Apple Losing Touch With the OS Community?

InfoWorldMike writes to tell us that InfoWorld's Tom Yager recently had the chance to sit down and chat with Apple about their closing the OS X Kernel. From the article: "The Mac platform is an overflowing basket of raw materials for innovators and creators of all stripes. It's what Steve Jobs would fantasize about if he still worked out of his garage, and you can bet that he'd be livid to find that the vendor locked some portion of his chosen platform behind a gate without a word of notice or explanation."

Part Deux

[American AC in Paris]

This is a follow-up to one of Tom Yager's earlier editorials [macworld.co.uk], which was discussed recently on Slashdot. [slashdot.org]

Oh, and nice headline. I'd even go so far as to call it a sensational headline. You get a slow clap.

Jobs upset?

[TheWanderingHermit]

I doubt Steve Jobs would have been the one to get upset about thing being closed off, since very little of the actual innovation, creative, and design work ever was his. I can see Steve Wozniak getting ticked off about it, but I imagine he'd hack away anyway.

Jobs may be great at pushing the designers to do more, but he was NOT the one who did most of the hacking. He even exploded when Woz asked if he could help with the Apple's analog port.

Re:Jobs upset?

[ultramk]

Honestly, I don't think either one would have given a damn about this: they would have been too busy creating their own system from scratch. Neither one has ever been content to play in someone else's sandbox, and I can't picture them whining about someone else's system being closed off to them.

FWIW, I believe that both Steves were necessary for the creation of Apple as a world-changing phenomena. Steve W. was (and still is) the prototypical alpha geek, who views a technical challenge as a personal quest, and who doesn't "work" on a problem, he plays with it. Steve J. had the vision of a world where technology was put in the hands of regular people, and knew enough to make seemingly impossible demands from the people who worked with him, and for him. The kinds of demands that once they were met, resulted in a revolution. I've briefly met both men (at different times), and I have deep respect for what each brought to the table in that fateful partnership.

Vision without ability is neutered. Ability without vision is sterile. The one thing both Steves have in common is the refusal to accept the idea that something's impossible.

m-

Get things straight...

[Chanc_Gorkon]

Apple hasn't released the kernel for x86...YET. See this forum post from a Apple Employee:

http://lists.apple.com/archives/Fed-talk/2006/May/ msg00105.html [apple.com]

In my opinion if I had to put companies on a list, Apple would stil be high on my openess with developer list. At least Apple has all of the developer tools (Xcode and others) free for the taking. You still have to pay Microsoft to write programs for windows unless it's a batch file.

Re:Surprising if true.

[AKAImBatman]

The platform sells well to people who want a Unix, not people who want to hack the OS.

What about people who want to hack up an OS using a Mac? (raises hand) Believe it or not, cross compiling to the x86 platform using a PPC Mac and QEMU actually works. It's actually a better development environment than Windows, because you don't have to work around Windows' lack of Unix tools.

If you're weird like me, check out the OS FAQ [mega-tokyo.com] for information on creating your own operating system, including the building of a cross-compiler. Bonefide [osdever.net] also has some great tutorials on getting going with your operating system construction project.

Only official Apple response

[daveschroeder]

To sum up Apple's objections, they felt I had given a year-old story a fresh coat of paint and sensationalized it for an audience that wasn't affected by it.

Yep, that pretty much sums it up.

To date, the only official response [apple.com] has been:

Just to be clear, Tom Yager was *speculating* about why we have -- so far -- not released the source code of the kernel for Intel-based Macintoshes. We continue to release *all* the Darwin sources for our PowerPC systems, and so far has released all the non-kernel Darwin sources for Intel.

Nothing has been announced, so he (and everyone else) certainly has the right to speculate. But please don't confuse "speculation" with "fact."

Thanks,
-- Ernie P.

Ernest N. Prabhakar, Ph.D. [phone]
Product Manager, Open Source & Open Standards; Mac OS X Product Marketing
Apple Computer; 303-4SW 3 Infinite Loop; Cupertino, CA 95014


and a response to a private message I sent:

Date: Mon, 22 May 2006 12:08:45 -0700
From: Ernest Prabhakar
To: Dave Schroeder
Subject: Re: [Fed-Talk] Apple [may not open] OS X Kernel for Intel

Hi Dave,

On May 21, 2006, at 11:41 AM, Dave Schroeder wrote:

When *will* something regarding a xnu source release on x86 be announced?

I know you probably can't answer this, so it's somewhat of a rhetorical question, but seriously: the lack of release of source for xnu on x86 represents a significant change in strategy to some customers with no corresponding announcement or roadmap. When will concerned customers be informed as to what is happening?

Generally speaking, when a final, irreversible decision has been made, we will find
_some_ way to let affected customers know about it.

If nothing else, the very fact I am telling you to *not* assume that something is true,
means *I* don't believe it is true. :-)

-- Ernie P.

Seriously, might there be kind of a, you know, huge developer conference coming up in a month and a half or so here where some of these questions might be answered? Especially since Tom Yager's speculation is just that - speculation - and extremely old news [slashdot.org] at that? Is it any wonder that both of Yager's articles are under "Opinion" headlines?

Fraction of a Fraction sounds about right

[Enrique1218]

I don't really see the issue about the kernel and everything. I am geek. Moreover, I am and Apple geek. The OS satifies my needs in current incarnation. I have never saw the need to recompile the kernel for this OS. I recompile for Linux more out of necessity than any masochistic desire. I rather not be bother with it which is why OSX is my primary OS.

Re:Personally

[sakusha]

I don't see MS locking down wma anywhere CLOSE to the way Apple has locked down AAC.

Open your eyes. AAC isn't locked down, it's an industry standard. You can use AAC without using DRM, but even if you do use DRM, it runs equally well on Windows and Mac systems. On the other hand, WMA protected files run only on Windows.
Oh, I see, you meant the iPod and AAC. I don't have a single AAC file on my iPod. I guess that's not locked down either.
Or maybe you meant the iTunes Music Store. You do realize you can burn those files to AIFF/WAV files on CDs and turn them into other formats, right?

Oh well, I guess there's no sense explaining this to someone who has their eyes shut.

Re:Proprietary != OSS

[Hungus]

Apple put no copy protection on the OS that I am aware of. In fact there is a family license for about 20 bucks more or so that gives you license to install the os on 5 machines. On the workstation version there is no serial number at all, just for the server version.

Re:Only official Apple response

[daveschroeder]

Tom isn't "speculating", it's quite honestly a fact that Apple has not released the source to Intel XNU.

To date, yes. I will 100% agree that the Intel xnu source is currently closed. However:

- Intel xnu source hasn't always been closed
- PowerPC snu source is still open
- The change happened with Intel-based Macs began shipping
- He is speculating as to the *reason* xnu on Intel isn't currently open source
- The implication in Yager's articles is that because it's closed now, it must/might be closed permanently

Also, in Yager's first article [infoworld.com], he uses the title:

Apple closes down OS X

Excuse me, but when was "OS X" ever open? And since when does one component on one architecture being closed constitute everything being closed, especially when all non-kernel sources that have been traditionally released to date continue to be released.

The first sentence is:

Thanks to pirates, or rather the fear of them, the Intel edition of Apple's OS X is now a proprietary operating system.

Again, huh? First, Mac OS X has always been a proprietary operating system. Nothing has changed. Second, all of the Darwin sources are still released [apple.com] on both architectures. With ONE, admittedly large, exception: the kernel (xnu) on x86.

The problem is exactly as Apple framed it in Yager's followup article:

- Yager presented this inaccurately and sensationalistically, making it seem to a broader audience as if "OS X" itself was previously "open", and is not "closed"
- Yager does not discuss the nuance of what the kernel being closed means from a practical standpoint
- Yager incorrectly asserts this somehow matters more now because Intel-based servers will be coming, because people who buy servers and equipment for enterprise will somehow have needs to use the kernel source, but Apple has been selling into this marketplace for over 4 years, and the fact that the server platform will be on Intel changes none of that
- Yager generally makes it seem like this "matters" to ordinary users in a broader audience

In any case, get your story straight. Either this "doesn't matter", because "nobody needs source code anyway", or "Apple has hit a problem releasing the source code but will do shortly, but cannot dare say such a thing in public because, erm, yeah, RDF! RDF! Our refusal to release source needs no justification, it "just works". Insert hypnotoad here".

There's no logical inconsistency in anything I've said, either here, or previously [slashdot.org]. Of course it matters. It matters to me. It matters to the people who actually want or need the source, which is an extremely small subset of Mac OS X users. (And no, users who don't even know what a kernel is don't receive a substantive benefit from others outside of Apple being able to see the kernel source.)

I'm tired of hearing pretty much every excuse from the insulting to the flat-out false. Maybe they will release XNU for Intel in the near future. Hey, guess what, MAYBE MICROSOFT WILL RELEASE THE SOURCE TO WINDOWS IN THE FUTURE TOO! Yeah, that's it! We can all start describing MICROSOFT as a FUCKING OPEN SOURCE COMPANY because they MIGHT release the source code under the GPL in a few hours!!!

Wrong.

The source code for Windows has never been open; the argument is not the same.

The source for xnu has been open, continues to be open on PowerPC, and is available in an earlier incarnation for x86 (parity with Mac OS X 10.4.0). Therefore, saying that a final decision might not have been made on current iterations of xnu on x86 is perfectly reasonable.

Further, if anything, MORE source is now released than previously: x86 sources for all non-kernel components are released with parity with Mac OS X releases for PowerPC and x86; previo

No official word; delay != closed

[raga]

If indeed it has been closed, it is going down the wrong philosophical path, and Steve Jobs should rethink the strategy on this (loss in revenues from pirating vs. goodwill from OSS.) But then I have not seen Apple indicate that it will be closed.

He actually has a desire to recompile the kernel and not get ad hits as far as this article appears.

I remain skeptical. How many Mac users are there who are recompiling xnu (not just to blog about it on Infoworld etc., but because they have actually studied it, and improved upon it?)

So all you xnu hackers, please stand up and be counted. As I recall, only three people even came close to understanding the kernel bug in the The Mac OS X Expert Challenge [kernelthread.com]. As previously reported on slashdot, "Also looks like other than these guys, nobody got anywhere with the problem." [slashdot.org]. So much for the myth of the legions of ubergeeks working with Macs.

Even the parts of the OS that are most useful for tinkering were an afterthought for Apple.

You lost me on that. The part of the OS that is "most useful for tinkering" is xnu, the kernel. For those who care, the only missing piece right now is xnu; it has not been updated - there is no "afterthought for Apple" (yet).

FWIW, xnu from Darwin released before the Intel switch six months ago is still available for Infoworld blogers to recompile to their heart's content.

cheers- raga

Re:Personally

[94229a]

AAC [wikipedia.org] is an ISO Standard. Really. The licensing terms can be found here [vialicensing.com].

I am quite surprised that so few other MP3 players support it. Some Sony-Ericcson phones [sonyericsson.com] do, as do Nokia [wikipedia.org], Motorola [wikipedia.org], Siemens [siemens.com] and the Sony PSP.

Maybe the reason why Creative, iriver, Rio and others don't support AAC is because they are too busy supporting proprietary formats like WMA.