Darwin 6.0.2 for x86 Released

Jos Louis writes "Apple has released the x86 version 6.0.2 of Darwin, the FreeBSD-based core of Mac OS X. You can download the bootable ISO on Apple's site."

Comments

Not a troll

[IamTheRealMike]

Could anybody give me a rundown of why you'd want to run this on your PC over say RedHat 8, which is also available for free, and under the GPL to boot?

Also, what is the hardware support like?

Re:Not a troll

[makisupa]

Two words, one hyphenated:

Multi-boot fetishists.

Re:Not a troll

[Anonymous Coward]

If you want to test server apps for MacOS X you develop but don't have a Mac, it's a good option.

Re:Not a troll

[bogie]

You wouldn't. When I asked this question before the answer I got was this is just Apple "Giving back" what they could.

I guess it's better than nothing, but I'd prefer they give back some of the interesting stuff, but that's never gonna happen.

Re:Not a troll

[Arkham]

I guess it's better than nothing, but I'd prefer they give back some of the interesting stuff, but that's never gonna happen.

What about Darwin Streaming Server and Rendezvous? What about OpenPlay and OpenDirectory? Apple is releasing lots of neat projects as open-source. This just doesn't happen to be one of them.

Darwin is great for OSX developers because it lets you look into the source of OSX and see how it works. It's not particularly useful as an x86 platform

Re:Not a troll

[Lussarn]

Exactly, apple wants to play nice with the opensource community and releases a kernel. Something the community got tons of allready, and very good ones too. If there where a penny to earn in apples kernel they wouldn't release that either, that is if it where better than Linux/FreeBDS etc.

They have some programs which clearly are better than what the opensource community has come up with but none of this is open source.

I don't think thats fair, and I think that matters.

Re:Not a troll

[Jezza]

Err... I'm guessing that to a lot of Unix guys - the kernel doesn't really matter a jot. Afterall we use the Unix calls and as long as it does what the man page says it should that's fine. So a lot of this is redundant.

But there are a lot of reasons you might want to use Darwin over GNU/Linux. You can add X to it, and compile up lots of software, from three feet away who'd guess it wasn't Linux or some other form of BSD?

If that's true (and I think it is) then why is this less useful than Linux?

Re:Not a troll

[Anonymous Coward]

If you want to test server apps for MacOS X...

...you get a mac.

Seriously tho; if you're writing software for the mac, don't *assume* it will work on a mac if you did all your designing, testing, updating on a DIFFERENT ARCHITECHTURE! (sorry for screaming)

Re:Not a troll

[bnenning]

Umm, a streaming server for a closed and proprietary media format

No. The QuickTime format is completely open and documented. It just happens that the most common codecs used with it are closed (Sorenson) or patent-encumbered (MPEG4). There's nothing stopping anyone from writing their own codec and having QuickTime support it.

Re:Not a troll

[proj_2501]

As the Ogg Vorbis folks have done. There are QuickTime extensions for Ogg Vorbis, albeit still not quite ready for production.

Re:Not a troll

[Inthewire]

There's nothing stopping anyone from writing their own codec and having QuickTime support it.

That's a relief! I thought the fact that I know dick about programming would have stopped me. Good to know that isn't gonna be a problem.

Re:Not a troll

[SumoRoach]

obviously, that's the safest path, but not always practical. Buy cheap x86's for your developers, but test on the final platform, so you don't *assume* that it will work on a DIFFERENT ARCHITECHTURE (sic).

Re:Not a troll

[multiplexo]

Hmmmmm, yessss, multi-boot fetishism. Black leather stilettos that lace up to the crotch. Cowboy boots with spurs, Chrome studded biker boots, FreeBSD, Win98, OS/2, BeOS. Ohhhhhhhh, I'm so hard now.......

Re:Not a troll

[benedict]

DSS can also serve MP3, according to its FAQ.
And although the FAQ doesn't mention it, I think
you can use DSS to serve RealMedia and other
formats as well.

As for OpenPlay, maybe they finished it. :-)

Apple released Darwin as open source in order to
help Mac OS X developers who are interested in
understanding and possibly closely integrating
with the OS. They're also hoping to get some free
maintenance and development. Anything else, like
a benefit to community relations, is gravy.

Nobody uses Mac OS X for its ideological purity.
But it is pretty darn open for a commercial
desktop operating system.

Re:Not a troll

[benedict]

Does the open source community have anything that
does what Rendezvous does, but better? Sun tried
with JXTA, IIRC, but I don't think it caught on.

Even if you were right, what exactly is unfair
about Apple keeping their good stuff to themselves?
What is it with Linux weenies that they think
people can pay software developers hundreds of
thousands of dollars a year and then give away the
results? Are you mad because Apple uses free
software? I have news for you: they abide by the
licenses of that free software. BSD developers
*want* commercial entities to adopt and improve
upon their software, even if the result isn't
open source -- that's why they use the BSD license.

Where's the source?

[Anonymous Coward]

There is no source. The source remains unreleased because the source does not exist. The real story is the part that Apple is not telling us: The fact that they have evolved an organic "force" capable of developing a kernel directly, as executable machine code, without human intervention. The implications are terrifying and profound.

Why do you think they called it.... DARWIN?!

Re:Where's the source?

[mr_z_beeblebrox]

The real story is the part that Apple is not telling us: The fact that they have evolved an organic "force" capable of developing a kernel directly, as executable machine code, without human intervention.

The part you were unaware of, is that MS is miles ahead of them as Outlook 2002 can automatically run organic machine code without human intervention.

Re:Where's the source?

[Anonymous Coward]

There is source, you just have to use cvs to get it, or download all the components separately via the html frontend somewhere on the Darwin pages @ apple.

Re:Where's the source?

[tswinzig]

The part you were unaware of, is that MS is miles ahead of them as Outlook 2002 can automatically run organic machine code without human intervention.

The irony here is that you are making fun of the first version of a Microsoft email program that does NOT give users access to executable attachments, and does NOT let an outside program use it to send email without approval gained from a popup window.

Re:Where's the source?

[Ilan Volow]

That's no AI. That's Clippy. He's been running in spite of human intervention for years.

Re:Where's the source?

[Anonymous Coward]

The irony here is that you are making fun of the first version of a Microsoft email program that does NOT give users access to executable attachments, and does NOT let an outside program use it to send email without approval gained from a popup window.

But it does eat babies.

Really! You can test this yourself. If you or someone you know has a computer with Outlook 2002 installed, leave it on overnight, and watch. At some point between midnight and 2 AM (what exactly triggers this feature is as of yet unknown, although the time of activation seems to oscillate within its two-hour window with the phase of the moon) Outlook 2002 will climb out of the computer, skitter to a window, climb outside, and hunt the streets until it finds a house containing a baby, at which point it will enter, consume the baby bloodily, and return to your computer. If during the entrance or exit of either your or the victim's house it encounters a locked window, it will shatter the glass and continue.

This behavior has been independently confirmed numerous times since the release of Outlook 2002, and yet so far Microsoft has declined to comment on it. I suggest you go read the rather long bugtraq thread on the subject.

Re:Where's the source?

[ncc74656]

The real story is the part that Apple is not telling us: The fact that they have evolved an organic "force" capable of developing a kernel directly, as executable machine code, without human intervention.

Considering that Steve Wozniak wrote Integer BASIC without the aid of an assembler, I'd guess that anything is possible...:-)

Re:Where's the source?

[.pentai.]

Changed in the registry, which means one of those hundreds of worms out there that get tossed from person to person can (if run) make this registry change, and then commence to send itself out to everyone (perhaps even changing the registry back afterwards, to hide itself better?)

Re:Where's the source?

[spongman]

you could always right-click on it, selection 'options' and uncheck the check-box that says 'use the Office Assistant'. not that hard, really.

for some.

Re:Where's the source?

[dylan_-]

But it does eat babies.

This is only the Default behavior by the way - this can be changed easily with a small change in the registry. I had to enable it for a company I did some work for because their custom software uses Outlook to automatically eat babies in the background. Poor Bastards.

(Sorry Tassleman ;-) )

It's getting closer

[boy_afraid]

It been rumored that Apple would bring OS X to the x86 and has a working version in secrecy under penalty of death to whomsoever relveals it. Apple knows that they have something that the Linux/Windows geeks really want, just like we want the iPod to work under Windows.

Steve Jobs is not smoking crack, just weed.

Re:It's getting closer

[derch]

and what crack are you smoking? People had hacked the iPod to work with Windows shortly after it's release. To boot, Apple now sells a Windows version.

PS - Even if Apple moved OS X to the x86 family, you'd still have to buy an Apple PC.

Re:It's getting closer

[Jezza]

And miss out on AltiVec, many of the iApps (iMovie, iDVD) need AltiVec to run at a reasonable speed. They would be far too slow on an Intel or AMD based system. This is of course deliberate on the part of Apple, to play to their strengths (AltiVec in this case).

Of course iTunes is also optimized to make heavy use of the AltiVec, but MP3s don't actually stress a modern processor at all so AltiVec isn't important here.

Re:It's getting closer

[kitzilla]

> Apple knows that they have something that the Linux/Windows geeks really want...

I'm a fan of Mac, but I don't know that the above statement is true. SOME people might want OS X running on PC hardware. I'd rather see it run well on Mac boxes. We have Linux for x86.

Re:It's getting closer

[dildatron]

And I think an Apple PC would be cool. It would provide a little more flexibility, and probably a pretty big cost savings. You would have to use limited hardware, but still, things like processor speeds could vary widely.

Re:It's getting closer

[pauljlucas]

And I think an Apple PC would be cool. It would provide a little more flexibility, and probably a pretty big cost savings.

How would it provide more flexibility, exactly? If Apple switched to x86, all that would change would be the CPU plugged into the motherboard. The rest of the G4 case and components would be exactly the same. That being the case, the cost savings would be nil.

Re:It's getting closer

[be-fan]

The only reason the iApps need AltiVec is because that is the only reasonably fast part of the G4. Plain C/C++ code optimized by a good SSE-enabled compiler (Intel C++) would perform just fine, especially given the fact that none of the iApps do anything that would take more than a (by today's standards) ancient 1 Ghz x86 proc anyway.

Re:It's getting closer

[be-fan]

The G4 case is probably around $100, and all the other components aren't that great. What isn't mediocre (hard drive, graphics card, etc) is crap (sound card, speakers, etc). The only nice bit is the GigE card, which is less than $50 on PriceWatch.

Re:It's getting closer

[Jezza]

No it's not, it's different - sure it has different strengths and might even be better for different jobs (this isn't a my Mac's better than your PC rant) but the way that Apple have solved these problems are deeply AltiVec.

That's also true for BLAST, and why the work was done on the PPC. G4s just have the edge over x86 in this. it's also true for Photoshop (that's why Apple often bring that to the demo table so often) it is written to make best use of whatever platform and it always does better on the Mac.

Re:It's getting closer

[Jezza]

Look at encoding video so it can be written to a DVD (and played on a normal player) this is one example where the AltiVec blows x86 away. Don't forget iDVD needs to do this, also the video transitions in iMovie are another area where the AltiVec helps a great deal, and extra speed is helpful. I agree that there are problems out there where the AltiVec isn't at all useful, but a suprising number where it is. The Mac hardware uses other "tricks" to get better performance - Quartz Extreem, makes use of the compositing in the GPU to lighten the load on the CPU.

The G4 also has a very short pipeline that helps on "normal" code as well. Lastly all the Mac Towers use 2 G4s, making use of the fact that Mach is well optimized for SMP.

Re:It's getting closer

[WatertonMan]

Once again I'm just going by how long between 10.0 and when all the Apps started coming out from the major players. Several key applications still haven't come out in OSX and many applications have only come out the past couple of months. And that was with those companies having betas long before 10.0 came out.

With the switch to a new processor you can expect many more things to break. Apple undoubtedly can recompile all their iApps. The issue is third party companies. (Especially MS Office)

I think Apple has been hedging their bets because of problems with Motorola. Thus they've been compiling much of Darwin on both systems and likely Aqua as well. But to run them on x86 will undoubtedly create all sorts of problems. Not to mention optimization problems.

It might be nice in the long term, but consider what the initial user experience will be those first six months to one year. . .

Further most companies will be forced to develop for both PowerPC and x86 version of OSX.

Unless IBM falls behind schedule I just can't see Apple being able to afford the costs of switching to the x86. The real problems of the PowerPC chip in terms of competitive speed though means Apple has to be able to if necessary.

Re:It's getting closer

[pauljlucas]

But the amount of processing power in the box would be much higher

Clearly, you've been fooled by the megahertz myth.

Re:It's getting closer

[Jezza]

That's actually part of my point SOME applications look better on the G4 than others (I'm highlighting Photoshop as an example that makes the G4 "look good" - of course if you run Photoshop a lot then this might be the metric that gives the best impression).

I'm sure someone out there will be able to find applications that really DON'T make the G4 look good. My point is that (warning: possible bad spelling) Sorenson Encoding is one thing that really needs the AltiVec to run well (another example is BLAST, but I guess fewer of us will want to run that).

As to a twin G4 1GHz being slower on benchmarks than a 2GMz P4 - yeah if you choose your benchmark carefully that'll be true (of course that "carefull choosing" might actually indicate the performance of the application you're interested in - it's the converse of the Photoshop example above).

Of course in most applications on modern machines G4s and P4s aren't the limiting factors. But that's whole different can of worms.

The only sensible advice about benchmarking is: test the application you want to use, with the operations that will take you the most time.

Re:It's getting closer

[Jezza]

Sorenson encoding is what we're talking about here (decoding is pretty easy - even quite feeble processors can do that) but encoding is hard.

G4s are encoding video at better than 1:1 speed last time I looked a high spec P4 was doing this at 1:25 (one hour of video encoded in 25 hours). Now that was a while ago, so I'd expect P4s to be doing better now (but are talking about 2GHz(ish) P4s here, they have got better since then).

This dramatic speedup for the G4 was the result of a "discovery" at Apple on how to do this. This lead directly to the SuperDrive. On the PC people use custom boards to do the encoding. Of course if you don't want to create DVDs that play in domestic players then this is a total non-event! But a lot of Macs are being sold on the back of this, from the eMac (cheap CRT all in one Mac, very much like the old iMac, but with a G4). Of course this also means that the SuperDrive won't appear in the iBook (no AltiVec).

Of course, you bring up another thing that requires AltiVec, FinalCut Pro RealTime Effects, again VERY important to Apple.

BLAST sells lots of Mac's sure you need to be a genetics researcher to need it, but in that field it's like MS-Office! Yeah, most people here won't care - but it's important to Apple - there are reasonable numbers of machines to be sold into this area, and it makes good marketing copy.

Supported Hardware

[Darth_Burrito]

Read this first:
http://www.opensource.apple.com/projects/d arwin/6. 0/install.x86.txt

Supported Hardware
------------------

IDE:
Only the PIIX4 IDE controllers have been found to work.
Attached devices must be UDMA/33 compatible or better.
Ethernet:
Intel 8255x 10/100 ethernet controllers are supported.
Video:
You must have a VESA 2.0 compliant video card. Almost all
modern graphics cards are VESA 2.0 compliant. However, emulators
such as vmware do not have VESA 2.0 compliant emulated video cards.
Successfully tested hardware:
All 440BX motherboards tested have worked with their internal
IDE controllers.
IBM ThinkPad A21m (with onboard Intel ethernet)
Known to not be supported:
All AMD and VIA based systems.

Re:Supported Hardware

[Kissing Crimson]

Odd. That's exactly the hardware in my (home) Red Hat server at the moment. Maybe I'll actually take a look at it tonight.

oh well..

[bm_luethke]

Known to not be supported: All AMD and VIA based systems.

There goes me trying it. seems to have a fairly small set of hardware it runs on.

Re:oh well..

[Jezza]

But it is open source - so these problems are in our power to fix, right? I mean Linux only ran on a couple of hardware configs to begin with, look at it now. (Runs on pretty much everything)

Serious question...

[Anonymous Coward]

Does anyone actually use Darwin on x86? I know it's good that Apple have kept open the core of their OS (although it'd be a good step to see more Aqua code), but surely if you want a PC-based UNIX you'd go for some BSD flavour or Linux in the first place.

Only reason I can see for running Darwin is for Mac hackers who want to enhance the OS -- but that poses another question: does Apple accept patches?

This is a serious question -- what are the benefits of Darwin being open?

Re:Serious question...

[DeltaSigma]

Depending on the license, the benefit might be that one could fork and make their own version of darwin, allowing mac (and x86 Darwing users) an instance of choice. ...and choice is a beautiful thing.

Re:Serious question...

[Raster Burn]

They say on their website that they maintain an x86 port just to ensure that their code is relatively portable - not really for our benefit.

GCC

[00_NOP]

I assume Apple don't build their production software with GCC, or do they?

Anybody care to comment - as best they can - on the difference between the GCC compiled code and code compiled with whatever tools Apple use?

Re:GCC

[chriswaco]

Apple uses gcc. I think they moved to gcc 3.x for the PowerPC builds. I don't know what version they are using for Intel builds.

There are probably some projects at Apple that use CodeWarrior or even MPW, but in general MacOS X is built using gcc.

Re:GCC

[Anonymous Coward]

The developers tools that ship with OS X do use GCC, and it is my understanding that Apple developers use these tools for their own software.

The dev tools that ship with OS X are excellent, and I would be very surprised if Apple put all that effort into making such kick-ass tools only to use something else internally.

Re:GCC

[mr_z_beeblebrox]

Anybody care to comment - as best they can - on the difference between the GCC compiled code and code compiled with whatever tools Apple use?

GCC is nearly ansi standard, I am sure that whatever they use for C it also is nearly ansi standard. Additionally, (correct me if I am wrong) but I do not believe it matters what you compile on. What will matter later on is what libraries are installed and that is a user/administrative choice. That is why you can get a Linux environment for your windows PC from Red Hat.

Re:GCC

[00_NOP]

I meant in terms of performance. Not the actual code base.

Re:GCC

[vranash]

Dude, that is so flamebait.

Go read the GPL, or hell, just read the GCC FAQ, I believe that covers it.

Code COMPILED with GCC is *NOT* required to be under the terms of the GPL, it is only required to abide by the terms of the libraries in question that it is linked again, libstdc++ and libc are both LGPL'd, which basically states your program may be licensed in whatever way you deem fit, however any changes you make to the LGPL'd libraries in question must be made availible for those who recieve a copy of the program/libraries. This is in no way intended to limit one's ability to use proprietary code as a frontend or backend to the library in question, it's simply meant to ensure that proprietary extensions don't get made to 'standardized libraries' and thus break compatibility with existing/future apps (of course sure breakages might occur anyway, but it prevents the microsoftian 'embrace and extend' philosophy).

Re:GCC

[Jezza]

It seems (from a developer's point of view) that they use GCC for Darwin and Cocoa and CodeWarrior for Carbon.

This makes sense - Mac developers have used CodeWarrior on "classic Mac OS" for quite some time.

NeXT always used GCC (in fact I think they helped to add Objective-C to GCC in the first place).

It's a safe bet they use GCC for the Intel builds too - that's how you get portability for Kernels (that's how Linux does it too).

It's always seemed to me that GCC was the unsung hero of open source, and Linux in particular.

Re:GCC

[Jezza]

Actually it matters a lot. Small differences they might be, but as the old saying goes: "The Devil's in the details". GCC is the best choice, because it's a damn fine compiler (I've used it for years, it's great) and because it's so widespread, especially useful if you want to create code that's portable (Linux is a great advert for portability of GCC developed code).

The other compiler that Apple advocate is CodeWarrior this has a long history with Mac developers, and is used for Carbon projects (again they're probably derived from Classic code originally developed with CodeWarrior. There is nothing wrong with that compiler either, so as the old saying goes "if it ain't broke don't fix it"). Mac developers are used to that compiler so switching doesn't make any sense. Again if you have an application like Office or Photoshop it doesn't make sense to get bitten by lots of compiler issues.

Re:GCC

[Jezza]

Your logic is flawless, and that is indeed true. You can use a GPL compiler and the binaries aren't GPL (bacause they aren't extentions to the code in the compiler, they are the output from the compiler). Linking is a grey area. Say you subclass an object that's GPL, then the subclass is also GPL even though your code is in a different file. But this is a library issue, not a compiler one, this could happen if you used a GPL library with Visual C++!

I think the issues are generally overstated, but of course you do have to check what you're doing. Of course if you're happy to use the GPL for your code then this isn't an issue!

Re:GCC

[AgtAlpha]

mwcc is the Metrowerks C Compiler. It's also the same compiler that is used for BeOS PPC (BeOS x86 uses gcc).

Re:GCC

[Jezza]

Heh, It's a suite of tools that includes a compiler - so I understand. I'm not from the Mac side of the tree so, I get stuff like this wrong. Anyway people from the Mac side use this stuff (whatever it's called) and I think Apple do too for stuff that's come from old style Mac OS.

Thanks for clearing up my confusion!

Re:GCC

[Jezza]

Yeah sorry - I was REALLY unclear.

They use GCC for all Darwin and TO IMPLEMENT (the words I forgot!) Cocoa.

They use the CodeWarrior toolset (here I am confused - if I still have the name wrong, I'm talking about the compiler that comes in the box that has CodeWarrior on it, I just assumed it was called CodeWarrior: my bad) for the old Mac derived stuff.

Well that's what they suggest developers do anyway. Darwin IS built with GCC.

Re:GCC

[sparkz]

It's always seemed to me that GCC was the unsung hero of open source, and Linux in particular.

Which is why you should be saying "GNU/Linux"

If you mean the statement, you should say GNU/Linux - if you do not agree with it, then just say "Linux"

Re:GCC

[solferino]

It's always seemed to me that GCC was the unsung hero of open source, and Linux in particular.

i agree with you (about gcc not getting the credit it deserves) however i would have worded my support somewhat differently, viz.

It's always seemed to me that GCC was the unsung hero of free software, and GNU/Linux in particular.

As i'm sure you know, GCC is a GNU project and was originated by RMS - if we want it to attain more recognition, perhaps referring to it as its authors request would be a good start

Cheers

Re:GCC

[borg]

I agree. The problem that I have had is that I'm too lazy to type four extra characters each time I mean to say GNU/Linux. I've solved this by implementing a perl script that scans STDIN for the phrase "GNU/Linux" and replaces it with "GNU/GNU/Linux." This is really handy for me, since I do with to honor the contribution of the GNU project to GNU/GNU/Linux, but I don't want to be bothered having to type GNU/GNU/Linux when just GNU/Linux conveys the same information.

I think there's still a few bugs in it, though. I'll post it to freshmeat when I've debugged it a bit more.

Re:GCC

[Uller-RM]

I don't know what they use internally... but the Developer Tools on the third Jaguar CD use the jam project manager (make alternative) to shell out to GCC for compiling. All the compiler options are available. In fact, there's even an option to choose between gcc-2.95 and the 3.1 beta.

It also wouldn't make much sense to use different IDEs for different APIs. Cocoa can be used with C++, and Carbon with Objective-C, you know. They're just different APIs, one that takes advantage of OO and dynamic typing wherever possible, the other that takes a more fragmented but backwards-compatible approach. Personally, I like both of them; I'm still more comfortable with C++ though than with Obj-C, so I use Carbon more.

(I used to hate Obj-C, mostly because it does dynamic type checking out the ying-yang and I'm an old Z80/8088 optimization hack, but I'm warming up to it now. It's somewhat like mixing C and Smalltalk, easy to learn but at the cost of orthogonality.)

Re:GCC

[benedict]

NeXT actually violated the terms of the GPL --
something about restricting distribution of their
version of gcc -- and the FSF had to threaten
them with a lawsuit to get them to play fair.

Re:GCC

[Cryptnotic]

It is not necessary to include GNU software with a Linux distribution. There are plenty of small Linux installs (such as for portable devices like iPAQ's or eMpeg's) which do not include any GNU utilities and do not even need to be linked with GNU libc (glibc). Busybox provides a rather usable shell and set of shell utilities (things like ls, cat, vi, sh, init, login, stty, etc). It can be compiled with uClibc (a free, lightweight libc) and it uses no GNU code.

Linux may be GPL, but it is not necessarily associated with GNU (except that it expects to be compiled with GCC-compatible tools).

(By the way, Linus has always praised GCC, without which, he says, Linux would not have been possible. Linus does not, however, request that his kernel be called GNU/Linux.)

Re:GCC

[solferino]

It is not necessary to include GNU software with a Linux distribution.

agreed, however almost the complete majority of os distributions using linux as the kernel heavily rely on GNU software including gcc - and i was responding to the original posters comment where he stated -

It's always seemed to me that GCC was the unsung hero of open source, and Linux in particular.

obviously, he is here referring to linux-kernel based systems that include and use gcc otherwise how could gcc be the unsung hero if it was not even present?

open source relates to a particular technocratic 'best-tool-for-the-job' philosophy - to use it in relation to a flagship project of a different philosophy (Free software) which, while holding mostly harmonious views, differs in the emphasis on certain priorities is just a little off-track - i was simply pointing that out, and saying that if you do want to give praise to gcc (as the original poster did) then you should word yr support properly

finally, to answer yr final paragraph

(By the way, Linus has always praised GCC, without which, he says, Linux would not have been possible. Linus does not, however, request that his kernel be called GNU/Linux.)

good for linus, i appreciate his tremendous contribution and his acknowledgement of the work of the shoulders he stood upon, and continues to stand upon - however i think yr point is irrelevant as no-one is asking that the kernel should be called GNU/Linux as you state - simply the request is that the overall os when it benefits greatly from GNU contributions should acknowledge that support

you seem to be a little confused on this point - i would suggest you visit the fsf site to check out their faq on this point

finally, i didn't want to pick a fight here - mostly we are all working together to build something of great philanthropical importance and ppl who use and support free software i count as my friends - my original post was simply to say that if you want to praise gcc, then best to refer to it as its creative team ask it to be - as free software and not open source - fairly straightforward point i think

cheers

Re:GCC

[Jezza]

If one is talking about Linux the kernel (as I am here) then I think GNU/Linux is wrong. If we're talking about a typical distribution then GNU/Linux is a better description (as in most there is a vast amount of GNU software). Of course there is always a danger that people who are less equainted with the subject will think that GNU/Linux is something different to Linux.

Re:GCC

[Jezza]

The point I'm making is that new applications are usually developed in Cocoa and then they use Apple's tool set (and GCC). Older applications are being updated to Carbon (so called Carbonized) and there is no merit in making that job harder be switching toolsets (you a start you need to learn the new ide) and so they stick with CodeWarrior.

Of course new Carbon apps are a different thing, but Apple are "encouraging" that new applications are written to Cocoa. For a new application then Apple's tools are a perfectly good choice. (And have the benefit of being free)

Maybe a major problem?

[jimson]

From the release notes:
* IDE drives may not work on x86. Try it, if it doesn't work, it's a known problem.
That seems like a pretty major problem to me.....

Re:Maybe a major problem?

[Jezza]

Open Source - hello!!! This is WHY it's open source so you (or someone else) can fix it. Do you think Linux just emerged supporting almost every configuration?

As this is a "known problem" then I guess someone is busy fixing it as I type this.

Re:Maybe a major problem?

[Jezza]

As far as I understand this is a problem that's limited to CD-Rom drives. I don't run Darwin on x86 so I'm not sure, but that's what I understand.

I'm confused...

[Dirtside]

Okay, every time there's an OS X story on Slashdot, someone asks when they'll be able to use it on x86 hardware, and someone else responds and says, "Never!"

So here we have "the x86 version 6.0.2 of Darwin, the FreeBSD-based core of Mac OS X". Huh? Is it just the user interface part of OS X that there's no x86 version of? And exactly how much stuff does this "FreeBSD-based core" contain? Is it just a kernel, filesystem, and some basic utilities, or what?

Re:I'm confused...

[IamTheRealMike]

So here we have "the x86 version 6.0.2 of Darwin, the FreeBSD-based core of Mac OS X". Huh? Is it just the user interface part of OS X that there's no x86 version of? And exactly how much stuff does this "FreeBSD-based core" contain? Is it just a kernel, filesystem, and some basic utilities, or what?

Darwin itself is just the kernel, some drivers, some tools (mostly bsd and gnu) and a few bits and bobs like NetInfo. Large parts aren't actually BSD however, like the microkernelness, and they have their own IO APIs (IOKit).

MacOS X is then everything else - not just Aqua as some suggest, but Quartz, Aqua, all the utilities/programs (finder, mac ui, control center), the iApps, Cocoa, Carbon etc. In fact, virtually everything that you need to have a useful OS.

From what I've seen so far, it seems that Darwin/x86 is at about the same level that Linux was in terms of features/hardware support in 1993.

Re:I'm confused...

[tswinzig]

Okay, every time there's an OS X story on Slashdot, someone asks when they'll be able to use it on x86 hardware, and someone else responds and says, "Never!"

And every time there's an article on slashdot telling of the latest Darwin release, someone posts a question very similar to yours:

So here we have "the x86 version 6.0.2 of Darwin, the FreeBSD-based core of Mac OS X". Huh? Is it just the user interface part of OS X that there's no x86 version of? And exactly how much stuff does this "FreeBSD-based core" contain? Is it just a kernel, filesystem, and some basic utilities, or what?

It's basically the kernel. Nothing on the level of the user-friendly OS X.

Re:I'm confused...

[lemkebeth]

Not entirely true.

Darwin includes the *entire* BSD layer.

This means pretty much all of your comman dline stuff.

An install of XFree86 is also provided (not a window manager though)

On PPC Darwin would make a pretty good server for serving stuff up.

On i386 there isn't enough hardware support for that. If someone really was interested in that they could fix some of that hardware support for i386.

The truth is that most people who decide to work on Darwin are people who work on PowerPC.

Re:I'm confused...

[John Siracusa]

MacOS X is then everything else - not just Aqua as some suggest, but Quartz, Aqua, all the utilities/programs (finder, mac ui, control center), the iApps, Cocoa, Carbon etc. In fact, virtually everything that you need to have a useful OS.

Linux lacks Quartz, Aqua the Finder, iApps, Cocoa, Carbon, etc. Both Darwin and Linux can run all the expected command-line Unix apps (emacs, apache, Perl, GNU utils, etc.) as well as X11 and the assorted window managers and GUI toolkits. So what is this quote saying? Is it saying that Linux lacks "virtually everything that you need to have a useful OS"?

Re:I'm confused...

[g4dget]

What he means is that Darwin lacks a GUI and GUI apps. Obviously, Linux has plenty of those.

Re:I'm confused...

[ciryon]

Darwin is not really a full featured distribution. It does not contain desktop apps like KDE or Gnome. It's a scratch UNIX distribution.

But I do wonder if drivers for say printers and cd-burners and digital cameras or included with Darwin or on a higher level in OS X?

Ciryon

Re:I'm confused...

[schwanerhill]

I'd say it means that Darwin lacks most of the stuff that sets Mac OS X apart from all the other Unixes.

Unhelpful answer.

[Anonymous Coward]

Run it if you like BSD and microkernels.

Some people like the technical approach; some people like it because it's "fun" to play with/develop what will be the basis for a true consumer product; some people like it for the same reasons others prefer penguins over platypi.

Myself, I'd rather run *BSD (in Free/Net/Open forms) over RH8 for a number of reasons- a few technical, a few based on rational-self-interest (the BSD tools and system layouts seem more 'intuitive' - always a dirty word - to interact with vs. some of the GNU-scene counterparts, but that's just personal preference), and some political.

I couldn't say how good it actually is, because I've been prejudiced against it by hearsay (QNX6 gets love from me, if I want to enjoy a microkernel OS), but that's why someone'd want it.

Anyone actually use Darwin?

[ryochiji]

Just out of curiosity, is anyone actually using Darwin? If you do use Darwin, why do you use it?

Re:Anyone actually use Darwin?

[Anonymous Cowrad]

I'm using it. The reasons I'm using it will get me modded down, though.

Quite simply, I think Linux sucks, and I choose not to run it on my x86 boxes. Almost all of my x86s are FreeBSD machines, but FreeBSD gets pretty boring after a while. Set it up, it runs, you're done.

So I play with Darwin now and again, just for the change of pace.

I don't know if 'change of pace' is the kind of answer you're looking for, but that's mine.

Re:Anyone actually use Darwin?

[Anonymous Coward]

I use it because my operating system, Mac OS X, runs on it.

Re:Anyone actually use Darwin?

[jdh28]

Care to say why you think Linux sucks?

(Not a troll, just interested.)

john

Re:Anyone actually use Darwin?

[IamTheRealMike]

Why do you dislike Linux so much? I haven't tried FreeBSD so I don't know why its users are so loyal. Considering it's lack of hardware support/ease of use compared to Linux, what makes it so much better?

Re:Anyone actually use Darwin?

[moderators_are_w*nke]

Its all down to personal choice really. I prefer FreeBSD to Linux because I like the ports system, I like the cleaness and I've had some bad experiences with Linux in the past.

It all comes down to what you like in the end really, I duel boot Windows 2000 and FreeBSD 4.7, both of which I prefer to Linux, but thats me. If you're interested, I would suggest you give FreeBSD a try - you might like it.

Re:Anyone actually use Darwin?

[SirSlud]

We have a box at work serving anywhere from 20 to 40 requests per second, 24/7 .. its been up, no reboot required, for over 2 years. Its been through software upgrades too, so its not like we havn't touched it in this time. (Warning: Thats just one example. I'm not looking to get into an 'uptime' pissing contest. I'm sure Linux can do this too .. or at least I think it can. :)

There are lots of differences between FreeBSD and Linux that are more esotaric in nature, but to me, FreeBSD represents the best balance between a stable, time tested code base and active development.

I have to imagine there is a reason why FreeBSD is used in so many ISPs and server farms (like Yahoo.) For us, the choice has been well worth it, save for a few troubles, like the lack of native threads.

To me, FreeBSD is the OS of choice for sysadmins who're well past the 'gee wizz aint this cool' phase of their computing life and just want something stable and tidy; even if its not bleeding edge with respect to hardware support. To me, FreeBSD is an experienced performer that does very little complaining, even if it can't do *all* the tricks Linux users might wish it did.

Re:Anyone actually use Darwin?

[Arctic Fox]

I agree. Going offtopic though. It's stunning how FreeBSD works. It's all there. No futzing with anything. After battling the Debian installer, installing Mandrake (8 maybe? been a while) and Redhat 7.x and fighting with it to no end to get the volume buttons on my no-name laptop working, I gave up.
FreeBSD 4.6 just worked. Take that back. I had to uncomment # device pcm in the kernel, and recompile and buttons worked!! I think 4.7 comes with that already compiled. DVD support through MPlayer kicks tail. Only reason to dual boot to XP is for games. Except the buttons dont work in XP! >:(

Re:Anyone actually use Darwin?

[Huge Pi Removal]

OK, I'm not the original poster, but...

I'm a FreeBSD person, I use it on my various servers (from web to mail to router to NAS). (Oddly, the thing that first got me into *BSD was that MkLinux wouldn't support the ethernet card on an old 68K Mac, so I put NetBSD on it instead.)

Now not that I've used Linux much, but I once heard the phrase "FreeBSD is an operating system, Linux is a kernel with some stuff attached". Certainly the way FreeBSD *feels* coherent, and is very natural to work with, makes that statement possibly true for me.

Linux doesn't suck. As such. It can just seem like a goddamn mess at times. In the same way that skins designed by people used to Windows never ever look as good as those designed by people used to Apples (just look at www.kaleidoscope.net), apt-get or RPMs just aren't as nice as ports (www.freshports.org). Where things get installed, where the logs are (fair play, RedHat's pretty good at logs as well), all sorts of stuff... whenever I've nearly put Linux on a server (because I want some bit of hardware that only has Linux drivers):

1) I look at a Linux distro, with its docs, and think "Oh my God, wtf...", and

2) Suddenly someone writes a driver for FreeBSD, and I am saved :)

There endeth the entirely xenophobic ramblings of Huge Pi Removal :) Honestly, I wouldn't kill myslef rather than use Linux, but I sure do like FreeBSD (come on guys, ipfw2... see the light!).

Re:Anyone actually use Darwin?

[Anonymous Cowrad]

I guess I'd better answer that, since my original post is being regarded as flamebait. I don't really want to get into why I don't like Linux for the same reasons I don't go to Mass and talk about what and asshole Jesus was.

One reason I've been largely uninterested in using Linux is that I see no compelling reasons to leave FreeBSD for it. All my FreeBSD boxes are headless/gpu-less, so I don't care about KDE, Gnome, etc. Linux may have the advantage there, I don't know. I don't care. If I want to work on a desktop, I have OS X.

I've tried many distros, and I've never found one to be as straightforward and simple to manage as FreeBSD. I'm looking into building a Gentoo box soon, though. I hear their ports system isn't half bad.

So, Linux sucks for me. As a server, it's far too complex for my needs. As a desktop, it's far too not OS X.

By the way, moderator types,it really sucks bad when one can't express a contrary opinion without getting marked as 'flamebait'. It's not a crime to think Linux sucks.

Re:Anyone actually use Darwin?

[Jeremiah Cornelius]

Man. My own little heresy time.

I will start another little blowtorch battle here, but... Here goes.

I grew up with Berkeley /etc/rc. I used to look at SysV init and think "God, how complicated! Why'd you do all that?"

Well, several years pass, and I start running some SGI's. I start building Debian and RedHat installs... I am slowly converted.

Everytime I look at OpenBSD or the like, I keep thinking: "Man! I dig this... But what an uncontrollably unmanageable init!"

Every freaking admin has their own way to hack another startup into the main .rc, and every one of 'em names their own .rc's what they like, and stick 'em in /sbin, or /usr/local, or god knows where! It's all legal.

I don't want to fault the OS for lack of policy adherence on the part of its operators, but heck - I just don't run into this in SysV land.

I always fantasize about hacking a Debian SysV init into OpenBSD, then reality re-schedules my free time.

Someone with the time to re-package the ports not to break should fork a system and do this. There was a "Debianized" FreeBSD port, but I think it's dead...

Re:Anyone actually use Darwin?

[dvdeug]

I had to uncomment # device pcm in the kernel, and recompile and buttons worked!!

So you said the magic words and it worked. Is there are reason to believe that your troubles in Linux were any more than not knowing the right magic words?

Re:Anyone actually use Darwin?

[Fweeky]

I always fantasize about hacking a Debian SysV init into OpenBSD, then reality re-schedules my free time.

You're obviously very ill. I prescribe large quantities of NetBSD's (and now FreeBSD-CURRENT's) rc_NG, maybe with a dose of slaming your head against a brick wall until the desire to use SysV init goes away :)

Re:Anyone actually use Darwin?

[Z4rd0Z]

Every freaking admin has their own way to hack another startup into the main .rc, and every one of 'em names their own .rc's what they like, and stick 'em in /sbin, or /usr/local, or god knows where! It's all legal.

I can't speak for every freaking admin out there, but by default, no scripts go into /sbin, ever! Always /etc. The only ones that go into /usr/local/etc are for third party packages, i.e., they aren't there by default. Also, you might check out NetBSD's beautiful rc system which runs an individual script in /etc/rc.d for each service. You still configure your services in /etc/rc.conf like before. There are no runlevels. FreeBSD 5.0 will be using this same system.

Re:Anyone actually use Darwin?

[benedict]

It's impolitic and rude to say "Linux sucks",
especially when what you mean is "I see no reason
to switch to Linux from FreeBSD."

I've been a FreeBSD weenie since 1996 and I get
totally frustrated when I try to use Linux. But
I'm not going to disrespect the Linux community
by tossing schoolyard insults at Linux. It's just
not necessary.

Re:Anyone actually use Darwin?

[benedict]

FreeBSD doesn't lack "ease of use".

The installer could be slicker, but once you've
used it a couple of times, it is pretty straight-
forward, and it's actually pretty cool in that it
supports installs from CD, DVD, FTP, NFS, and I
think local filesystem, though I've never tried
that one.

Everybody raves about the Ports system, so I won't
go on about it, but it is really quite good. And
the combination of cvsup and make as system
updating tools is quite good too. I updated today
to the latest stable code using just a couple of
short, simple shell scripts that I put together.

People talk a lot about FreeBSD's lack of hardware
support, but in practice, it's not something I
notice. I've only once found myself with a
computer I couldn't install FreeBSD on, and it was
weird, with the CD-ROM and modem both attached to
the sound card. And if I'm actually buying
hardware, I just check the hardware compatibility
list before I buy. It's not a big deal at all.

Re:Anyone actually use Darwin?

[benedict]

You should check out the rc system that comes
with NetBSD 1.5 (and will come with FreeBSD 5).
It's similar to SysV rc, but instead of making
the dependencies implicit in the numbers in the
script names, they're encoded explicitly at the
beginning of each script.

% head fsck
#!/bin/sh
#
# $NetBSD: fsck,v 1.2 2001/06/18 06:42:35 lukem Exp $
#
# PROVIDE: fsck
# REQUIRE: localswap

These scripts also pull in a common set of sub-
routines, so instead of having everybody define
their own start and stop routines, you just set
a couple of variables and let the system do the
rest. The routines can be overridden with one's
own when more control is needed.

Mac OS X uses a similar scheme. Within a few
years, monolithic rc is going to be nothing but
a bad memory in BSD-land.

Re:Anyone actually use Darwin?

[Jeremiah Cornelius]

You are right!

I just checked out the link provided by Fweeky - I am very impressed. I really like the documentation of actual approaches used by the FreeBSD team to arrive at this - and appreciate the use of library shell functions to standardize the format of inits. RedHat has a set of standardized script called as regular executables from /sbin in their scripts, but I agree that this needlesly clutters /sbin. It also reduces portability of init scripts to non-RedHat derived boxes, and does nothing to enforce scripting style by the init system itself.

I will investigate further - but Kudos to the NetBSD folks! You have provided the functionality I enjoy with SysV and bettered it.

Re:Anyone actually use Darwin?

[Twirlip of the Mists]

A desktop OS is supposed to be a general-purpose environment for running user-interactive software. Linux, even with eye-candy from such outfits as Ximian and Red Hat, is not an acceptable desktop OS. FreeBSD is not an acceptable desktop OS. Solaris is not an acceptable desktop OS. HP-UX is not an acceptable desktop OS. IRIX is not an acceptable desktop OS.

The only acceptable desktop OS's currently available are Windows XP-- or, equally well, Windows 2000 if you can find somebody to sell it to you-- and Mac OS X. Earlier versions of Windows and Mac OS are acceptable to varying degrees.

Anybody who thinks FreeBSD (or Linux, or Solaris, or et cetera) is an acceptable desktop OS is either not doing very many things with his computer, or is happy using substandard tools and employing workarounds.

Re:Anyone actually use Darwin?

[g4dget]

By the way, moderator types,it really sucks bad when one can't express a contrary opinion without getting marked as 'flamebait'. It's not a crime to think Linux sucks.

Yeah, as opposed to the "even handed moderation" from Macintosh zealots? Anything remotely critical or even questioning of Apple is immediately modded down by the Macintosh crowd.

Ultimately, users will decide, no matter what power trips or delusions certain user populations have.

Re:Anyone actually use Darwin?

[Twirlip of the Mists]

It's not about speed, friend. It's about what you can do with it. I'd like to see you play a game with FreeBSD, or run a decent word processor (Open Office, AbiWord, and Lyx need not apply), or run an accounting package, or run Photoshop. As I said before, FreeBSD is an exceptional server operating system. It is an unacceptable desktop operating system, unless your needs as a desktop user are so trivial as to be virtually nonexistent.

Re:Anyone actually use Darwin?

[Eil]

If you ever think of giving Linux another shot, you might give Slackware a try. It's a very no-frills distro, which is (IMHO) a good thing. You dive in and make things the way you want them instead of wading through layers of GUI and buggy config tools. I've been hearing good things about Gentoo also.

One thing that would make me switch to FreeBSD instantly is a framebuffer console. On my desktop, being able to adjust the console resolution is nice, but on my laptop it's essential. 640x480 expanded to 1024x768 w/o anti-aliasing is extremely painful to look at. Painful enough that I won't use an OS that can't display it's console at the native LCD panel resolution.

Re:Anyone actually use Darwin?

[Twirlip of the Mists]

I respect this opinion, but I find it atypical. I know a reasonable cross-section of computer users-- business types, creative types, programmer types, non-computer-user types who just need to surf and do email-- and none of them would find any of the currently available Linux or BSD flavors acceptable desktop operating systems. They each have their own reasons, of course, but the conclusion is the same.

Darwin is BSD

[yerricde]

but surely if you want a PC-based UNIX you'd go for some BSD flavour

Darwin is a BSD flavour. A long time ago, the BSD source forked; on one side, we have FreeBSD and Darwin, and on the other, NetBSD and OpenBSD.

(I could be wrong; corrections are appreciated.)

If I can get this to work...

[rickthewizkid]

... do I get a "Darwin Award?"

Heaven knows I'll be spending enough time hacking on this that I won't have *time* for kids... :)

Seriously, kudos to Apple for releasing this... it was a good treat for the day my DSL came back to life!

RickTheWizKid

*not* FreeBSD based dammit

[Leimy]

Its XNU... mach+a bsd personality in the same address space with some FreeBSD userland tools.

Re:*not* FreeBSD based dammit

[AIXadmin]

Since Mach can have multiple personalities, could you create and run a Linux personality and a BSD personality at the same time?

Re:*not* FreeBSD based dammit

[Leimy]

Normally you would be correct but in this case the personality is not a "mach server" but in the same address space as mach. This just means they both access the same memory regions and the BSD layer isn't as "protected" as it would be if it were a userspace implemented personality.

The reason for putting XNU and BSD together [I bet] is that it reduces the amount of IPC [mach messages and ports] to do Unix things. This should improve the latency of system calls and other things.

Unfortunately the improvement isn't that much from what I can tell. Latency is still pretty high on XNU compared to NetBSD or Linux on the same hardware [I don't know if my friend published his results of some tests he has done so I won't point there]. Anyway I don't know if the version of Mach used in XNU has all the various improvements I have only read about like Mach Continuations [Unix Internals: The New Frontiers Uresh Vahalla] but such things would further improve OS X in general.

The nice thing is if we want to play with this stuff we can. Its !Linux and !FreeBSD so it is good exposure to something else with a slightly different perspective and design. Also if you go to OpenDarwin you can get the source pretty easily and tweak it yourself.... I did a little and ran my own custom kernel on my TiBook for months before I got Jaguar. [UFS implementation logging and exploration... nothing terribly fascinating].

Re:*not* FreeBSD based dammit

[Fafhrd]

Since Mach can have multiple personalities, could you create and run a Linux personality and a BSD personality at the same time?

Yes, it would be possible. In fact, Modern Operating Systems (great reference book by Tanenbaum) mentions the possibility of running a MS-DOS personality together with the BSD personality on top of Mach.

Indeed, the first Linux version to run on Macs, MkLinux, was actually a version of Linux adapted to be hosted on top of Mach... a "Linux personality", so to speak. The now usual monolithic PowerPC Linux came later.

Of course, running these two different personalities side-by-side would entail very careful resource management, and would probably not be very practical.

Re:*not* FreeBSD based dammit

[benedict]

You're not quite right there. The BSD personality
contains a lot of code from FreeBSD. So there's
FreeBSD in the Darwin userland and kernel.

I'd call it NeXTStep-based before I'd call it
FreeBSD-based, but both are accurate to some
extent.

Re:*not* FreeBSD based dammit

[kcbrown]

XNU/Darwin??

I submitted the above anonymously...

[Anonymous Coward]

...because it's just not all that funny. Why the hell are you people modding it up?! Get a grip!

Re:I submitted the above anonymously...

[Anonymous Coward]

because my self-doubt is a hellish see-saw of conflicting emotions and self loathing. Stop modding me up - I can't deal with the pressure!

15 mins later

[MrFreezeBU]

15 minutes and the reg server is spouting 500 server errors...Are they running X-servs?? j/k

Open Source for Apple

[swagr]

Basically this just means that developers who write console applications, servers, etc can now port and test with ease.

This is good for Apple because they get more for free.

What exactly do you get?

[mcraw]

If I were to install this on my machine, what exactly would I get?

Would it be just a basic BSD-like OS? Would any OS X features be there?

In short, how much of a running Mac OS X install is OS X and how much is Darwin?

Not quite...

[ryochiji]

>it boils down to the fact that Simpler means Stupider

Not when you also have access to a CLI (through Terminal.app), on top of the "elegant and beautiful" GUI. Or perhaps you're actually saying that the free development tools (IDE, GCC, GDB) bundled with MacOS X are somehow "Stupider" too, and that anyone who happens to use this highly-usable version of UNIX is somehow "Stupider" than you?

Re:Not quite...

[pythas]

Here's what I do:

1. Insert CD
2. Install developer tools

Hope that answered your question.

Re:Not quite...

[Twirlip of the Mists]

No. They're included with new computers that come with Mac OS X, and they're included in the retail Mac OS X box. The only time you have to download the tools is when new releases of them become available.

And signing up for the Apple developer program is about as tricky as signing up for Slashdot. If you want to do it anonymously, get yourself a free email account and go for it.

CD-ROMs?

[bongholio]

Also from the release notes it looks like you're lucky if you can get a CD-ROM to work at all...

* SCSI CD-ROM drives are not supported

* IDE drives may not work on x86...

Myth: Viral nature of the GPL

[Charles Dodgeson]

Due to the viral nature of the GPL, any software compiled using GCC could be considered a "derivative work" of the GCC, thus forcing that software to be open-source under the terms of the GPL

I find it remarkable that anyone could actually read the GPL and believe this myth. See the GPL FAQ which specifically addresses this question:

Can I use GPL-covered editors such as GNU Emacs to develop non-free programs? Can I use GPL-covered tools such as GCC to compile them?

Yes, because the copyright on the editors and tools does not cover the code you write. Using them does not place any restrictions, legally, on the license you use for your code.

Re:Myth: Viral nature of the GPL

[mindstrm]

Which of the standard C or C++ libraries are those?

And your original statement "Due to the viral nature of the GPL, any software compiled using GCC could be considered a "derivative work" of the GCC, thus forcing that software to be open- source under the terms of the GPL"

Is still totally wrong. Any software including GPL libraries that come with GCC, yes, they would be restricted by the GPL... but you quite clearly stated that all software built with gcc is automatically GPL, which is BS.

Re:Myth: Viral nature of the GPL

[Jason Earl]

Yes, RMS wants to phase out the LGPL, especially in those cases where Free Software has created a library that doesn't have a commercial software equivalent (libreadline). So far he hasn't been particularly persuasive. However, it simply isn't true that the GNU C++ libraries are GPLed, at least they aren't on my machine.

In fact, libreadline is the only example I can think of off hand of a GNU library that is GPLed. Even the Gnome and GTK libraries are LGPLed (as opposed to QT which is GPLed).

Re:Myth: Viral nature of the GPL

[Arandir]

as opposed to QT which is GPLed

Qt is dual licensed under the QPL and GPL. The dual licensing means that it is NOT viral.

Re:Myth: Viral nature of the GPL

[Jason Earl]

What planet are you from? Have you actually read the QPL. Not only does it require that you distribute source code with your derivative works to the people you distribute binaries to, but it also requires that you make this software available to TrollTech should they ask for it. In other words the QPL is even worse than the GPL in that reguard. I can write a GPL program and sell it to a client and only myself and my client will have access to the source code. However, if I use QT with the QPL then TrollTech can demand a copy of the software as well, and they could theoretically resell this software to my client's competitors.

Here's the relevant clause:

If the items are not available to the general public, and the initial developer of the Software [TrollTech] requests a copy of the items, then you must supply one.

The QPL also requires everything that the GPL requires. If you choose to use the QPLed version of QT then you have to distribute your work with source, and you have to license your software in such a way so that the people receiving your software are also allowed to distribute the software (as long as they distribute the source code as well).

In short once you distribute software linked against the QPLed version of QT then it can never be closed again (much like the GPL).

Now, you can buy a commercial license for QT and then distribute commercial software based on QT. However, the QPL is every bit as viral as the GPL. In fact, it's worse.

Free as in...

[intermodal]

*cues up canned RMS speech* The views expressed in said canned speech are not my views. More power to them. Now if only they'd port the whole OS for more hardware, Microsoft might start losing sleep...

Re:Free as in...

[MalleusEBHC]

Now if only they'd port the whole OS for more hardware, Microsoft might start losing sleep...

For the billionth time, they would lose more sleep than M$ would. Yes, M$ makes a good chunk of money off OS sales, but they also make an assload off other sales (especially Office). Apple makes 90%+ of their money from hardware sales last time I checked. For Apple to release an OS X build for generic x86 hardware would be economic suicide. Steve knows where his bread is buttered.

older version

[nuttyprofessor]

Must be an older version than the binary
that ships w/ 10.2.

% uname -a
Darwin *.*.*.* 6.1 Darwin Kernel Version 6.1: Fri Sep 6 23:24:34 PDT 2002; root:xnu/xnu-344.2.obj~2/RELEASE_PPC Power Macintosh powerpc

They do it to maintain the balance of power

[Ilan Volow]

The name of the game is deterrance. It's like cold war.

• Apple works on a partial x86 port of OS X (Darwin x86). Not that they are ever going to deploy a full x86 OS X, but they want to let microsoft know they can do so at any moment.
  
• Microsoft keeps "re-evaluating" whether they should be releasing Mac versions of Office. Not that they are ever going to cancel Office for Macintosh, but they want to let Apple know they can do so at any moment.
  

Re:They do it to maintain the balance of power

[mindstrm]

Apple works on a partial x86 port of OS X (Darwin x86). Not that they are ever going to deploy a full x86 OS X, but they want to let microsoft know they can do so at any moment.

They very well may release it; if intel processors get far enough ahead, apple will most likely make a new mac based on an intel processor. It won't be a version for the common PC, but it will be a box with an intel chip in it.

Re:They do it to maintain the balance of power

[sparkz]

Same as Solaris_x86 was developed way before it was ever released, and continued to be developed after it was announced that it would no longer be available ... keep your options open, make stuff, and if it suits the business, then make it available, but whether you currently want to make it available or not, you need to keep working on it.

Re:They do it to maintain the balance of power

[SmittyTheBold]

They very well may release it; if intel processors get far enough ahead, apple will most likely make a new mac based on an intel processor.

I suppose they could release x86 Macs, but they will NOT replace the PPC line. To do so would be suicide.

Think of it this way: Developers tagged along with Apple on the PPC transition somewhere around seven years ago. The same developers followed the OS X transition, but they took more time with that move - some bog houses don't have 'real' OS X ports out yet.

Now, both of these were rather simple transitions for developers to make. Somewhere on the order of a 5-15% code change for most apps, last I knew.

A move to x86 is a BIG move. A lot of software will have to be rewritten. AltiVec instructions have to be replaced with SSE (, MMX, SSE2, etc) blocks.

Even worse, there is no clean-cut way to interoperate with older PPC code. The PPC is relatively good for emulationg other architectures, but the x86 is rather poor at emulating anything modern - if for no ther reason than an astounding lack of general-purpose registers. Macs have a way of runnign even your old software in a satisfactory manner. I can run games from 1993 on an OS X PPC machine, and it still runs pretty well.

If Apple released the OS X x86 machine, it would run NONE of the old Mac software, nor would it run new OS X software that hadn't been expressly recompiled yet.

Apple got developers to transition from 68k to PPC, then they got developers to transition from OS 9 to OS X. There is no way the same developers would transition again a mere two years later.

Re:They do it to maintain the balance of power

[sql*kitten]

Apple works on a partial x86 port of OS X (Darwin x86). Not that they are ever going to deploy a full x86 OS X, but they want to let microsoft know they can do so at any moment.

More likely it is to keep Motorola/IBM on their toes. They could have Apple over a barrel because they control the PowerPC supplies. Apple screwed Motorola royally over Mac clones, and Moto would love to get its revenge.

Re:Supported Hardware - USE THE FORK!

[Jezza]

Wasn't there some legal reason for this? (I'm pretty sure there was) So while I agree with you, I don't think Apple are to blame.

As I remember it was something about someone under that age could be held to the licence or something?

Someone help me out here.

Byte compares linux and ox s performance

[mojorisin67_71]

Byte released a comparision of linux and OS X at here

Re:Byte compares linux and ox s performance

[soellman]

great article. I was surprised that osx even approached the performance of linux, and it seemed to be consistently about 75-90% as fast. Now of course, that comparison only tested osx 10.1.5, and there has been significant work on darwin since - I'd love to see an updated shootout.

DSS supports open formats

[benwaggoner]

Darwin Streaming Server can stream MPEG-1, MPEG-2, and MPEG-4. It can also stream with codecs like H.263 that are compatible with open source solutions. And QuickTime streaming itself is based on RTP/RTSP/SDP, etcetera. The only thing propritary is QuickTime specific codecs, and DSS doesn't know from codecs and clients.

As Charles Wiltgen says, "Darwin Streaming Server is the Apache of Streaming."

For the freaking 10,000th time...

[maggard]

Darwin is not MacOS X.

It is much of the core of MaxcOS X but it lacks many of MacOS X's layers and services like Carbon, Cocoa, Quartz, and of course Aqua.

If you don't know what those mean don't post the same damn lame question to /. asking for these to ONCE AGAIN be explained. There's a search function: Goddamn use it as the topic has been discussed NUMEROUS times on /.

Or show some minimal level of initiative and look it up for yourself (hint: World Wide Web) where it's all well detailed in ways even the lamest "explain-this-to-me" poster would get in a dozen places online.

Next, yes, there is an x86 port of MacOS X inside Apple. Will it ever see the light of day? Not likely (sorry PC fanboys).

Does this mean Apple plans to use MacOS X'86?

Yes, they use it every day to make sure that MacOS X remains true to the portability of it's predecessor Openstep (which was on 5 platforms.) Undoubtedly Apple figures if MacOS X can be kept running on the very different PPC & x86 platforms then they're good for about anything.

Why x86 over some other processor? First off Openstep (or whatever you want to call/capitalize it) was updated by Apple as part of their aborted "Rhapsody" strategy to both PPC & x86 so it was little effort to keep it going to MacOS X. Furthermore this helps keep Apple from getting caught in any PPC-isms in the future. Cross-porting helps show up any problems early, keeps everyone honest, provides valuable insight into many problems.

What good is Darwin? Well, it does run a lot of code, including things like Apple's free streaming media server. It gives MacOS X developers a look into the heart of MacOS X. That it comes out for x86 just lets that many more folks play with it.

Finally, to respond to the next half-dozen whinges that come up every time:

• Yes Darwin is Open Source.
• No Apple isn't going to give away the rest of MacOS X. As much as many folks go gimme-gimme-gimme-for-free Apple's management has fiduciary responsibility to keep the company profitable; giving away MacOS X in its entirety will not further that goal.
• No QuickTime does not lockout Linux or any other users. QuickTime is a file-format and some libraries, not a codec (clue phone ringing!) Yes Apple licensed a codec, you want access to it find someone willing to pony up the cash like Apple did. That other folks use that codec is lovely but there's no gun to their head preventing them from using any of the other codecs.
• No Apple would not do well selling or giving away MacOS X'86. If you think you've got some novel bit of reasoning that makes this a good strategy for Apple go pitch it to their board, don't bleat about it on /. again where nobody well-informed is buying it.
• Is Darwin "better" then XYZ? Who knows, depends entirely on for what to whom with which criteria.
• Do PPC/Mach/micro-kernels/aesthetics/etc matter? Well as Apple seems to be one of the few PC vendors doing well and MacOS X is now the best-selling Unix then yeah, apparently Apple is doing something right.
• Last, but not least, love 'em or hate 'em Apple makes waves and does interesting things. None of the other vendors constantly do as interesting things or generate so much controversy, gotta love 'em for that.

Re:For the freaking 10,000th time...

[g4dget]

No Apple isn't going to give away the rest of MacOS X. As much as many folks go gimme-gimme-gimme-for-free Apple's management has fiduciary responsibility to keep the company profitable; giving away MacOS X in its entirety will not further that goal.

Well, what will further that goal, then? Do you really believe that software developers are going to switch in droves to a proprietary, single platform set of APIs that requires the use of a 20 year old unsafe programming language? APIs that are only implemented on the hardware from a single company, which ships a very limited range of machines?

I certainly don't: I think Cocoa and Quartz are a dead end, with no prospect of widespread adoption by software developers, outside a die-hard community of Mac developers. Calls for Apple to open source the GUI have nothing to do with "gimme-gimme-gimme", they are simply a reflection that most developers and companies don't want to commit a lot of time and effort to a set of APIs that stand and fall with the fortunes of a single vendor. Open sourcing Cocoa and Quartz wouldn't make the APIs technically more attractive, but at least they would ensure their continued existence.

In any case, I don't actually want Apple to open source Cocoa and Quartz--I think it would just prolong the agony. I think Cocoa and Quartz will have to be replaced within a few years with something very different--unless Apple goes out of business first.

Re:For the freaking 10,000th time...

[alannon]

Several replies to this:
Do you really believe that software developers are going to switch in droves to a proprietary, single platform set of APIs that requires the use of a 20 year old unsafe programming language?
1) I can only assume by '20 year old unsafe programming language' you refer to C. Since you talk about only Cocoa for the rest of what you say (not Carbon) I can only assume you are referring to it when you say 'single platform set of APIs'.
Since the 'standard' for Windows programming is in C++ (arguably LESS safe than C, since you can over-ride operators), and the 'standard' API is MFC, I fail to see your point, when compared to Windows.
Since the 'standard' programming language for Linux is C, and OSX and Linux share the vast majority of the basic API's and the rest (X11) are available as a free download, I fail to see your point in comparison to Linux/BSD.

I think Cocoa and Quartz are a dead end, with no prospect of widespread adoption by software developers, outside a die-hard community of Mac developers... Open sourcing Cocoa and Quartz wouldn't make the APIs technically more attractive, but at least they would ensure their continued existence.
2) You state that Cocoa and Quartz are a dead end, with no prospect of adoption software developers outside the 'die-hard' community of Mac developers. First, Quartz is rarely accessed as an API on its own, unless you are doing eye-candy. Usually it is called by the application-level API you are writing in (Cocoa or Carbon). You are really only showing your lack of familiarity with these APIs by mentioning it in the same context.
Second, the Cocoa API is more or less source-code compatable with GNUStep. What is GNUStep? It's an open-source implementation of the Objective-C OpenStep APIs on top of X11. What's OpenStep? It's the open standard that NeXT released and implemented and eventually became Cocoa. You can write full-fledged OSX applications that cross-compile for GNUStep on Linux TODAY.
Third, as a professional developer who is experienced on (Classic) MacOS, OSX, UNIX/Linux and Windows, I will state my opinion that I find the Cocoa APIs to be the most attractive to use. If these APIs were unattractive to use, why would anyone have gone to all the trouble to do a complete re-engineering of them in the form of GNUStep?

In any case, I don't actually want Apple to open source Cocoa and Quartz--I think it would just prolong the agony... unless Apple goes out of business first.
Damn! I've been trolled.

Re:For the freaking 10,000th time...

[tialaramex]

"Yes Apple licensed a codec, you want access to it find someone willing to pony up the cash like Apple did."

My understanding (and so this is 2nd hand, but I don't see you doing any better) is that when you call Apple they say "Ask Sorenson" and when you call Sorenson they say "Ask Apple".

Yes, they can /legally/ give everyone the run around like this, but I don't think it's /ethical/ for them to do so. It may even be bad business.

Worse, experience suggests that if a better alternative is developed and made free Apple AND Sorenson will do everything in their power to discourage their users from accessing this improved technology.

The duplicability versus scale problem hits big for this kind of stuff. Everyone hiring one or two academics, making them sign NDAs and keeping them apart is not the most effective way of delivering improved compression to real users. That's why we have a dozen competing "next generation" streaming audio codecs, none of which are universally available. [It would be nice to think that MS & Apple will see sense and ship Vorbis but don't count on it]

This /is/ rocket science and if Apple were serious about delivering for media people they'd put their codec money together with other people's and have the technology developed in the open. They can differentiate elsewhere without wasting all these resources on duplicated effort.

Re:For the freaking 10,000th time...

[g4dget]

I fail to see your point, when compared to Windows. Since the 'standard' programming language for Linux is C, and OSX and Linux share the vast majority of the basic API's and the rest (X11) are available as a free download, I fail to see your point in comparison to Linux/BSD.

I didn't make that comparison between OS X and Linux, BSD, or Windows. And I agree with your analysis: all three systems, OS X, Linux, and Windows, are based on roughly the same 20 year old technology.

The difference is that many Linux and Windows developers have no illusions about that. But Apple claims that they have a system for the future and don't show any intention of wanting to move forward to something better.

My question to Apple is and remains: where are you moving forward to in terms of software architecture; Cocoa/Quartz/Objective-C is not a satisfactory answer to me.

You can write full-fledged OSX applications that cross-compile for GNUStep on Linux TODAY.

Yes, and the miniscule use that GNUStep receives today should tell you something about the desirability of the GNUStep/OpenStep/Cocoa APIs--even when people get it for free, they don't want to use it.

Re:For the freaking 10,000th time...

[benedict]

> Well, what will further that goal, then?

Exactly what they're doing. No, it's not a
strategy that's going to let them take over the
world. I doubt that there is any such strategy.

If Apple gave away all of their software, what
would they sell? The hardware? Give me a break.
It's nice hardware, but PC hardware is faster and
cheaper. Of course, it is worth something to be
free from 20 years of questionable architecture
decisions.

Re:For the freaking 10,000th time...

[g4dget]

Of course, it is worth something to be free from 20 years of questionable architecture decisions.

Macintoshes are mostly PC hardware (PCI, AGP, IDE, USB, etc.) with a PPC processor. And the Macintosh software is mostly NeXTStep with an OS9 and BSD compatibility layer, all of which go back to the early 1980's. So, what is Apple actually "free from"?

Re:For the freaking 10,000th time...

[benedict]

I like not worrying about IRQs. I know it's not
the biggest deal in the world, but I just don't
miss 'em.

I don't particularly miss ISA slots, RS232 serial,
or floppy drives, either -- I know I don't reap
any benefit from their absence, but it gives me
warm fuzzies not to look at that stuff when I'm
not at work.

Re:For the freaking 10,000th time...

[Lars T.]

Sorenson gave Macromedia basically the same codec for Flash MX. AFAIK there is no version for Linux yet. QED.

whats the diff?

[Pan_God_of_Gods]

What is it that makes darwin any different from the other FreeBSD kernals? Without Aqua in front of it darwin just isnt that cool..

Something only Apple could do?

[CathedralRulz]

I wonder if the reason that Apple is willing to make it's core OS open source is because, as long as software developed with it is running on a Mac (that's the idea), and they are the only folks selling Macs, the more the merrier.

Apple started doing this in 1999 at great expense and effort. In this time, has it paid off? I really don't know, so please enlighten me if it has or hasn't.

Re:Supported Hardware - USE THE FORK!

[kin_korn_karn]

There is no fork.

er... that's not right...

It's Obvious to me...

[maccroz]

Okay...maybe this has already been posted and I missed it, or maybe I'm way off...but here's my take on the situation.

The reason Apple is releasing Darwin for x86 is because Motorola has very little incentive to keep up in the processor wars. Granted, the PowerPC chipset is fantastic, it's just not a big player anymore as Motorola has better things to do than to cater to Apple's whims.

Just as they released an infantile Darwin for PowerPC, they are releasing it now for x86 so that people can port their drivers to Darwin. Once Apple has enough hardware support, they are one step closer to porting Aqua and all the higher layers of MacOS X towards the x86 architechture and having a way of escaping the sluggish Motorola chips.

MacOS X is a fantastic operating system and unfortunately it is far from living to its potential due to inferior and expensive chips from one provider. This is one step in the direction that people have been encouraging apple to make for the past 15 years. We all know that Apple is slow to respond, they like to take their time and make sure it is the right decision before they do anything drastic.

Imagine how much cheaper an x86 Mac would be and how much of a heavyweight they would finally be if MacOS X became an option to the other 95% of the computing population. I believe that this is just Apple keeping all their options open with a miniscule investment on their part.

I'd still like to see two buttons on the iBook, or heaven for bid they put a wheel on their stupid buttonless mouse. I love Apple, but sometimes they just have their thumbs up their asses. Hopefully this is a move in the right direction.

Re:It's Obvious to me...

[AnamanFan]

Well, Apple seems to be weaning away from Motorola and going to IBM for the IBM PowerPC 970 (A nice tech-spec here). Yet it's still a PowerPC architecture.

Something tells me there are more actions at play here than simply changing processor manufactures in this x86 release. What exactly, I don't know. But still, there's something more here than meets the eye.

Re:It's Obvious to me...

[benedict]

It seems to me that that's the only way Apple
could move to x86. It's a tired refrain at this
point, but once again, they make their money on
hardware sales, and if they went head-to-head
with Microsoft by going to a software-only model,
they'd get eaten alive.

Re:It's Obvious to me...

[Lars T.]

Motorola recently released a PDF (which it has pulled since then) with info about the new integrated processor MPC8245. This PDF also includes new processor roadmap wich shows a "7457 (0.13mu Cu SOI) 867-1833 MHz" to ship in (early) 2003, and a proposed "7457-RM" (with integrated DDR interface) 1.3 - 2.0+ GHz later.

OS X to be on x86 in 2003, according to Giga

[cweiblen]

Giga predicts Apple will offer OS X on x86 next year:

http://maccentral.macworld.com/news/0210/28.intel. php

Re:"That's Horseshit Jack!"

[lugonn]

"What? I'm supposed to buy this shit! 2,000 clock cycles and you can't find an OS to fit the bill? Come on Dave...you must be doing something seriously wrong."

Apple is a OEM. Why would they build OS X to run on a x86? To make 50,000 geeks happy, or to show Billy how to make a real OS from a BSD kernel? Where is the capital gain for Apple to switch Hardware platforms?

Re:And among other things, READ YOUR OWN POST

[veddermatic]

Yup it's not multisuer. Until you read the last part of your own post and "...create additional accounts using standard UNIX tools.

Dunno if you are attempting a misguided Apple bashing, or didn't bother to read what you copied and pasted, but golly....

Practicality: Anything else vs. HURD or Darwin

[intermodal]

I'm not saying that's incorrect. I'm just saying that it's not exactly going to convert anybody to be able to run OSX with dubious results on a PC. Running a Darwin kernel is like running the HURD kernel. Either way it's going to be a pain in the ass. Sure, it may be decent and viable. But until it's working on more hardware (esp. AMD/VIA) then it's a really niche thing, and not something that can be done without great efforts to find the right hardware unless you're very lucky (think modems on Linux)

Re:But where is the source code to the Carbon libs

[mcc]

I'm not going to address the question of whether Apple should be open-sourcing more than they already have-- i don't feel like arguing right now-- but let me try to clear a couple things up.

it would be really nice if **Carbon** was a platform independant library

I don't think Carbon is what you mean.

Mac OS X natively supports three APIs:

1. Cocoa, OS X's "standard" API. Cocoa is basically a newer version of NeXTStep with a new name. It consists of an extremely elegant object-oriented GUI application API sitting on top of POSIX. This is what Apple wants people to program new applications in.
2. Carbon. Carbon is a "transitional" API, that basically consists of an updated version of the old Classic Mac OS Toolbox, with anything related to unprotected memory, cooperative multitasking, or such things removed, and a lot of API cruft in general cleaned up. Apple estimates that about 20% of an average classic mac os program will have to be changed in order for it to work under Carbon. The recommended use for this is that if you have an existing codebase written for the Classic Mac OS Toolbox, you won't have to rewrite from scratch-- you can just carbonize it. Writing new applications in Carbon makes Apple sad, and it isn't as pleasant as writing in Cocoa, but people do it anyway becuase unlike Cocoa programs, Carbon programs can run under both OS 9 and OS X.
3. Java 1.4.

A program can have different components from each of the three groups above. Anyway, while i am not altogether certain abotu this next bit, it's been implied that due to some slightly legacy code, Carbon will NOT be supported away from the PPC or even if apple releases an x86 OS X. At any rate, unlike NeXTStep, Carbon was not designed as a platform-independent API, it's full of a LOT of macintosh-specific idiosyncracities, a small number of incorrectly-constructed Carbon apps will actually break if you put them on a non-forking filesystem, and it just wouldn't work very well on other OSes, i don't think. And besides this, it just isn't as good an API as Cocoa. You don't want it.

That said, Cocoa actually is available as a GPLed, cross-platform API! GNUStep is a third-party reimplementation of NextStep/Cocoa that follows Cocoa closely enough that porting between the two is somewhat trivial. There is no reason why you cannot use this right now.

Apple keeps complaining when OSS programmers emulate the look and feel of a Carbon application instead of calling the real thing.

No, apple keeps complaining when skin developers for other OSes copy the exact textures of the skins in Mac OS X. They also complain if people release applications whose interfaces are straight copies of iApps. I haven't seen them complaining about "Look and Feel" in a long time.

But if you want your application to not be tied specifically to MacOS X then your better off using winelib or wx for your widget set

Umm, why not use Java 1.4 and Swing? That's about as crossplatform as it gets. Wx would be ok too but Winelib doesn't seem like a great idea to me.

If Apple wants OSS programmers to use the real thing then they should provide the real thing to OSS programmers.

While it would be really cool if Cocoa were a cross-platform API like it once was, Apple really doesn't seem concerned with exploring that avenue right now. They seem to be of the opinion that if you want to write an OS X application and have it not tied down to OS X, that's what Java's for. Sorry.

Darwin

[Anonymous Coward]

When you leech Darwin, you get the BSD core, a few services, an apache webserver, and some modifications and tweaks that apple has done. That's about it. So, if you think you're going to load it up with a beautiful looking GUI with all the pretty buttons, realize that the CARBON toolkits (as i understand it, carbon powers all the pretty buttons and all the nifty crap that makes OS X so attractive) don't run at all.

NOT FreeBSD -based!

[otuz]

Darwin, the FreeBSD-based core of Mac OS X

The core is NOT based on FreeBSD. The userland is a port of the FreeBSD userland, yes. It would not be based on Linux, if the default userland were GNU-utils.
The core of Darwin/MacOS 10 is a Darwin (or NeXT Step) -based kernel/"core" running on Mach-nanokernel.. more alike GNU/HURD than FreeBSD (or Linux).

Re:NOT FreeBSD -based!

[karlm]

Good post, you just missed one small point: Mach is a microkernel. It's a big dirty-old man of a microkernel at that. Mach is a big reason GNU/HURD isn't performing as well as hoped. OS X gets reasonable performance out of Mach by using a monolithic server (as opposed to HURD's more modular multiserver approach) and running the server in the same address space as Mach itself (thus it's not a Machsever in the strictst sense).

I've run some fast microkernel/nanokernel OSsses on my x86 machine (BeOS, QNX, L4Linux) that all use much lighter-weight kernels and servers that run in user space. L4 and the QNX kernel each weigh in at about 1/10th the size of my maximally pruned Linux 2.4.18 kernel (everything compiled as modules, except IDE and ext2 support). Mach itself without the BSD personality probably is slightly bigger than my Linux kernel. I nuked my GNU/HURD partition last weekend, so I can't tell you for sure. L4-Hazelnut and the QNX kernel each have about 32k of compiled assembly and 32k of C++ code. Hopefully OS X will eventually migrate to a nanokernel and/or runing multiservers in userspace.

BTW - I wouldn't recomend L4Linux, at least a year agoit was less stable than Mac System 7. I think it was due to a poor job of making linux into a monoserver, as the debugging counters would keep rolling in the corner of my screen and there were no L4 panics/ Maybe in a couple of years L4-HURD or L4-Linux will be up to par. There are some reeally nice things going on in that area of research.

Re:NOT FreeBSD -based!

[karlm]

I neglected to mention a side benefifit of a nice clean nanokernel : good cache utilization. I had to run some memory benchmarks on several different architectures as part of a systems engineering class. Just for kicks, I ran the tests on both Linux and QNX on my home machine. Linux transitioned much more smothly to higher latency L2 cache and main memory usage. QNX had very rapid and clear transitions at higher thresholds. In short, QNX thrashed the cache less than Linux. It certainly helped that half the kernel fit into my L1 instruction cache. If you hand-linked and rearranged the Linux kernel, you could probably get close to the same effect by increasing locality of reference, but your time is better spent doing other things.

Darwin is no longer a micro-kernel

[AIXadmin]

Darwin is no longer a true micro-kernel. Maybe people will stop bitching. Apple has taken the best of both worlds and combined them.

Read Apple's Docs on there developer site:
"The core of any operating system is its kernel. The Mac OS X kernel is also known as XNU. Though Mac OS X shares much of its underlying architecture with BSD, the kernel is one area where they differ significantly. XNU is based on the Mach microkernel design, but it also incorporates BSD features. It is not technically a microkernel implementation, but still has many of the benefits of a microkernel.
Why is it designed like this? Pure Mach allows you to run an operating system as a separate process on the system that allows for flexibility, but can also slow things down because of the translation between Mach and the layers above it. With Mac OS X, since the desired behavior of the operating system is known, BSD functionality has been incorporated in the kernel alongside Mach. The result is that the kernel combines the strengths of Mach with the strengths of BSD.

http://developer.apple.com/techpubs/macosx/Darwi n/ PortingUNIX/additionalfeatures/The_Kernel.html

Re:Darwin is no longer a micro-kernel

[Aqua OS X]

Interesting... mod this guy up.

Re:Darwin is no longer a micro-kernel

[Lars T.]

Cough!

PC to Mac

[u19925]

I downloaded Darwin and loaded on my computer. Upon reboot, it evolved into Mac.

Re:"That's Horseshit Jack!"

[NetGyver]

50,000 Geeks * 99$ OS copy = 4.95 million dollars. How's that for capital gain?

At any rate, there's probably more geeks than that out there willing to spring for a x86 version of OS X and apple will probably charge a bit more. So the milage may very.

It's not just geeks either mind you. Apple has some serious skills when it comes to the GUI department, and could attract A LOT of joe-sixpacks if it'd move to the x86 platform.

I don't like to compare Apple to Windows, because, well it's like comparing apples to oranges (pun included), but IMHO Apple has just as much intuitiveness and ease of use as windows does, if not more.

Hypothetically speaking, if Apple moved today and pushed OS X on to the x86 and seriously worked on it, If they did their marketing right, i seriously believe Apple would have bitchslap privledges on Microsoft and could widen the playing field.

However, Apple seems to be fixiated on hardware sales, the OS itself it just a "tool" to push hardware. So who knows? It doesn't look like it's going to happen anytime soon though.

Although i'm a serious newb to linux, i did take the inituitive and download Red Hat 8, and i'm having fun learning the more technical aspects of it.

However I do sometimes long for the OS that is drop dead easy to install and use, stable as a rock, tons of apps and works great. All at the *same* time.

Re:Serious question... (serious results?)

[tyler_larson]

Only reason I can see for running Darwin is for Mac hackers who want to enhance the OS -- but that poses another question: does Apple accept patches?

This is a serious question -- what are the benefits of Darwin being open?

A little research reveals that Apple does in fact accepts patches and hopes to see real real help and real results from the open source community with their kernel.

So the real question about their open-source philosophy is, Does it actually work? In other words, are they actually seeing results, and are we really trying to contribute.

I know for a fact that I don't ever plan to contribute to Apple's open-source projects because:

• Darwin is relatively useless on x86
• I can't afford to waste that much money to buy a mac just to play with it. I can build a quality x86 box for just a few hundred bones.
• All of the fun Apple projects (Aqua et. al.) that I would be interested in tweaking are closed-source.
• There's already another high-quality open-source UNIX-compatible kernel that's much more widely supported and understood. If I want to do any hacking, I do it with Linux.

I realize that Apple has reasons for not opening their other projects, and I don't expect them to change their minds any time soon. But how much help can they really expect when they don't give us any incentive to work with them?

Did Apple decide to take this road because "open source" was just one of those buzzwords that translated to "free labor" in the minds of management? Do they really have any intention of listening to what hackers want, or do they just expect us to work on anything that calls itself "open source"?

Re:Supported Hardware - USE THE FORK!

[Twirlip of the Mists]

That's exactly right. People under 18 (in the US; varies in other jurisdictions) can't legally waive their copyright-derived rights over their works. In order to release software under a license like the APSL or the GPL requires that the licensor waive certain rights that would ordinarily be protected by copyright. (I'm not sure if this is true of the BSD license, but I don't believe it is. The BSD license doesn't require the licensor to give up any rights, as far as I know.)

The important side-note to this is that any GPL'd software that includes contributions by persons who are or were under 18 at the time is being distributed illegally. By the letter of the law, anyway.

Darwin

[Zen Programmer]

Only Apple would have the balls/ingenuity/vanity to name their product Darwin.

In other news, Windows has released a competing OS codenamed 'God'.

An Intel CPU does NOT mean PC hardware!

[pauljlucas]

Hypothetically, if Apple were to switch to using Intel CPUs, i.e., x86s, that does not mean that you'd be able to install and run OS X on any old PC box.

All it would mean would be that Apple would unplug the PowerPC CPU from the motherboard and plug in an Intel CPU (plus whatever other motherboard tweaks were necessary to make this actually work: the pin-outs are different, for example). The G4 towers, iMacs, PowerBooks, and iBooks would all look exactly the same. You'd still be buying Apple machines.

Apple is in the business of selling hardware: their hardware. Plus, you can bet that Steve would never let OS X run on anything as aesthetically unpleasing as a typical PC box.

Re:An Intel CPU does NOT mean PC hardware!

[scrod]

The G4 towers, iMacs, PowerBooks, and iBooks would all look exactly the same. You'd still be buying Apple machines.

Heh, no they wouldn't. They'd be a lot thicker, bulkier, and noisier due to the extra fans they'd need.

Apple is able to develop these unique designs because of the efficiency of the PowerPC processors they use.

Re:An Intel CPU does NOT mean PC hardware!

[g4dget]

Well, except for the CPU, what's different? Macintosh uses the same bus, the same graphics cards, and most of the same I/O ports.

Re:An Intel CPU does NOT mean PC hardware!

[pauljlucas]

Well, except for the CPU, what's different?

The case design, obviously.

Re:"That's Horseshit Jack!"

[BinxBolling]

50,000 Geeks * 99$ OS copy = 4.95 million dollars. How's that for capital gain?

For a company the size of Apple, not very good. That $5 million probably wouldn't even cover all of the development costs. Good engineers don't come cheap, and the engineers are just one part of the process of making a real software product.

Re:Serious question... (serious results?)

[jaoswald]

All of the fun Apple projects (Aqua et. al.) that I would be interested in tweaking are closed-source.

Hmmm. Perhaps Apple is keeping all these things closed source so you CAN'T indulge your interest in "tweaking." Perhaps these elements are fundamental enough to Apple's core goal for OS X that they don't want to be distracted by hundreds of amateurs submitting what they think are improvements. Instead, these tweakers will proliferate yet more useless "skins" for Linux desktop environments.

The kind of thing I'm sure Linux hackers would love to add: X window "compatible" cut and paste behavior, various redundant widgets with unpolished appearance and behavior, font rendering "optimizations" that gain 10% in throughput while adding 100% in butt-ugliness, etc. Thank god you can't add those to Aqua.

Re:Ports, Linux Emulation, Layout but...

[friedmud]

One word:

Gentoo

I actually have a friend that switched from FreeBSD last year to Gentoo Linux.

It is a complete ports system (compiling everything from source, with auto-dependency resolving). It also has a very cool init system (with dependency checkin and resolving) along with other cool features.

Give it a try - it is GREAT, I simply can't use anything else now.

To install KDE X and everything they depend on you just do:

emerge kde

It then downloads the source for KDE, X and everything they depend on, compiles them for your hardware (mine are all athlon-tbird optimized), then installs them.

All very nice and tidy.

Check it out - you will be amazed.

Derek

GNUStep versus GNOME/KDE?

[cpeterso]

Second, the Cocoa API is more or less source-code compatable with GNUStep. What is GNUStep? It's an open-source implementation of the Objective-C OpenStep APIs on top of X11. What's OpenStep? It's the open standard that NeXT released and implemented and eventually became Cocoa. You can write full-fledged OSX applications that cross-compile for GNUStep on Linux TODAY.


I've been researching GNUSteplately and wondering why it doesn't get more high profile attention. The GNUStep framework seems to solve many of the same problems that GNOME and KDE are trying to solve. However, instead of reinventing the wheel, GNUStep uses a time-tested API design that is source compatible with Mac OS X (a platform many people consider the pinnacle of user-centric Unix). What can GNOME and KDE do that GNUStep/OpenStep cannot?

I believe the biggest problem for GNUStep is that few people use Objective C. That is a big speedbump to people adapting their legacy code.

Re:GNUStep versus GNOME/KDE?

[int69h]

I believe the biggest problem for GNUStep is that few people use Objective C. That is a big speedbump to people adapting their legacy code.

Not really. Objective-C is a superset of C, so C code is not a problem. You can also instantiate and use C++ objects from Objective-C, so C++ code is not a problem. As a matter of fact, you can instantiate and use Java objects in Objective-C code. Anyone who's familar with Java or C++ should be able to pick up Objective-C in an afternoon.

Re:GNUStep versus GNOME/KDE?

[cpeterso]

cool, I did not know you could call C++ code from Objective-C. Is that true of all Objective-C implementations or just Apple's Cocoa environment?

Re:GNUStep versus GNOME/KDE?

[mcc]

I've been wondering the same thing for a really long time. It seems like GNUStep could be fantastic for The GNU World, but they seem continuously resource-strapped and unable to get public attention. I really don't know what gives, but i think the best possible answer, you've already said:

I believe the biggest problem for GNUStep is that few people use Objective C. That is a big speedbump to people adapting their legacy code.

You really probably have hit on about 90% of the problem right there. We will just have to see what happens with this and projects like it-- i mean, it looks very promising, but then you have to compare it with Cocoa, where objects written in Objective C and Java (and now, due to third-party efforts, i think even Perl and Python are 100% supported) can interact with each other and the OS APIs seamlessly and without any fuss.

Come to think of it, ProjectBuilder, the thing that links all those Java and Objective C thingies together when you're coding for Cocoa, supposedly is just a front-end to the GCC. Supposedly all the backend stuff is open-source utilities that are part of Darwin. Hm. Is Apple's Objc-java bridge open-source or no? If it's part of the closed-source Cocoa stuff, how does that work out? (And won't even cocoa services run in Darwin as long as you stay in the Foundation and don't touch the AppKit where all the GUI stuff lives? What about if wrote that service in Java?) If it's open-source, then why hasn't GNUStep integrated it yet? Why haven't they integrated the Perl/Python bridges to Objc? Is this just the whole continous GNUStep-limited-resources problem, or what? Isn't there some thing going on where GNUStep doesn't use objc_msgsend(), it just pretends to? Is that a problem here?

I dunno. I really think if more people could get over their intense fear of the [square brackets]; they would fall in love with NeXTStep :) The ability to write in multiple languages would at least make people feel more secure about the thing..

I keep meaning to try to port a couple of the open-source Mac OS X apps, probably one of the excellent GPLed AIM clients, to Linux using GNUStep as a proof-of-concept, or something, but i keep not getting around to this. Hopefully i'll be able to give it a try in a couple weeks, after the mononucleosis gets a little better :(

Does anyone know if the InterfaceBuilder analogue in GNUStep is anywhere near as good as Apple's? The whole nib-file concept is i think the most brilliant and useful thing about Cocoa.. eh.

Re:GNUStep versus GNOME/KDE?

[g4dget]

I believe the biggest problem for GNUStep is that few people use Objective C.

That, and the fact that its imaging system is based on Postscript.

That is a big speedbump to people adapting their legacy code.

Why should they bother? Gtk+, KDE, and X11 are much more widespread, and they give you a much wider variety of languages to program them in.

Re:GNUStep versus GNOME/KDE?

[bnenning]

Hm. Is Apple's Objc-java bridge open-source or no? If it's part of the closed-source Cocoa stuff, how does that work out? (And won't even cocoa services run in Darwin as long as you stay in the Foundation and don't touch the AppKit where all the GUI stuff lives?

The Java bridge isn't open source, and unfortunately neither is Foundation, so even Cocoa command-line tools won't run on a stock Darwin install. That's unfortunate, and Apple really should release Foundation, especially since most of it is apparently a thin wrapper over CoreFoundation, which is open source.

I really think if more people could get over their intense fear of the [square brackets]; they would fall in love with NeXTStep :)

I agree completely. I too had an irrational distaste for Objective C based on the "weird" syntax, but after using it you realize that it does a lot of things right that most other languages do wrong. It's interesting to compare it to Java, where Sun made some design decisions that were technically inferior but probably necessary to get developers on board, such as:

• Method syntax. Java had to look like C and C++ so as not to frighten anyone, but Objective C is more readable once you're used to it. "[table setObject:foo forKey:bar]" is much more self-documenting than "table.put(bar, foo)".
  
• Factory methods vs constructors. Java continues the inane tradition of treating object creation as some mystical act that requires its own operator and syntax, while to create an object in ObjC you just send the appropriate message to the class object. Compare "new Integer(1)" with "[NSNumber numberWithInt:1]". Java requires you to name the exact class to be instantiated, as well as hard-coding the act of allocating a new object. The ObjC method can return an instance of a private subclass and/or a cached object, and is much cleaner from an OO perspective.
  

I could go on for a while, but I'll stop there. And in fairness there are some things I'd like to see ObjC take from Java as well, specifically namespaces, garbage collection, and lack of header files. But overall Objective C is a great language, and Cocoa is an even better API.

I keep meaning to try to port a couple of the open-source Mac OS X apps, probably one of the excellent GPLed AIM clients, to Linux using GNUStep as a proof-of-concept

My game Gridlock runs on Mac OS X and GNUStep. I think there are 3 "#ifdef GNUSTEP" occurrences in around 5000 lines of code, all to work around minor GNUStep UI and Gorm bugs. It doesn't look nearly as nice on GNUStep as on Mac OS X, but like you I wanted to do the port as a proof of concept.

Does anyone know if the InterfaceBuilder analogue in GNUStep is anywhere near as good as Apple's?

Gorm looks and acts very similar to IB, but the last version I used was very buggy. In particular, it had an annoying habit of corrupting files when it saved them such that it could no longer open them, even though they would load fine at runtime. There's a new version out that I haven't tried yet.

Re:GNUStep versus GNOME/KDE?

[BlueGecko]

That, and the fact that its imaging system is based on Postscript.

The DPS backend hasn't been used for over a year. The current backend is raw Xlib; no PostScript involved.

GNUStep versus GNOME/KDE?

[cpeterso]

I've been researching GNUSteplately and wondering why it doesn't get more high profile attention. The GNUStep framework seems to solve many of the same problems that GNOME and KDE are trying to solve. However, instead of reinventing the wheel, GNUStep uses a time-tested API design that is source compatible with Mac OS X (a platform many people consider the pinnacle of user-centric Unix). What can GNOME and KDE do that GNUStep/OpenStep cannot?

If people truly do write new Mac OS X apps in Cocoa, then GNUStep could easily give those developers cross-platform support for Linux (and other GNUStep supported platforms). Don't Linux users want more "native" apps?

I believe the biggest problem for GNUStep is that few people use Objective C. That is a big speedbump to people adapting their legacy code.

Re:its all about the libs..

[Arandir]

This is actually a myth. Yes, the FSF believes it. No, no one is willing to challenge them in court. But dependency does not define derivation.

Mac OS X 10.2 uses a custom GCC 3.1

[truth_revealed]

Mac OS X 10.2 uses a custom version of GCC 3.1 according to this GCC mailing list announcement.
It was apparently tweaked to accomodate the GCC 2.95 C++ ABI for certain drivers and libraries.
I, for one, hope the GCC steering committee will allow them to add Objective C++ to the GCC main branch.

OT - Re:Anyone actually use Darwin?

[little_fluffy_clouds]

Hey you like, stole my .sig :]

Rhapsody X86?

[mbourgon]

Time to go back a little in the Wayback machine. I have a set of Rhapsody CDs, which (as I understand it) was Darwin + Mac_goodness, basically NeXTStep + Mac stuff. Now, I don't have the floppies for it, so I've never been able to test it. But, would anyone like to comment on how well it works, the state of Pre-Aqua on X86, etc, etc?

Re:"That's Horseshit Jack!"

[Ponty]

50,000 iMacs at $1100 = 55,000,000.

Mac OS X for commodity PCs = no Office for Mac = dead Apple. It's that simple.

Want an OS that is drop dead easy to install and use, stable as a rock, tons of apps and works great? Get a Mac.

Re:Supported Hardware - USE THE FORK!

[timster]

I don't see how you can come to the conclusion that the GPL requires the licensor to "waive rights" where the BSD does not. Can you give some sort of rational explanation for what you said?

In the meantime -- the GPL doesn't require you to waive rights, as it doesn't prevent you from doing anything with the work (licensing under other terms, etc). And if people under 18 can't waive their rights to SOME extent then it's not legal for them to distribute their works at all (which makes the copyright pretty pointless...)

My understanding is that the Apple thing came about because someone who is 18 cannot enter into a legally binding contract, and therefore they could not agree to _Apple's_ license.

Re:be Very careful with this release

[MouseR]

That's so miss-informed it's not funny anymore.

In large friendly letters

1. Fact:
2. Macs ship with a one-button mouse.
   
   Fact: Darwin (and Mac OS X) supports multi-button mice.

Re:"That's Horseshit Jack!"

[spitzak]

Apple does not have to work on "installation" so I doubt the result will be that easy to install. On the variety of PC hardware I doubt they can do much better than the best Linux or Windows installers.

In the real world, Windows is infinitely easier to "install" because for most people it is on the machine when they buy it. It is not physically possible to make installation of any operating system easier than that. Apple has this same "easy installation" for their small segment of the market.

Most people here think that if Apple makes an 86 OS/X they will also make their own boxes to run it. You will not be able to run it on Windows pc's.

Re:FreeBSD rocks

[vga_init]

I've made several attempts to use FreeBSD, but it never worked out. ;_; I will never give up on the OS and still insist that it rocks, but no matter what I do X-Windows does not configure right for me whereas in all linux distros it does. Hmm...I guess because linux is more often geared towards the more inexperienced people (that would be me).

If I were a guru I bet I could make it work, but alas, woe is me...

Re:atheist!=agnostic

[Snocone]

Actually, you've got it wrong too, agnostics DO believe something: They believe in unknowability.

Agnostic = "I believe is impossible to know whether there is a God."

Some modern commentators call this 'strong' agnosticism as opposed to what you think it is being 'weak' agnosticism, but that's silly.

Re:Supported Hardware - USE THE FORK!

[Twirlip of the Mists]

I don't see how you can come to the conclusion that the GPL requires the licensor to "waive rights" where the BSD does not.

The GPL requires all parties, including the author, to agree not to distribute derivative works under a different license. That's waiving a fundamental right of authorship-- the right to distribute one's works. The BSD license requires no such waiver.

LMBench results for Darwin, NetBSD, and Linux

[cpeterso]

http://clustermonkey.org/~laz/pbook/rob.lmbench.tx t

Here are some LMBench benchmarks that someone did comparing multiple releases of Darwin, NetBSD, and Linux. The general pattern is that (surprise surprise) Linux is the fastest. NetBSD is about 2x slower than Linux. Darwin is about 2-3x slower than NetBSD.

Re:LMBench results for Darwin, NetBSD, and Linux

[Leimy]

You didn't try NetBSD current. I have a friend with other benchmarks showing context switching and pipe performance of NetBSD to be higher than Linux... He has to publish it though. :)

Re:LMBench results for Darwin, NetBSD, and Linux

[Leimy]

NetBSD 1.6 beats linux in some areas [on PPC]... now if I could just legally get those benchmark results :)

Dave

Re:its all about the libs..

[Arandir]

but if you remove the library how useful is a program that can't run?

Completely irrelevant. Dependency != Derivation. If there is no derivation then copyright law, and thus the GPL, does not apply.

Re:But where is the source code to the Carbon libs

[spitzak]

You are mostly correct except for this:

Umm, why not use Java 1.4 and Swing? That's about as crossplatform as it gets. Wx would be ok too

It does not matter how well those work, for any real application development "native widgets" does not work, no matter how much you wish it did. The differences are just too great. Simple things like order (system a needs A before you can send it B, while system b needs B before A) can make it impossible to port your code without the differences percolating directly to the highest level. How else do you explain that virtually all Open Source development uses toolkits (Qt, GTK, FLTK, Mozilla, Fox, Tk, ...) that draw things at a low level and bypass any native widgets.

In any case, GNUStep, if it works, would be a very good idea. I don't have a good explanation as to why it does not seem to be succeeding, I know Gnome was looking for a toolkit at one time and they don't seem to have considerd using it. It may also be that it was too hard to make Windows-like programs using GNUStep.

Possibly the popularity of OS/X will help. GNUStep should make their #1 priority to clone Cocoa as closely as possible in such areas as widget sizes and shapes so that portable programs will work. If they do not do this then it will be just like wx where it is not much use for portable programs except for small demos.

Darwin Switch Commercials

[Oculus Habent]

I don't think we'll see any Darwin/x86 Switch commercials...

I was typing this shell script on my dad's RedHat box, and it was a really good shell script, and I went to save and it was like "bleep bleep bleep bleep" and, like, half of it was gone. And I was like, "unnnh?". And I had to write it over, but it wasn't as good 'cause I had to do it fast...

Re:There's a reason for this...

[spitzak]

Actually there is nothing stopping the Carbon emulation layer from working on x86. It just needs to be compiled and the programs using it recompiled. I think you are thinking about the OS9 *emulation* which actually runs programs that are compiled for the PPC, that probably won't be ported. But all the sample programs you list are actually compiled on OS/X.

Most people think that if Apple made an 86 version, they would also make their own hardware. So I don't see how that would be different to MicroSoft than the current situation. Then again, it might be easy to get a Wine-like emulator on the system, so maybe MicroSoft would be worried about it.

Excellent, another OS to play with...

[PyroX_Pro]

Linux, BSD, MacOS, Darwin, Minix, as long as it is not a Microsoft crapplication, who really cares?

It all comes down to your happiness. If you can use Mac OS X and get everything done you need to, that's wonderful, good for you! *BSD vs. Linux, does it matter? Does what your neighbor use really matter to your productivity or usage?

Both will exist, for a long time, both have a large group of supporters.

The only time to get upset is if your workplace decides to force everyone to switch from a Unix based OS, to an Win/NT solution.

BTW: I am getting Darwin just give support of its availability. I love supporting non-Microsoft software of any kind.

But my preference is Linux, either Slackware, or RedHat depending on Server or desktop.

98% downloaded, time to go play ;)

Re:If for some reason

[Anonymous Coward]

actually that is not correct. You are probably still thinking of the days when the OS required Apple ROMS on the logic board. This has not been the case for 4 years now.

The main issue is one of "endian-ness". PCs are little-endian and Macs (and many other non-PC platforms) are big-endian. This byte-switch either has to happen in software or hardware. In the case of graphics cards its better to have it happen in hardware for performance reasons. If a manufacturer does not include the appropriate code for the byte-switching then you can't use a PC card on a Mac (and vice-versa). Some however have been able to "flash" the ROMs on certain cards (ATI mostly) since they actually manufacture and write the drivers for their cards. Unlike NVIDIA which only make the chip set, other manufacturers make the cards. Apple is either making the cards for their Macs or contracting someone else to make them.

Make sense now?

Re:Supported Hardware - USE THE FORK!

[benedict]

% perl -v

This is perl, version 5.005_03 built for i386-freebsd

Copyright 1987-1999, Larry Wall

Perl may be copied only under the terms of either the Artistic License or the
GNU General Public License, which may be found in the Perl 5.0 source kit.

Re:Will Linux EVER support HFS+?

[90XDoubleSide]

The FSF doesn't like this part of the APSL:

(c) You must make Source Code of all Your Deployed Modifications publicly available under the terms of this License, including the license grants set forth in Section 3 below, for as long as you Deploy the Covered Code or twelve (12) months from the date of initial Deployment, whichever is longer. You should preferably distribute the Source Code of Your Deployed Modifications electronically (e.g. download from a web site)

Whether that stops other people from using the code is another question.

Why I think Apple releases Darwin/x86

[PinkX]

There already is a huge base of x86 users around the globe, and inside it, an ever-growing base of GNU/Linux, *BSD and all sort of *NIX and opensource enthusiasts. That gives them a pretty good testbed for debugging the system, and making it more compatible.

Backporting it to a completely different platform from what it was originally developed (FreeBSD/x86 -> Darwin/PPC -> Darwin/x86) is a pretty good sanity check in order to see they didn't break what was already there, gives them a good shot on portability (think byte order endianness) and gives them a nice try on moving from their current platform (Motorola PPC) to some future versions (IBM Power4). By making the base system more portable, it's just a matter of recompiling the upper layers (think GUI, APIs, etc.) to asure potential future compatibility.

I don't think the Darwin/x86 release is due to enter the *nix market which is already dominated by the various *BSD flavors and GNU/Linux. Besides what I've previously said, it shows commitment from the Apple people to the OpenSource community.

x86? NOO!! it's really IA64

[Nicolay77]

Think about it. They compile their kernel to x86 only to know if the source is still multiplatform.

So we probably have already got everything we could ever get from apple for x86 from OS X.

But if they're going to switch to intel, they better do with IA64. It's better, x86 compatibility means nothing for them, and having the options open means that they can push IBM to offer good prices, same for Intel if ever taken in consideration.

Re:One acronym: API

[SmittyTheBold]

The problem is, unless there is a DAMN GOOD REASON, nobody will upgrade to a system that breaks all previous compatibility. People migrated to PPC because their old software would work. People moved to OS X when they could use their old software. Peopel upgrade to new versions of Windows because it usually doesn't break old software. It's not a problem for currently-developerd software, but I sure don't want to lose all the software I've built up over the years. A good portion of it has not been updated recently, and only works now because the last few (major) releases of my OS have not broken compatibility.

Re:Serious question... (serious results?)

[snowlick]

All of the fun Apple projects (Aqua et. al.) that I would be interested in tweaking are closed-source.

And thus you've hit upon the main reason open-source stuff survives: someone finds it to be fun. Visual stuff gives the most immediate reward, so it gets the most attention. Other things, like darwin streaming, deserves a good deal of attention too.

Why switch?

[BalkanBoy]

My name is Jeremiah Cohick. I bought my first Mac (an Apple 466 iBook Graphite SE) on January 30, 2001.

I'll be honest. Prior to actually using a Mac, I was a basher (the people who scoff at anything that doesn't say Linux on it). I would probably still be a basher today if my C++ teacher hadn't dared me to deviate from my GNU/GPL world. The first time I used Mac OS X, I wished I had never spoke a single word against Apple. Apple enlightened me beyond Linux and refined my thinking. Java and UNIX *really* are great things. Who knew?! In July 2001, I moved to a Power Mac G4. Now, I'm developing with Mac OS X Server and it is the most phat system in the world.

It doesn't freeze. It doesn't crash. It doesn't require me to reboot when I change network settings. It boots fast. Its sleep function actually works. I don't need virus software. There is no system tray. It lets me run a gazillion media and programming applications without taking a performance hit. And (even though my fellow programmers won't admit it) it is beautiful.

My Mac never requires technical support. It is fast, efficient, and compatible with the rest of the world. Everything is better on the Mac overall.

Warm regards, Jeremiah Cohick

Re:Supported Hardware - USE THE FORK!

[karlm]

The GPL requires all parties, including the author, to agree not to distribute derivative works under a different license. That's waiving a fundamental right of authorship-- the right to distribute one's works. The BSD license requires no such waiver.

If the author is the original and sole author this is not true. This is only true if the work contains GPLed code written by others. Most of the software produced by the Kompany is dual-liscenced.

Re:be Very careful with this release

[commodoresloat]

uhhh, yeah, I know. Sorry, it was a dumb joke, based on the fact that every time someone mentions Macs, somebody whines that they want another button on the mouse. But, yeah, it wasn't very funny, I suppose. Oh well, back to clicking my one button.

Re:I can respect your opinion about Linux, but

[troc]

He's entitled to his opinion - as long as he's not reciting it just to get a rise out of people.

i.e. he's stating an opinion "I think Jesus is an Asshole" rather than the flamebait "Jesus is a Fucking Asshole"

They are both very contentious statements and will result in serious and heated replies but one is valid in that it is constructive - it gives an opinion and a place from which to discuss - whilst the other is simply an invitation to flame back and generally piss people off.

Now I don't actually know Jesus so I personally can't comment on whether he is an Asshole or not. I was just using this as an example.

People taking offence, purely for the sake of taking offence are as bad as flamebaiters IMHO.

Troc

Re:Ports, Linux Emulation, Layout but...

[friedmud]

I already use it. I was saying that I know someone who used FreeBSD but now uses Gentoo - and is very satisfied with the move.

But I do use it on a day to day basis. It is my desktop at home and at work - and runs all of my servers. It is a great multipurpose distribution because you install exactly what you need and no more.

Derek

iso mirrors available

[Jean-Pierre]

opensource.apple currently redirects users to opendarwin.org to fetch the iso, but unfortunately opendarwin.org has been taken offline. in the interim the following mirrors are available...

darwin/x86 iso
http://web.mit.edu/darwin/www/darwinx86-602.iso.gz
http://enigma.us.itd.umich.edu/darwin/www/darwinx8 6-602.iso.gz

darwin/ppc iso
http://web.mit.edu/darwin/www/darwinppc-602.cdr.gz
http://enigma.us.itd.umich.edu/darwin/www/darwinpp c-602.cdr.gz

md5 checksums
MD5 (darwinx86-602.iso.gz) = d4e9a94c48d900736fa9f77d42707d50
MD5 (darwinppc-602.cdr.gz) = 07d4614c4e3b417f0022a97cf941ad97

installation instuctions
http://www.opensource.apple.com/projects/darwin/6. 0/install.x86.txt
http://www.opensource.apple.com/projects/darwin/6. 0/install.ppc.txt

Don't be sorry sweetie.

[BoomerSooner]

Just don't do it.

(end of Payback reference).

You are correct however, why would you bother developing something for an architecture you don't run? Seems kinda silly to me.

Why did they do this?

[jmanning2k]

Simple. Free testing of their ix86 based code. Rumor has it that they are keeping that option open should PPC architecture fall far behind that of ix86. Making this release just allows them to have it publicly tested and improved.

It's a back burner project anyway, but having to make these releases makes it a bit more formal and forces them to achive certain usability goals. Plus, it helps ix86 development keep up with their main Darwin core.