologies get pulled at the last minute for all kinds of reasons. Careful perusal of even the first public Darwin sources would have shown massive x86 support throughout, but it was of no use in predicting when Intel Macs would appear, or what form they would take. If a technology is committed to appear in Leopard, there will be a press release, it won't be hidden away on a man page. Stan 23:29, 11 May 2006 (UTC)


 * Well said - and with WWDC just around the corner why don't we just wait and see? It's not as if the information has to be here, right now - after all, it's not wikipedia's job to be yet another rumour mill for Macs. Graham 10:52, 12 May 2006 (UTC)

Darwin kernel on Intel
In relation to these two edits, I remember reading somewhere that the PPC version of the Darwin kernel compiles on Intel as well like it always has. However you end up with a kernel that runs fine but just doesn't match the Mac OS X on Intel kernel. Can't remember where I read this though. Anyone have any ideas? AlistairMcMillan 18:01, 17 May 2006 (UTC)


 * My understanding is that the source for the x86 aspects of Darwin (kernel and drivers) have not been released by Apple, but the rest of it has. I'm not sure if this is still the case or not.  Some interesting related reading:  Warrens 20:53, 17 May 2006 (UTC)


 * Up to and including Mac OS X 10.4.3 Apple were releasing Darwin packages that were supposed to compile on both PPC and x86. From 10.4.4 onward they are releasing separate packages and the x86 specific packages don't include XNU and some others.  You can see this for yourself here: http://www.opensource.apple.com/darwinsource/


 * If you compare the sources (and I know this original research on my part, which is why I'm posting here asking for source) you can see that Apple haven't removed any i386 code from XNU since starting separate x86/PPC releases. So, in theory, it should still compile on x86 machines. AlistairMcMillan 21:32, 17 May 2006 (UTC)


 * I'm leaning towards think something along the lines of "officially the source for the kernel and drivers are not available in the X86 distribution" - there still hasn't been an actual official announcement as to future plans, but currently that seems to hold up. (so far it is just on slashdot and some news sites, but it could be everywhere pretty soon). I like warrens link :) - here are a couple more (from the same post but still interesting):
 * http://apple.slashdot.org/comments.pl?sid=185992&cid=15351035
 * http://listserv.cuny.edu/Scripts/wa.exe?A2=ind0602&L=macenterprise&T=0&P=58970
 * It is as it always was T 22:04, 17 May 2006 (UTC)


 * Okay, interesting. Hmm... postings from relevant mailing lists should be a good way of getting reasonably safe citations for this, because surely someone has gone down this path and posted about it.  Including such information in the Darwin article would be a good idea, too, once we can wrap our heads around what exactly is going on.  Warrens 23:28, 17 May 2006 (UTC)

This isn't the source I was thinking of, but at least it says it is possible. Apple employee and OpenDarwin contributor Kevin Van Vechten: "Can you compile the 10.4.4.ppc sources and boot them on a generic i386 system?Yes." AlistairMcMillan 23:37, 20 May 2006 (UTC)


 * Be that as it may, nonetheless without a complete release of the Intel branch of Darwin, we should mention it and I'd respectfully ask for a valid argument to sustain the reverts you made to my additions on the subject. --Ministry of Truth 04:53, 11 June 2006 (UTC)


 * Sorry but I disagree. For a long time now, the kernels that Apple have been releasing as part of Darwin haven't matched up with the actual Mac OS X kernel.  In other words, you couldn't just compile the Darwin kernel and swap it out for the Mac OS X kernel and still have a working operating system.  (If you don't believe me on this point please look through the archives of the Darwin mailing list, you'll find numerous examples of people complaining about this.  Or read "A Brief History of Apple's Open Source Efforts" by Rob Braun.)


 * All that has happened recently is that the kernel sources that Apple have been releasing for six years (which compile for both PPC and Intel) have now been tagged as "PPC". They still compile on Intel just as well as they did before being tagged as "PPC".


 * The thing that Apple haven't done is release the exact kernel source that is shipped with their Intel machines. But they haven't released the exact kernel source that ships with their PPC machines for ages either.  People get confused because they think the Darwin operating system is supposed to be standing shoulder to shoulder with the FreeBSD or Linux operating systems.  It isn't and no-one uses it that way.  It is just there to make it easier for people to develop stuff for Mac OS X.  That is why although people on the mailing lists discovered this "issue" in February at the latest, the news didn't break until about a month ago.  AlistairMcMillan 02:06, 12 June 2006 (UTC)


 * I agree with you that we need to be careful to correctly convey the information, nonetheless I believe that the wide-spread coverage you acknowledged is reason enough to include the undisputed facts and would welcome any language suggestions to complement the "most" in "most of Darwin is open source". --Ministry of Truth 12:31, 15 June 2006 (UTC)


 * AlistairMcMillan, I tried to fix the wrong edit too, but you were faster. Can you live with:
 * source_model = Closed source (Darwin foundation is open source, but part of the Intel branch, including the kernel, has not been published from 10.4.4 on.)
 * ? --Ministry of Truth 17:59, 15 June 2006 (UTC)

Short summary:
 * 1999 Apple releases Darwin operating system (including kernel) as open source.
 * 1999 Apple do NOT release Mac OS X operating system (including kernel) as open source.
 * 2006 Apple still releasing Darwin operating system (including kernel) as open source.
 * 2006 Apple still NOT releasing Mac OS X operating system (including kernel) as open source.

The ONLY thing that has changed:
 * 2006 Apple put Darwin kernel (that compiles on Intel and PPC) under a heading titled PPC.

Let me say it again. The Darwin kernel source that is available NOW compiles on Intel. Apple did not rip out the Intel code from the kernel source they now call "PPC".

If you want to write a paragraph about how in 2006 a bunch of people got their panties in a bunch over nothing, then by all means go ahead. AlistairMcMillan 18:02, 15 June 2006 (UTC)


 * Actually Alistair, one thing that has changed is that you can't compile the sources for Intel and replace the kernel of your Mac OS X install, if you are running an Intel-based Mac. That means that developers who previously relied on the ability to test kernel-level code in Mac OS X can't do this on new Macs anymore. It only affects a small portion of the developer community, but it's not "nothing." MFNickster 19:35, 15 June 2006 (UTC)


 * Are people able to do this on PPC? AlistairMcMillan 20:20, 15 June 2006 (UTC)


 * Yep! MFNickster 20:31, 15 June 2006 (UTC)


 * Great but this guy doesn't actually say he has done this, just that he thinks other people did and now can't. Do we have an actual example of someone who has?  AlistairMcMillan 20:40, 15 June 2006 (UTC)


 * Here's a page with instructions on how to do it. MFNickster 23:10, 15 June 2006 (UTC)


 * Great but there are instructions all over the place on how to do it. The question is, does anyone?  Has anyone recently? AlistairMcMillan 23:52, 15 June 2006 (UTC)


 * I don't know. But what people are upset about is that evidently, if you buy a new Mac now, you no longer have an option but to run Apple's kernel. You can't run a custom kernel on Mac OS X or Darwin anymore if you have an Intel-based Mac. You're right, though, that Apple never promised you could - some are just pointing out that it's a bit disingenuous to call it "open source" if you can't build it on the majority of shipping Apple machines. MFNickster 00:22, 16 June 2006 (UTC)


 * Firstly do we have an examples of people trying to build Darwin on their Intel Macs? Do we know that Darwin fails on Intel Macs? Secondly why  "disingenous"? I don't remember reading any Apple material promoting the idea that you could compile your own kernel to replace the one that Apple ships. AlistairMcMillan 00:40, 16 June 2006 (UTC)


 * I think the openDarwin thread you quoted mentions that you can't build Darwin on the Intel Macs. The same post by Kevin Van Vechten that you quoted also says: "Can you use DarwinBuild on a Mac OS X 10.4.4 system with an Intel processor? Not currently." Is there a viable alternative to using DarwinBuild?


 * When I say it's disingenuous, I mean that the purpose of providing the source is so that you can compile it, with or without modification. The purpose of Darwin is to serve as the core OS for Mac OS X, as Apple has said. That's why offering source which is forked out of sync with the machines they sell is just paying lip-service to open source - it defeats the purpose. As Rob Braun wrote, "Critical sources are missing which preclude the possibility of continuing Darwin. Namely the xnu sources and the drivers." He specifically addresses the criticism that "all the sources are there," and spells out why the sources provided are not usable for the new Macs. Mac OS X's core OS is now de facto a closed-source proprietary fork of Darwin.


 * As Rob wrote in his article (posted by User:RN above): "One has to wonder why Apple even bothers to release non-GPL'd source at all, if it is unwilling to cooperate with external developers to increase their return on investment and accept external bug fixes and features. Even worse, one has to wonder why people would want to donate their time to such a fruitless and pointless cause." MFNickster 04:29, 16 June 2006 (UTC)

Let's get back to the point:

Can anybody refute this statement:

"Up to and including Mac OS X 10.4.3 Apple was releasing Darwin packages that were supposed to compile on both PPC and x86. From 10.4.4 onward they are releasing separate packages and the x86 specific packages don't include XNU and some others. You can see this for yourself here: http://www.opensource.apple.com/darwinsource/"

If not, what objections exactly are there to include the following wording into the info box:


 * source_model = Closed source (Darwin foundation is open source, but part of the Intel branch, including the kernel, has not been published from 10.4.4 on.) ?

I appreaciate the high-level discussion on what compiles where or not, but these two questions are the only ones relevant to build a consensus on this subject that needs to be correctly represented in this article due to wide-spread and in part misleading press coverage of the non-publication. --Ministry of Truth 21:28, 15 June 2006 (UTC)


 * Again. Apple have never released the exact kernel code that ships with their Intel machines.  So saying "not been published from 10.4.4 on" is totally wrong for a start.  And if I remember correctly Apple have never released the exact kernel code that ships with their PowerPC machines.  Or if they did they haven't for a long time.


 * Also the kernel code that is currently sitting under the heading "PPC" for 10.4.4 is exactly the same code that was available for 10.4.3, except for bug fixes.


 * Everything that Apple were releasing six months ago they are still releasing. Please try to understand, Apple release the Darwin kernel, not the Mac OS X kernel. Nothing has changed.  AlistairMcMillan 21:37, 15 June 2006 (UTC)

Keep better-known Konfabulator name in "Prominent features" section
I think the well-meaning update that did away with the mention of Konfabulator is unfortunate.

We need to use the widely known term (Google: 2,190,000 for Konfabulator -wikipedia vs. 260,000 for "Yahoo! Widget Engine") in any case, otherwise people will not immediately understand what we're talking about.

Whether or not we also mention the Yahoo name or leave that to the redirect is up for debate and I'd like to hear your comments. --Ministry of Truth 00:16, 12 June 2006 (UTC)


 * I rewrote it as follows:


 * Dashboard (introduced in version 10.4) supports small applications ("widgets") that can be called up and dismissed in one keystroke. The inclusion of this feature resulted in contraversy is it was seen by some reviewers and commentators as being a "rip-off" of an existing application, Konfabulator.


 * I rewrote it as such for two reasons: I feel it's important to define what the thing is before we mention the contraversy surrounding it. Also, Konfabulator was the name of the application at the time of the contraversy.  Warrens 00:35, 12 June 2006 (UTC)


 * I was the one who did it - how about rewording that last bit above as "Konfabulator (now known as Yahoo! Widget Engine)"? The point is, the name is Yahoo! Widget Engine, and Konfabulator will redirect to that page anyway. Put both names on - it'll be both accurate and understood. –-  kungming·2 | (Talk ·Contact) 01:56, 12 June 2006 (UTC)


 * I'm marginally against noting the name change, but only out of concern for brevity and clarity; it's a feature list of OS X, after all, and it's good not to drift too far from that for too long. Warrens 04:46, 12 June 2006 (UTC)


 * I'd side with Warrens to say that the new name lacks relevance given the desirable brevity of list items, but won't insist as long as Konfabulator and the significant "rip-off" debate are mentioned. The ars technica source should probably be kept, otherwise, some zealot might dike the whole rip-off part out as "unsubstantiated", how about:
 * "* Dashboard (introduced in version 10.4), supports small applications ("widgets") that can be called up and dismissed in one keystroke. Its similarity to the pre-existing Konfabulator started a controversy. " --Ministry of Truth 08:19, 12 June 2006 (UTC)
 * Good point - since the controversy is already covered in the Dashboard article, and is completely irrelevant to the list of Mac OS X features, I recommend striking it from this article completely. Oh, and "rip-off" is a weasel word. MFNickster 08:31, 12 June 2006 (UTC)
 * I concur - as I believe the similiarities have been noted in quite a few articles - the Konfabulator article, Dashboard, Mac OS X 10.4 Tiger, and even the Microsoft Gadgets article all mention something along the lines that "A controversy has arose over reputed similiarities betweeen Dashboard and Konfabulator.". Do we need that many mentions of this? It's the Mac OS X article - maybe this issue should be left (among Mac-related articles) to the OS X 10.4 Tiger article and Dashboard only. –-  kungming·2 | <font color="#999999">(Talk ·<font color="#999999">Contact) 06:59, 13 June 2006 (UTC)
 * I just took out the weasel word and redacted the whole thing to be shorter without cutting information. Hopefully a step forward on the way to consensus on this one. --Ministry of Truth 06:37, 14 June 2006 (UTC)
 * Actually, upon further reflection, I think this might merit a mention in the Tiger section of the article, but not in the list of features. I think including it in this article at all shows an implicit bias. MFNickster 19:50, 17 June 2006 (UTC)

BSD Based?
As a former student of NEXTSTEP I would like to strongly dispute the tidbit that claims OSX is based on BSD. OS X is effectively the same as NEXTSTEP. The BSD compatibility is just stacked on top (the same way you can have linux compatibility in FreeBSD), it's native threads are not 'posix' threads per se. Thus this is a Nextstep based os, not a BSD based os. —Preceding unsigned comment added by Mineralè (talk • contribs)


 * The article does say Mac OS X is based on Mach and BSD, by way of NEXTSTEP. Are you saying you disagree that NEXTSTEP was based on Mach and BSD (i.e. that BSD was just stacked on top of Mach)? Mach is pretty useless by itself without a userland, just as the BSD userland is pretty useless without a kernel. MFNickster 01:14, 14 June 2006 (UTC)


 * Slight correction. Some of the stuff imported from BSD is included in the kernel.  For example the virtual file system stuff I believe comes from NetBSD as well as the UFS itself.  AlistairMcMillan 02:45, 14 June 2006 (UTC)


 * Yep - not so much a correction as a clarification! And it's clear from this page that Darwin is a BSD, with a Mach- and BSD-derived hybrid kernel. MFNickster


 * Yup, it's BSD-based. The networking stack above the drivers, the VFS layer, and several of the file systems that plug into the VFS layer (UFS, NFS, smbfs, devfs) are BSD-based, and much of the code in the bsd/kern directory is BSD-derived as well.  Much of the userland also comes from BSD, even if some of it came from BSD via NeXTStEP.  That doesn't mean any of it is identical to FreeBSD or NetBSD (the VFS layer is different, especially in Tiger), but it's definitely BSD-based.


 * There's no "BSD compatibility" in the sense of FreeBSD having Linux compatibility; that would imply that there was some kind of "native" API, with BSD compatibility being a secondary API. The API for file and network access is the BSD Unix API, and most of the API for process manipulation is the BSD API as well.


 * As for the native threads not being "POSIX" threads per se, I'm not sure the kernel-level threads in FreeBSD (or NetBSD) are "POSIX" threads per se, either - the threads are implemented atop Kernel Scheduler Entities in FreeBSD and atop scheduler activations in NetBSD. Guy Harris 03:21, 14 June 2006 (UTC)


 * My suggestion would be to simply change it from 'BSD' to Unix Like... <span class="user-sig user-Mineralè">—Mineralè 2006-06-14 18:01Z 


 * Sorry but not when it is based on BSD. Please consult Apple's website or the kernel code itself. Please note the directory called "BSD" which includes code from BSD Lite, FreeBSD, OpenBSD and NetBSD. AlistairMcMillan 18:04, 14 June 2006 (UTC)


 * My suggestion would be to leave it as it is, because it's not just Unix-like, it's "BSD-like" - over all, it more closely resembles the other BSDs than it does, say, Linux or Solaris (although the command set is a bit more Linuxy than that of {Free,Net,Open,DragonFly}BSD, e.g. Bash rather than the Almquist shell is the Bourne shell replacement). Guy Harris 18:08, 14 June 2006 (UTC)


 * So, would it be safe to change NEXTSTEP to BSD instead of Unix-like? <span class="user-sig user-Mineralè">—Mineralè 2006-06-18 17:11Z 


 * Probably. If the APIs and commands included BSD APIs and commands of the era, and a lot of the code came from BSD, and so on, I'd put it into the BSD subspecies of the Unix-like species. Guy Harris 18:05, 18 June 2006 (UTC)


 * Isn't NEXTSTEP already in the BSD subcategory of Unix? It also has the "Unix-like" template, which includes both official systems and clones. MFNickster