Talk:MacOS/Archive 14

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

Archive 10

←

Archive 12

Supported platforms including PPC -- a mistake?

I can't tell if this was intentional or an oversight, but as of right now, the info box states supported platforms includes PPC.

Is this an oversight or was it on basis of the Rosetta emulator? Does supported platform refer to hardware or software platform? (Wasn't clear to me.) Thoughts? Dsf (talk) 12:20, 29 August 2009 (UTC)

This article covers all the versions of Mac OS X, including those for PPC. The Mac OS X v10.6 article correctly shows only x86 and x86-64. MFNickster (talk) 14:30, 29 August 2009 (UTC)

Ah, ok, I see. Thanks! Dsf (talk) 14:44, 30 August 2009 (UTC)

RFC: Mac OS X is not Based on UNIX

This discussion has been closed. Please do not modify it.

The following discussion has been closed. Please do not modify it.

Mac OS X is based on NextStep: [1], and NextStep is based on code (subsequently rewritten) from BSD and Mach. I dare anyone to show me a single module of AT&T code in Mac OS X. I imagine they will have a hard time doing so, because OS X is written in objective C, not C, like UNIX. It also uses a PostScript display API. The agreement between the OpenGroup and Apple relates to licensing: [2]. It is not a certification that OS X is based on UNIX. In fact, Z/OS is also listed on the OpenGroup's site, even though Z/OS cannot be more different from UNIX: [3]. Windows also has a POSIX API, by the way. Should we call it UNIX, too? Unless I am soon given convincing evidence to the contrary, I will remove the ridiculous claim that OS X is based on UNIX from this entry.--Validbanks 34 (talk) 01:05, 15 September 2009 (UTC)

Well according to this [4], OS X is it's own branch off of the Mach 3.0 kernel, which gets roots from BSD 4.2, which gets roots from 2BSD & UNIX 32v. According to this same timeline, only OS X Server are direct decedents of NextStep, which still based of of Mac 2.5 & BSD 4.2. This article talk about OS X's BSD Unix roots.[5] The Apple Developer's site even addresses UNIX Development on OS X.[6] And you are going to disregard the ref already in the article that Apple even says it's built on UNIX.[7] The Unix wiki article even addresses OS X in it. I fail to see how OS X does not have strong UNIX roots. 『 ɠu¹ɖяy 』 ^¤ • ¢ 02:54, 15 September 2009 (UTC)

Validbanks, I'm assuming on good faith that you are not the same guy who argued this same point further up on the page. We certainly don't need to open this discussion again. NextSTEP was also based on Unix; in fact, that was touted as a major selling point back in 1988 when it was introduced. NextSTEP and Mac OS X are not a separate family, they are derived from UNIX code and UNIX designs, and are part of the UNIX family tree. Furthermore, most of the low-level system is not written in Objective C. MFNickster (talk) 03:05, 15 September 2009 (UTC)

Gu1dry: No, Mach is a microkernel, whereas BSD uses a monolothic kernel. And Darwin is now a hybrid kernel.

MFNickster: I have used almost every operating system in existence, so trust me when I say the following. Only a noob would ever think that BSD and UNIX are the same thing. BSD was written at Berkley and UNIX was written at Bell Labs. HP-UX, Linux, BSD, AIX, Solaris, etc. are all much more similar to System V UNIX than Mac OS X. Besides the characteristics mentioned above, the file system (HFS+) is unlike any other file system in use. Other non-UNIX characteristics include the use of file extensions and the bizarre file system layout. How is having a /Private/etc folder UNIX-like? Windows has a C:\WINDOWS\system32\drivers\etc folder, too. It must also be a member of the UNIX family, huh?--Validbanks 34 (talk) 03:22, 15 September 2009 (UTC)

When NeXTStep was created, BSD was referred to as BSD Unix and contained AT&T code, which NeXT licensed. It was UNIX, and Darwin was later based on it. The fact that designs were changed and the modules rewritten doesn't change that history. MFNickster (talk)

If I claim to be a member of a family, I should have at least a single gene in common with my supposed ancestors. Mac OS X has no UNIX genes at all. The code isn't there. There is no common DNA. The supposed relation is nothing but an illusion, and even if there were one, it could not be boiled down to a three-word statement: "OS family: UNIX."--Validbanks 34 (talk) 04:00, 15 September 2009 (UTC)

Well, I think you're arguing UNIX purity for the sake of it, as if UNIX never can and never did change. I have no intention of slogging through all this again; please read the discussion at the top of the page titled OS X is not just UNIX based - all of it, including the hidden parts - and then if you have sound refutations of all the arguments and sources that support "UNIX-based," and if you can provide authoritative sources backing up those refutations, then you might be in a position to add a note to the article that the claim is controversial, and call for a vote of consensus on whether to change the "OS family" entry in the info box. MFNickster (talk)

Unfortunately, I have no intention of compromising on this issue. The absence of a relationship is proof that it doesn't exist. It doesn't need proof to be evident. The only thing that must be proven is that something exists, not that it doesn't exist.

Therefore, the claim that OS X is part of a UNIX family must be cited by a reliable source saying exactly that. Improperly-cited material may be removed at any time.--Validbanks 34 (talk) 04:46, 15 September 2009 (UTC)

You must be trolling. Unfortunately for you, your claims must be verifiable and not original research. "This includes unpublished facts, arguments, speculation, and ideas; and any unpublished analysis or synthesis of published material that serves to advance a position." MFNickster (talk) 04:57, 15 September 2009 (UTC)

The only person making any claims here is you. And your claim that Mac OS X is a member of the "UNIX family" must be proven. I don't intend on adding any claims to the entry. I will only remove improperly-cited claims.--Validbanks 34 (talk) 04:59, 15 September 2009 (UTC)

So you say citations in the article & countless others in the talk, mean it's not properly cited. The only making any frivolous claims right now is you ([[User:Validbanks 34|Validbanks 34]). I, sir, think you need your head checked. All you rebuttals make no logical sense. I serious doubt you have any true knowledge of how OS X, BSD, Linux, etc really work. I would advise you take MFNickster's advice and read thru the full discussion above. 『 ɠu¹ɖяy 』 ^¤ • ¢ 05:06, 15 September 2009 (UTC)

Show me a statement from a reliable source claiming that OS X is a member of a "UNIX family." Marketing material from Apple and obscure web sites don't count. I cited a source above claiming that OS X is based on NextStep. The source was published by Addison-Wesley. I find it ironic that you would question my knowledge of operating systems. The only reason you people use Macs is because you don't want to be bothered with how your computer works. And I can see that attitude here. "I don't want to talk about my operating system. Just show me the button I press to check my e-mail." You probably can just barely turn on your computer to check your watch list, and yet you have the gall to claim that Mac OS X and UNIX are the same.--Validbanks 34 (talk) 05:14, 15 September 2009 (UTC)

I'm not denying OS X is based of NextStep (I even stated the correlation of OS X with NextStep), but is NextStep based off of what? UNIX! *confetti* Who even stated I used only Mac? Because I never saw it. 『 ɠu¹ɖяy 』 ^¤ • ¢ 05:23, 15 September 2009 (UTC)

And NextStep used parts of BSD -- not UNIX. And I'm descended from Adam and Eve. That must mean I'm a member of the "human family," right?--Validbanks 34 (talk) 05:40, 15 September 2009 (UTC)

If you're going to ignore evidence such as that I mentioned earlier (BSD was called "BSD Unix" when NeXTStep was created), then arguing is futile. MFNickster (talk) 05:45, 15 September 2009 (UTC)

Do any of your sources state explicitly that Mac OS X belongs to a UNIX family? All I see is claims that they're somehow related. I don't see anything about a family, which is what the entry mentions.--Validbanks 34 (talk) 05:50, 15 September 2009 (UTC)

No. Unless you have some new sources to bring to the table covering this debate, I think we're done here. Bye. MFNickster (talk) 05:51, 15 September 2009 (UTC)

No. We're not done. I still have to remove that ridiculous claim in the entry. See you tomorrow.--Validbanks 34 (talk) 05:53, 15 September 2009 (UTC)

Here is the current version of Apple's document supporting the statement that Mac OS X is "Unix-based": [8] There is nothing improper about it. I've updated the citation. MFNickster (talk) 05:08, 15 September 2009 (UTC)

Note: guys, feel free to disagree and debate, but no personal attacks. Keep it civil. MFNickster (talk) 05:17, 15 September 2009 (UTC)

This is obviously the same guy as before. He is making exactly the same mistakes: (paraphrasing) "Nextstep and Mac OS X are written entirely in Objective-C", "BSD is not Unix", "if it doesn't have a monolithic kernel it isn't Unix", etc etc etc. By discussing this issue with him all you are doing is wasting your time and keeping the troll entertained.

Consensus here is clear. All the sources are clear. If the troll starts editing the article, ignoring all the evidence, his account will be blocked. AlistairMcMillan (talk) 07:50, 15 September 2009 (UTC)

Hi, Alistair. I was willing to give him the benefit of the doubt until there's evidence of sock-puppetry, but it is awfully familiar. MFNickster (talk) 14:10, 15 September 2009 (UTC)

Let me get this straight: You're calling me names, and I'm supposed to be the troll? Every single statement you just said is BS. The sources are not clear. I am not a sock-puppet (although your editing patterns are indicative of a bad-hand sock). And I will remove that statement. Just because I agree with someone doesn't mean that I am that person. Is this making any sense to you at all? What's wrong with you?--Validbanks 34 (talk) 16:41, 15 September 2009 (UTC)

I refer the honourable gentleman to my previous statement. AlistairMcMillan (talk) 21:33, 15 September 2009 (UTC)

I still think the whole thing is ridiculous -- again. Dravick (talk) 05:07, 16 September 2009 (UTC)

Multics Family

Moved from Talk:FreeBSD

– and others:

--Tothwolf (talk) 19:44, 16 September 2009 (UTC)

UNIX is based on Multics. So why don't we just say that it's a member of the Multics family? Why stop at UNIX (only three levels deep) when we can drill down four layers?--Validbanks 34 (talk) 23:43, 15 September 2009 (UTC)

Cite? MFNickster (talk) 23:47, 15 September 2009 (UTC)

[9]--Validbanks 34 (talk) 23:50, 15 September 2009 (UTC)

I'm sorry, where does that say UNIX is based on Multics? MFNickster (talk) 01:52, 16 September 2009 (UTC)

OK. How about this one: "UNIX is based on Multics": [10].--Validbanks 34 (talk) 02:05, 16 September 2009 (UTC)

That's a good reference. Now, applying your own logic, does UNIX (say, circa 1974) share any "genes" with Multics? Are you going to propose changing all the articles in the category Unix variants, or should I do it? MFNickster (talk) 02:10, 16 September 2009 (UTC)

Yes. They share source code. And, yes, you can change all of them to read Multics if I can change all of the Mac articles.--Validbanks 34 (talk) 02:12, 16 September 2009 (UTC)

On second thought, why don't you just replace all instances of the word UNIX in the Mac articles, too? No sense in killing two birds with two stones.--Validbanks 34 (talk) 02:15, 16 September 2009 (UTC)

Hello?--Validbanks 34 (talk) 02:23, 16 September 2009 (UTC)

Your source says that "features" and "ideas" for UNIX were borrowed from Multics; do you have any evidence that actual code was used? MFNickster (talk) 02:26, 16 September 2009 (UTC)

But you said it doesn't matter that OS X doesn't share source code with UNIX. All that matters is that they are "related." I'm just trying to put myself in your shoes. Multics is also related to UNIX, and if Mac OS X is related to UNIX, then it must be related to Multics, right?--Validbanks 34 (talk) 02:33, 16 September 2009 (UTC)

Uh huh. What happened to "I have no intention of compromising on this issue" and "I won't stand by while readers are given inaccurate information"? MFNickster (talk) 02:36, 16 September 2009 (UTC)

I have no intention of allowing the word UNIX in that table row, no. But I realized today that the claim that Mac OS X is a member of a UNIX family is inaccurate in two ways--not just one. Not only is UNIX not the great-grandparent of Mac OS X, but it is wrong to name the supposed family after UNIX, when UNIX is derived from Multics.--Validbanks 34 (talk) 02:43, 16 September 2009 (UTC)

There are a lot of articles in the "Unix-like" category. Why don't we suggest they all change to "Multics-like" and see what the consensus is? Any bets on the answer? MFNickster (talk) 02:48, 16 September 2009 (UTC)

Sounds good. Would you like a JavaScript to automate the act?--Validbanks 34 (talk) 02:49, 16 September 2009 (UTC)

Go for it. :) MFNickster (talk) 02:54, 16 September 2009 (UTC)

OK. I'm kind of busy, so maybe you can finish this. You'll paste it into your monobook.js file. I wonder how we'd collect the URLs of all of the pages in the category, though:

window.onload = replaceText;

var articleViewURL = document.URL;
var endString = "&action=edit";
var editURL = articleViewURL.replace("/wiki/", "/enwiki/w/index.php?title=") + endString;

function replaceText()
{
  if (document.URL.indexOf('action=edit') != -1)
  {
    var innerTxt = document.getElementById("wpTextbox1").innerHTML.replace(/UNIX/g,"Multics");
    document.getElementById("wpTextbox1").innerHTML = innerTxt;
    document.editform.submit();
    setTimeout(window.location = 'http://en.wikipedia.org/wiki/Special:Random',900)
  }

  else
  {
     window.location = editURL;
  }
}

Right now it just goes to random pages, though.--Validbanks 34 (talk) 03:02, 16 September 2009 (UTC)

Sorry I can't fix it, since I don't do JavaScript. However, rather than editing the info box, the proper approach would be to post a new section on each talk page titled "Multics family" and saying something like "I propose that we re-categorize this article under 'OS family: Multics,' for the reason that Unix is based on Multics." MFNickster (talk) 03:04, 16 September 2009 (UTC)

OK. Go for it, man.--Validbanks 34 (talk) 03:11, 16 September 2009 (UTC)

I started with the major ones: BSD, FreeBSD, NetBSD, Solaris, SunOS, and NeXTStep. Let's just sit back and see what people say. MFNickster (talk) 03:22, 16 September 2009 (UTC)

All right! Sounds good. But you missed Talk:IBM AIX, talk:UNIX System V, and Talk:HP-UX.--Validbanks 34 (talk) 03:31, 16 September 2009 (UTC)

UNIX System V doesn't have an info box, but I got the other two. MFNickster (talk) 03:36, 16 September 2009 (UTC)

Wait. What about Linux? http://en.wikipedia.org/wiki/Category:Linux_distributions. They also have the "UNIX-like" text in their info boxes.--Validbanks 34 (talk) 03:44, 16 September 2009 (UTC)

Oh, I forgot; I did Linux too. The score there is already 2-0 against. MFNickster (talk) 03:52, 16 September 2009 (UTC)

No, don't worry about it, man. You're doing great. I think you stumped them with that last post. But Linux isn't an operating system. It's just a kernel. So you might want to hit the major distros at Talk:Ubuntu (operating system) and Talk:Fedora (operating system).--Validbanks 34 (talk) 03:56, 16 September 2009 (UTC)

It's just wrong, I can also find other sources that say Unix is not "based" on Multics. The initial version Unix was written by the Multics developers [11]. Some of the Multics concepts were used in Unix (but not the code), but then Multics also borrowed other concepts in earlier operating systems. So using this logic, we then need to say that Multics is in the OS/360 family. Oh, and OS/360 borrowed concepts from even earlier operating systems, so every single operating system in the world is based on IBM 7090/94 IBSYS!

It is reasonable to say that Mac OS X is based on Unix because some of the code in OS X is from BSD Unix. Raysonho (talk) 05:47, 16 September 2009 (UTC)

OK. I think we have a consensus going that Mac OS X is based on Multics. I will change the entry tomorrow to match this belief. That's assuming, of course, that certain fascist elements on this site do not attempt to silence me first. I just got a message from someone who said that I'm being "disruptive." I guess free speech is kind of disruptive, but isn't it my right to edit an encyclopedia that anyone can edit? How can you have free speech if disruption is prohibited? I feel like I'm in Nazi Germany.

Anyway, trust me on this one, guys. I'm actually a neutral party. I don't use any of the above operating systems. I'm a Microsoft fan, so I'm probably the only one who can see this without any sort of bias!--Validbanks 34 (talk) 06:41, 16 September 2009 (UTC)

How can you say we "have a consensus going" when you are the only one who agrees with you? MFNickster (talk) 07:44, 16 September 2009 (UTC)

I oppose the Multics-like movement. Majority of reliable sources have OS X based off of UNIX, with no mention of Multics. 「 ɠu¹ɖяy 」¤ • ¢ 07:46, 16 September 2009 (UTC)

My mistake. MFNickster claimed he agreed with the statement that it was based on Multics. Thus, I will simply remove it--Validbanks 34 (talk) 07:54, 16 September 2009 (UTC)

Actually, I said you had a good reference, and proceeded to query other editors what they think of it. MFNickster (talk) 08:04, 16 September 2009 (UTC)

But you said, "I propose that we re-classify this article under 'OS family: Multics' in the info box, for the reason that Unix is based on Multics." That contradicts what you just said.--Validbanks 34 (talk) —Preceding undated comment added 08:25, 16 September 2009 (UTC).

I suggested that you do it yourself; since you didn't want to, I did it on your behalf. MFNickster (talk) 08:30, 16 September 2009 (UTC)

I suggest going through them all again and referring everyone back here. There are already fragmented discussions going on and really they need to be co-ordinated instead of perhaps twenty isolated discussions. I suspect you will have a hard time convincing the wider Unix community. There are more authoritative source than that cited, such as the Peter Salus book, which goes into considerable detail on this issue. Describing Unix as being based on or derived from Multics is probably overstating things: it is like saying MS-DOS is part of the CP/M family: it was an influence but not much more. In any case I would question whether Multics amounts to an OS "family". CrispMuncher (talk) 14:57, 16 September 2009 (UTC)

I oppose suggested change to Multics. Three reasons;

Debatable genealogy. If anything Multics is a "great aunt" of OS X. Focussing on this one OS is undue weight and ignores other ancestors in the family tree with equal or better claims.
Overwhelming weight of cites available state that OS X is Unix derived. If this is disputed among reliable sources, then cite the opposing claims and let the reader decide. If there are no reliable sources to be cited on this, it is a non-issue and shouldn't be on the article.
It is original synthesis. i.e. One cite says OS X is based on Unix, another says Unix is based on Multics, therefore OS X is based on Multics. A + B => C. Original Synthesis is not permissible.

--Escape Orbit ^(Talk) 17:21, 16 September 2009 (UTC)

Then, using that policy page as a guide, it is also original research to say that Mac OS X is part of a "UNIX family." All I've seen is marketing material from Apple that makes the claim that it is based on UNIX, not that it's a part of any family.--Validbanks 34 (talk) 18:51, 19 September 2009 (UTC)

Where is the evidence that Unix contains Multics code? The one source I can see here simply says Unix is based on Multics in the sense that the guys who created Unix previously worked on Multics and a lot of the decisions they made in designing Unix were based on their experiences working on Multics.

Where is there a single source that says Mac OS X has anything to do with Multics? How can you even suggest making such an edit given the blatant rule against original research? AlistairMcMillan (talk) 18:06, 16 September 2009 (UTC)

Utter nonsense. I haven't heard of "Multics family of OSes" this is WP:OR at its best. man with one red shoe 21:56, 16 September 2009 (UTC)

I suggest that Validbanks 34 (talk) is a troll and bad faith editor and should find somewhere else to peddle his self-indulgent time-wasting. 59.167.42.2 (talk) 22:04, 16 September 2009 (UTC)

I also oppose this. There is no "Multics family of OSes". Multics was a mainframe OS, Unix is intended for both servers and workstations. It originally ran on the PDP-7. It was inspired by Multics but was intended as a smaller, faster OS for non-mainframe machines. Multics never had a "family" and Unix was never considered to be a part of this non-family. This whole discussion is a waste of time, started by someone who is apparently working only from text and not from experience. That's the only way a glaring mistake like this could possibly be proposed. Yworo (talk) 16:05, 17 September 2009 (UTC)

But, surely, if we keep tracing back we'll find they all are based on original ideas from Colossus, and you can trace that back to some textile loom. And before that was the Abacus. Then there was the palaeolithic hand axe that started all this. So, I know, who can write some JavaScript to go through every single article and add a statement that the technology described is derived from palaeolithic hand axe... I think some people are winding each other up for some large-scale vandalism here. It's best stopped before one of them actually learns JavaScript, I fear. (I didn't link 'JavaScript' in case that gave them a head-start) --Nigelj (talk) 17:53, 17 September 2009 (UTC)

Why are we allowing all of these tainted sources to be used as citations? None of them say explicitly that Mac OS X is part of a UNIX family. One of them is a brochure from Apple and the other is a licensing agreement between Apple and the OpenGroup.--Validbanks 34 (talk) 18:53, 19 September 2009 (UTC)

You're splitting hairs - the family tree I posted above is certainly not a "tainted source"; it was compiled by Eric Levenez - a Unix "guru" who worked on kernel development. If "family" means a line of descendants with modifications, that is certainly where BSD, and NeXTstep, and Mac OS X belong. You still have not cited any sources to the contrary. Remember that Wikipedia articles should rely mainly on published reliable secondary sources. MFNickster (talk) 19:15, 19 September 2009 (UTC)

Who proofread his work? I only cite reliable sources that have been published by a neutral party.--Validbanks 34 (talk) 19:35, 19 September 2009 (UTC)

You haven't cited anything that supports your POV. If you ever do, then and only then will you be in a position to argue. MFNickster (talk) 20:09, 19 September 2009 (UTC)

You believe that I need to cite my claim that the aforementioned claim is improperly cited? I didn't know we needed citations to prove that something isn't cited.--Validbanks 34 (talk) 20:14, 19 September 2009 (UTC)

No, you just need to provide something - anything - in the literature that contradicts the sources already given in the article, then maybe you can add a note about how a notable author/columnist/researcher disagrees with the categorization. MFNickster (talk) 20:37, 19 September 2009 (UTC)

License

The breakout box at the top of the article says that OS X has a "proprietary EULA" license, but I can't find a link to the license itself. 06:00, 3 October 2009 (UTC)~ —Preceding unsigned comment added by 118.209.47.146 (talk)

Here it is: EULA, however I don't see any reason to link it in the article, it's not saying what the text of the license in in the infobox, rather what the type of license is.--Terrillja _talk 20:35, 3 October 2009 (UTC)

Requested move of Mac OS X v10.3, v10.4, and v10.5

I have posted a request for the Panther, Tiger, and Leopard articles to be moved from eg. Mac OS X v10.3 to Mac OS X Panther. If you'd like to weigh in on the discussion, it is taking place here. —INTRIGUEBLUE (talk|contribs) 06:37, 15 October 2009 (UTC)

Stacks

This article doesn't appear to mention Stacks. I was wondering where it would be appropriate to mention Stacks and place an image of Stacks with a fair-use rationale rationale for this article. --NerdyScienceDude ^{(talk to me)} 00:43, 9 December 2009 (UTC)

We have generally tried to avoid pictures so far, but go right ahead if you feel it's appropriate. Dravick (talk) 17:45, 9 December 2009 (UTC)

I'm not so sure that going into detail about a sub-feature of the Dock is the best thing for such a broad article. The feature was one of the major changes in Mac OS X Leopard though, so that's where it's discussed. Kinda like how Aero Peek is discussed in the Windows 7 but not in Microsoft Windows. Althepal (talk) 21:29, 9 December 2009 (UTC)

I decided to place it in the article Dock (Mac OS X). There was no mention of it there, and it was already covered in the Mac OS X Leopard article. --NerdyScienceDude ^{(talk to me)} 14:41, 10 December 2009 (UTC)

Good stuff. Just checked out the Dock article. I think it's funny how the criticisms there says that it's too large and that it only says the program name on mouse-over, two things that Microsoft adopted in the Windows 7 taskbar! Althepal (talk) 20:17, 10 December 2009 (UTC)

Mac os x 10.7?

I'm wondering: Is there any information about an upcoming os x? (or any os made by apple)? did apple again conceal everything? I can't imagine mabdul 16:42, 2 March 2010 (UTC)

Yeah, I forget where I read it so I can't link at the moment, but I read that some internet monitoring company reported computers logging on to sites with an OS code that if interpreted using Apples current OS numerical coding scheme translates to Mac OS 10.7. The site also said that the IP addresses of the computers running 10.7 matched up with known IPs from Apples headquarters in Cupertino. Which makes sense. So there is no official word from Apple, or any other details, but at this point it time it appears that it is indeed under development. I do think though that this isn't worth mentioning in the article. Its not much information, and what I had read only confirms what is obvious: Apple is going to make more operating systems. But when details from reliable sources come out, I'm sure they'll be added to the article accordingly.

Security?

I noticed Microsoft Windows has an extensive discussion regarding security, but nothing is here in this article. Does anyone think it would be inappropriate to add such a section to this article? ElBenevolente^(talk) 21:05, 4 March 2010 (UTC)

The German article has the section, too. Should we translate it? --89.204.153.65 (talk)

I looked at the German article about Mac OS X security and it's not very informative, so don't even bother.

--King of Dreamers (talk) 07:39, 28 May 2010 (UTC)

There should be some discussion of Mac security. It isn't invulnerable after all. 71.166.54.115 (talk) 20:55, 30 July 2010 (UTC)

GA nomination

I have nominated this article for GA. Everyone is welcome to make improvements to the article. NerdyScienceDude :) (✉ click to talk • my edits • sign) 14:43, 7 March 2010 (UTC)

Unfortunately the review is a disappointment. Hekerui (talk) 01:03, 11 March 2010 (UTC)

Talk:MacOS/Archive 14/GA3

Mac OS X only programmed in C?

The resumé box says that Mac OS X is programmed in C. Isn't it also largely programmed in Objective C and maybe also C++? —Preceding unsigned comment added by 87.64.57.97 (talk) 08:04, 2 April 2010 (UTC)

As far as I know, Mac OS X is hugely in C, while the exposed APIs (i.e. Cocoa) are in ObjC, with C++ filling special tasks like some sound stuff or video stuff. Yes there is some ObjC and C++, but it is mostly insignificant compared to the parts in C. Dravick (talk) 16:08, 2 April 2010 (UTC)

I don't think so. When I was a compiler engineer for Apple ca 2000-05, I had occasion to look at most of the source files (if only to figure out why the compiler didn't like them), and I would say that at the time, no single language was predominantly in use. Lower levels coming from NeXT and BSD are all C, higher levels coming from Classic Mac OS are all C++, higher levels coming from NeXT are ObjC. The form of the API doesn't necessarily tell you the underlying languages - part of the tangle in /System/Library derives from Cocoa APIs calling into classic Mac C++ code for some services, which in turn call into Mach for various reasons. Some of the mid-level subsystems are huge, dwarfing xnu for instance. Stan (talk) 20:43, 2 April 2010 (UTC)

Then again, I was never a compiler engineer for Apple. I was mostly thinking about the BSD and NeXT lower levels, but now I'm not sure I can guess right about how large that part is compared to others. Dravick (talk) 05:02, 3 April 2010 (UTC)

I assume we agree to include C++ and Objective C into the box? GoldRenet (talk) 14:06, 3 April 2010 (UTC)

Given that we don't seem to have a single reliable source on this, I'm removing it entirely. If someone finds a reliable source that we can cite, please restore and add the source. AlistairMcMillan (talk) 18:10, 3 April 2010 (UTC)

Yeah, hard to say if a reliable source is even possible, I'm sure I'm skating on the edge of my confidentiality agreement by saying as much as I have... :-) Stan (talk) 22:49, 3 April 2010 (UTC)

I've removed the source cited by User:Kkm010. For two reasons. Firstly the author of that source doesn't distinguish at all between Mac OS and Mac OS X. Secondly, he doesn't explain at all what he is using as sources and, as far as I'm aware, doesn't have a "reputation for fact-checking and accuracy". AlistairMcMillan (talk) 16:31, 4 April 2010 (UTC)

Neither the apple.com nor the Lextrait source are relevant - the languages supported by Xcode are for the benefit of external developers, internal developers don't necessarily even use Xcode. There's been a number of published interviews with Apple bigshots over the years, I'm sure at least one makes reference to OS X implementation languages, most likely around the 10.0-10.2 timeframe. Things might easily have changed too, I've been gone from Apple for over three years now; maybe they've rewritten all the C++ parts into ObjC! :-) Stan (talk) 12:28, 5 April 2010 (UTC)

I Think the programming languages should be removed until we have good and reliable references. GoldRenet (talk) 13:17, 5 April 2010 (UTC)

Good idea. The answer.com reference reads like it was written by a third-grader - "objective-C ... which is mac's version of C++" hahahaha... Stan (talk) 16:41, 5 April 2010 (UTC)

Another good resource is Amit Singh's Mac OS X Internals[12] Given that Apple's materials indicate "The Cocoa frameworks are primarily written in Objective-C"[13], "the I/O Kit framework...is written in a restricted subset of C++" and "the Carbon interfaces are written in C" [14] I don't see any problem with listing C, C++ and Objective-C as (at least some of) the languages used to build Mac OS X. MFNickster (talk) 00:32, 6 April 2010 (UTC)

Also: CoreAudio "Frameworks are implemented in C and C++ and present a C-based function API." http://developer.apple.com/audio/pdf/coreaudio.pdf] MFNickster (talk) 00:59, 6 April 2010 (UTC)

Excellent that you found that! The second and third reference you give (the one about Cocoa and the one going to the PDF about the technology overview of Mac OS X) seem to be enough proof to me. I wouldn't use the first reference, as that's from 2003 already, nor would I use the Core Audio reference as that's from 2001. GoldRenet (talk) 08:52, 6 April 2010 (UTC)

The CoreAudio reference may be from 2001, but there is no newer version that I could find - it appears to be the most current available. MFNickster (talk) 04:03, 7 April 2010 (UTC)

I actually reviewed my copy of Singh yesterday, and decided it wasn't a reference for the closed-source parts of the system; Singh carefully limits himself to describing only the sources of open-source pieces (for which IOKit is indeed an example of extensive C++ usage). The line about "Carbon interfaces are written in C" refers specifically to the public header files. As a reality check, just do this on a stock Leopard system: "nm /System/Library/Frameworks/Carbon.framework/Frameworks/HIToolbox.framework/HIToolbox | grep _Z | c++filt | grep ::". You should see about 9000 symbols that are clearly from C++ code. If you look at the whole symbol list, there are also a great many ObjC symbols, per my point above about how the different modules intermix code these days - especially ironic in this case, since HIToolbox is descended from the original Mac toolbox that was written in asm and Pascal. Stan (talk) 15:29, 6 April 2010 (UTC)

I would remove all current references (apple.com/macosx/developers/; Lextrait and Answers.com --> that one is just too ridiculous :p) and replace them with two of the ones that MFNickster mentioned:

http://developer.apple.com/Cocoa/overview.html

As MFNickster mentioned, the above reference says that the Cocoa libraries are mainly written in Objective-C.

And better:

http://developer.apple.com/mac/library/documentation/MacOSX/Conceptual/OSX_Technology_Overview/OSX_Technology_Overview.pdf

I went through the document and I think it has enough statements that C, C++ and Objective-C are used for components of the Mac OS X-system.

Some examples of which some have already been mentioned by MFNickster:

"the Carbon interfaces are written in C";

"Core Animation is a set of Objective-C classes";

"Image Kit framework is an Objective-C framework";

"For the most part, the interfaces of the Core Audio frameworks are C-based, although some of the Cocoa-related interfaces are implemented in Objective-C";

"the QuickTime Kit provides an Objective-C based set of classes for managing QuickTime content";

"the Input Method Kit is an Objective-C framework";

"Although it is written in Objective-C, you can use the classes of the PDF Kit in both Carbon and Cocoa applications";

"It is written in a restricted subset of C++. Designed to support a range of device families, the I/O Kit is both modular and extensible";

etc.

Do we agree? GoldRenet (talk) 16:35, 6 April 2010 (UTC)

Please correct me if I am wrong, but saying something has an Objective-C interface or something is an Objective-C framework or is a set of Objective-C classes, doesn't prove the thing is written in Objective-C does it? That just means you are supposed to use Objective-C to talk to it, right? AlistairMcMillan (talk) 17:43, 7 April 2010 (UTC)

That's right. However, it would be inordinately difficult to build the outer layer of an ObjC framework in anything other than ObjC, and in practice Cocoa frameworks are ObjC. The Darwin part of the system obviously uses loads of both C and C++, so all we really need is an interview with any Apple exec or senior engineer where they say that Cocoa frameworks are mostly written in ObjC. Stan (talk) 19:16, 7 April 2010 (UTC)

The Cocoa (API) article itself notes that the frameworks are "written in Objective-C." MFNickster (talk) 20:09, 7 April 2010 (UTC)

Yes. But the Cocoa overview on developer.apple.com specifically says: "The Cocoa frameworks are primarily written in Objective-C". Additionally and as I mentioned before, the Mac OS X technology overview PDF has two clear statements: 1) "Although it is written in Objective-C, you can use the classes of the PDF Kit in both Carbon and Cocoa applications" and 2) "Darwin offers an object-oriented framework for developing device drivers called the I/O Kit framework. This framework facilitates the creation of drivers for Mac OS X and provides much of the infrastructure that they need. It is written in a restricted subset of C++. Designed to support a range of device families, the I/O Kit is both modular and extensible.". GoldRenet (talk) 10:10, 8 April 2010 (UTC)

OK, the developer.apple.com overview looks like a good source, no waffly words there. So we have cites for three languages, and can legitimately mention all of them, but *not* make any assertions about relative amounts of each. Stan (talk) 15:45, 8 April 2010 (UTC)

All right, just added the references! GoldRenet (talk) 08:46, 9 April 2010 (UTC)

The Objective C runtime reduces all ObjC to C function calls. So anything Objective C is, at bottom, C. — Preceding unsigned comment added by Pantergraph (talk • contribs) 13:06, 29 December 2010 (UTC)

And the original cfront C++ compiler emitted C code, so, with that compiler, anything C++ was, at bottom, C. That's also true of anything C++ when compiled by Comeau C/C++. And the C, C++, and Objective-C compilers reduce all C, C++, and Objective-C code to machine code, so anything C, C++, or Objective-C is, at bottom, machine code. It's not clear that any of that is actually interesting in this context, however. Guy Harris (talk) 21:46, 29 December 2010 (UTC)

There is nothing interesting on wikipedia. This is just to waste time — Preceding unsigned comment added by Pantergraph (talk • contribs) 15:54, 29 December 2010 (UTC)

Initial release

i searched the archives for this before asking, but..is there a particular reason the initial release in the infobox is listed as 24 January 1984 (1984-01-24), with the wikilink to mac os? to me this should conform to the standard set by other similarly detailed articles, as per Windows XP, Windows 7, etc., in that it should be the release date of os x, not the mac operating system in general. if someone wanted to know the initial release date of the entirety of windows, it can be found under the category box..same should go for the release date here (as in the table further down the page). Impasse 18:42, 29 April 2010 (UTC)

Mac OS X is a different sort of OS than Mac OS, so I have changed it to the release of Mac OS X 10.0. You cannot compare the Mac OS (X) releases with the Windows releases though. I consider Mac OS X to be a series of OS'es on itself. If someone disagrees, feel free to discuss. GoldRenet (talk) 07:42, 30 April 2010 (UTC)

Discussions from Marklar project

I am redirecting Talk:Marklar project here since Marklar project redirects here, so I have moved the following conversations here from its talk page. - EdoDodo ^talk 17:39, 1 June 2010 (UTC)

I changed OSx86's description a bit, the old version gave implications like OpenSuse's community edition - a commercial product being freely released to the people, which the Developer Transition Kit versions of OS X certainly are not. --Niteice 22:08, 13 November 2005 (UTC)

Also - was Rosetta included all the way? I'm pretty sure QuickTransit has only been around since 2004, making it impossible for Apple to have it ready until they decided to switch. Someone with more knowledge on that (if any) should fix it. --Niteice 22:10, 13 November 2005 (UTC)

Does anyone know if the name a South Park reference to Marklar? Rehevkor 01:55, 10 February 2006 (UTC)

Why nothing on 10.7?

Why is there no placeholder for 10.7? Also, the next release ought to be called Lion, since there is a lion on the poster for the October 20 event. That will likely not be the only tidbit of information of the next major release of one of the worlds most common OSs. Bonus pater familias (talk) 18:12, 13 October 2010 (UTC)

Hold on your horses! The "Back to Mac" event was announced today, and there is nothing official yet. It is unclear that Wikipedia should talk of rumors. It would look pretty stupid (though admittedly unlikely) if the event is not about a new release. Ratfox (talk) 18:59, 13 October 2010 (UTC)

Wikipedia shouldn't be talking about rumours really, as it is an encyclopaedia. They do seem to have confirmed that 10.7 will be previewed, but there's nothing but speculation about the name so far. As someone has pointed out on The Register, it could just as easily be panthera leo nubica, etc. Gkmotorsport (talk) 19:46, 13 October 2010 (UTC)

Quite simply, WP:CBALL. Ms. Cleo doesn't work here.--Terrillja _talk 20:10, 13 October 2010 (UTC)

Memory allocation limitation

Hey techies, i ran into this one at the Blender web site, where i read that "OSX only assigns processes a memory space of 2 GB", and "only uses the 2nd half of the 4 GB range". Where in the various OS X articles would this fit ? --Jerome Potts (talk) 19:28, 20 December 2010 (UTC)

I think this is a limitation of non-Intel versions of Mac OS X: Mac OS X 10.0 up to Mac OS X 10.3, and Mac OS X 10.4 and 10.5 when run on PowerPC CPU's (as you probably know, Mac OS X 10.6 only runs on Intel CPU's). Intel versions of Mac OS X use the PAE-system, so a 32-bit computer can use more than 4 GB of RAM (this technique supports up to 64 GB). As a consequence I think (I'm not 100% sure) this not only gives all your running processes together more than 2 GB, but can even allocate up to 4 GB per running process.

The article you refer to is from February 2006, so I presume they use a PowerPC system, most likely a PowerMac.

Still, looking at the wider picture of 32-bit, this issue is not something specifically notable about Mac OS X in itself, but about 32-bit in general. Therefor, the way Mac OS X handles RAM on PowerPC systems would suit better on an article about the limitations of 32-bit processing. Gold Renet 15:34, 27 December 2010 (UTC)

PAE has nothing to do with how the virtual address space is laid out, it only affects how much physical memory a 32-bit x86 system can access. An OS using PAE could, for example, put kernel code and data into the upper half of the address space and leave only 2GB for user-mode code and data - and an OS not using PAE could keep the kernel code and data (or, at least, most of it) in their own address space and allow up to 4GB of user-mode code and data (even if there isn't enough physical memory for 4GB).

But, no, none of that is OS X-specific; Windows also puts the kernel-mode code and user-mode code into the same address space, and it can either split the address space into two 2GB chunks or into a 3GB user-mode chunk and a 1GB kernel-mode chunk, depending on a boot-time flag. I'm not sure what Linux does, but I think it might also keep the kernel-mode and user-mode code in the same address space; I don't know what Solaris, for example, does.

And none of this applies, of course, when running 64-bit user-mode code. Apple no longer sell 32-bit x86-based machines, and I suspect the only 32-bit x86-based desktop/notebook machines you can buy these days are netbooks - I think the only 32-bit x86 processors Intel sells are Atom processors and embedded x86 processors, and I don't know whether AMD sell any 32-bit x86 processors at all.

In any case, I don't think the information from that Blender article is of enough interest that it belongs in an OS X article. Guy Harris (talk) 20:20, 27 December 2010 (UTC)

Thank you for the explanation! Gold Renet 08:31, 28 December 2010 (UTC)

and/end users??

Do you mean "Apple end users" here: "Mac OS X v10.6 is usually referred to by Apple and users as "Snow Leopard"." ?? —Preceding unsigned comment added by 62.199.181.214 (talk) 13:53, 2 January 2011 (UTC)

No. What it means is that Apple as well as its users use the name "Snow Leopard". Gold Renet 15:58, 2 January 2011 (UTC)

Would "Mac OS X v10.6 is usually referred to by Apple and by users as "Snow Leopard"" be clearer? Guy Harris (talk) 23:09, 2 January 2011 (UTC)

"Usually referred to" is a strange way to put it. It would be clearer just to say "Mac OS X v10.6 is named 'Snow Leopard'" or "carries the moniker 'Snow Leopard.'" MFNickster (talk) 01:02, 3 January 2011 (UTC)

That depends on whether the sentence in question is intended just to say "the OS releases have, in addition to a version number, a "big cat" name", or whether it's intended to say that and that the "big cat" name is what's usually used to refer to it. In the former case, "Mac OS X v10.6 is named 'Snow Leopard'" suffices; in the latter case, "usually referred to", or something such as that, should be used. Guy Harris (talk) 01:12, 3 January 2011 (UTC)

Ah, okay. In that case, it would make sense to say it is "usually referred to simply as 'Snow Leopard'," but that would be a difficult assertion to back up without a source. MFNickster (talk) 02:17, 3 January 2011 (UTC)