Slashdot Mirror
OpenDarwin Project Shutting Down
niabok writes "According to a message sent by Rob Braun to the OpenDarwin mailing lists, the OpenDarwin project will be shutting down, saying that 'OpenDarwin has failed to achieve its goals in 4 years of operation, and
moves further from achieving these goals as time goes on.' The project's servers will remain online long enough to allow developers to move their various projects elsewhere."
I guess they needed more intelligent design.
Help poke pirates in the eyepatch, arr.
ClosedDarwin?
With a PageRank of 8 and an age of 4 years, that domain will sell to some SEO company very VERY fast. I wonder what they'll get for it.
Stay tuned!
I'm not fat, just big boned...
I personally use Fink (and love it, for all of its flaws), but it's sad to me to see a good alternative source for OSS on OS X bite the dust. The only reason I'm able to enjoy a proprietary OS like OS X is because of the availability of many of the best OSS packages (if not all), and the compatability this affords me with linux-based environments. Hopefully Gentoo on OS X will go somewhere - does anyone know how it stacks up against Fink right now?
Too bad their dreams did not work out, but frankly, they will not be missed.
Sure, they ported fink and some libs to Darwin, but that's pretty much it. ODP has been dorman for years, since 2002, pretty much.
Is Apple to blame for their luck of support? I do not think so; since they do have a neat thing going with http://developer.apple.com/opensource/
Obama likes poor people so much, he wants to make more of them.
Apple never supported the open source version of darwin in any way beyond lip services, some server space, and releasing source packages in mostly unbuildable form. They took from many open source projects but returned precious little to the community. At the end of the day Apple does what immediately benefits Apple. It's sad, but it's likely the threat of hacking OS X to run on white box computers likely is the greatest reason for Apple to not release vital parts of the latest OS X source code. Yet this will still happen. In the meantime, Linux continues to grow and become better all the time. There just was no need for OpenDarwin without Aqua. If all you want is a unix-like OS to run servers, Linux suits the bill just fine.
If I were a better programmer, I could make the project I've wanted: an update of GNUstep to be more library-compatible with OSX, and an OS using it with Darwin.
-uso.
What you hear in the ear, preach from the rooftop Matthew 10.27b
you loose.
Well, I don't really get how much "open source" there is in Apple's effort. To me it looks more like "open-source compatible". In other words, with tools provided you can compile your open-source software (read: linux code) for Darwin. However I don't really see a full open-source effort. To me open-source means that you have to release the source one way or another, and Apple doesn't release any piece of source code. It's not enough to be based on FreeBSD to acquire the status of "Open-Source".
Sorry, this is another of those marketing schemes of Apple's. In fact it's one of the main reason I am staying away from it.
P.S. What is Microsoft did the same?
I sometimes wonder why this isn't an ongoing project like Wine. I think it's two things: 1) Few developers are interested in both Macs and asm. 2) It's such a narrow target. I mean, what Mac apps would you want to run on Linux that you can't find a windows version to run on Wine or some windows emulator.
How we know is more important than what we know.
I wonder what will happen to DarwinPorts.
Uttering logically derived and empirically supported truths to the disciples of the orthodox establishment.
I work with a lot of users and all they talk about is how great it is but when I mention Linux they respond with "But OS X is Linux" I try to tell them that OS X actually uses the BSD kernel with some parts of open source projects(mostly KDE) but they say "same thing" it really pisses me off(especially as a Linux user). I cannt imagine what the BSD developers/users feel. Anyway I always though the Apple commitment to open source was half-ass/shady. blah I just hope more Apple users smarten up and switch to Linux or a real BSD system.
I guess I am just dense, but I actually googled around for a bit before realizing that I had been had.
I tip my hat to you on that one, even though it should have been obvious, ya got me....
It is your personal duty to fight for what is right on a daily basis. Ignoring injustice is identical to approving
When it is the uninformed user who keeps upsetting your equilibrium? All Apple has done is release source, adopted open source products internally, and contribute to other open source projects.
If Apple users smarten up, it in no way changes what Apple has done and continues to do: use open source when beneficial.
GPL Deconstructed
I wonder if they're afraid that people would try to use the opendarwin kernel with mac os x for intel to run the whole thing on any machine.
As a Darbat (L4/Darwin) developer this is sad, and will be a bit of a set-back. We were hoping to try and become involved with the OpenDarwin community. I'm really sorry to see that this really handy resource will be going.
3) and you can get more binaries to run with wine than sources to recompile with gnustep...
Advogato readers will know the true shame is that opendarwin is closing and not gnu-darwin.
Smarten up...A real unix system...Linux/BSD is the only way...
As a Linux user of four years who has recently bought their first, personal, mac laptop I wish to call bullshit. I'd like to point out that OSX still plays a very important part in Linux development (less so in BSD) - specifically in regards to new features. Take for example xgl/compiz and xcompmgr which will be in full deployment for when Vista ships to compete with the M$ eye candy...Sure it only came into the lime light when Vistas beta's started shipping and the glory project status moved to them though without earlier projects like luminocity (etc) which was an attempt to add mac grade eye candy to Linux there wouldn't have been the ground work or the test case for this. And even now look at xcompmgr with transett or compiz - they just basically fashion themselves after inbuilt mac effects or 3rd party add ins that have existed for awhile under OSX.
On top of that mac make computers end users like and OSX Just Works(tm) which for a Linux user is really handy some days when Debian sid decides it wants to blow the heads off all the toys. It also interconnects flawlessly with my other Linux boxes through ssh, samba, nfs, vnc and everything I need (I use Fink for random unix tools I need).
Lastly OSX shows every day users that there are Real! alternatives to Windows that don't have the stigma of To Hard attached that they can try and enjoy. So really outside the RM ethos of everything should be open (to which, hypocritically in context of the above, I subscribe) there isn't really much reason for a mac user to smarten up and switch, try maybe, to Linux (and a mac user wouldn't touch BSD).
I ate your fish.
[Re: GNUStep]
I sometimes wonder why this isn't an ongoing project like Wine.
Well, basically, it's like this: the people who know enough to work on it are, for the most part just using Mac OS X, and most of the Linux crowd can't really tell the difference between GNUStep and Gnome (ie, they actually believe Gnome is good enough).
The upshot is that the contributors to GNUStep are a very small number indeed, and it's amazing how far they've gotten with so few people working on it.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
We feel "who cares what other people are running?". I feel the same about OSX users as I do about windows users, I don't care at all. Linux users on the other hand do bug us, since they do such stupid shit to their servers, and then when we take over when they get fired, it takes forever to clean up the mess.
I started out using Fink but it never felt quite right. Then I tried DarwinPorts and I've been happy ever since. As a result, when I saw this story my first thought was, "What will happen to DarinPorts?" I checked the Darwinports Mailing List Archive and found this comforting post. To summarize, DarwinPorts is alive and well and will continue. Time to start using www.darwinports.org rather than www.opendarwin.org.
OpenDarwin is dying. Another crippling bombshell hit the beleaguered OpenDarwin community, dot dot dot
Listen, it's been over twenty minutes since this story was posted, and I haven't seen a Netcraft confirmation post yet. How do I know it's really dying? C'mon, people, get on the ball.
No need for amarok, just do this. ls * > ./playlist && mplayer -shuffle -playlist ./playlist
Or
ls */*.mp3 > ./playlist && mplayer -shuffle -playlist ./playlist :)
mplayer for life, bitches.
Linked site says "Page Not Found"
Duct tape is like the Force. It has a light side, a dark side, and it holds the universe together.
Most of the linux crowd don't know why gtk+ Qt because they believe C > C++. What's funny is that I've heard of Python users who believe this. Last time I looked the PyQt bindings were way cleaner than the PyGtk bindings. Blah, whatever.
:)
Personally, I have no idea why people want to run any proprietary software on their Linux box, except maybe games and "shit you can't live without". Maybe no-one who uses a Mac ever migrates to Linux.
How we know is more important than what we know.
to switch to GNU Darwin (http://www.gnu-darwin.org/)
I haven't used it myself, but it seems to be more of a full system (with GNOME and WindowMaker) and more actively developed than OpenDarwin ever was.
Gee... You mean lifting large chunks of code from other free operating systems to create a slow and very limited OS, and then imposing restrictive license terms on that free code, somehow doesn't automatically lead to an OSS project everyone wants to jump on???
I'm SHOCKED! Shocked I say!
Somehow I don't think the end of OpenDarwin is going to mean Apple will stop lifting code from the BSDs.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
I try to use Fink when I can, but I like to stay on the bleeding edge. Everything in Fink seems so out of date. Ruby is still version 1.8.2, for example. That's the version that ships with Tiger in the first place!
I don't like to complain, though. Fink is still a wonderful concept. I just wish its admins didn't already have so many things they need to dedicate their time to.
In the meantime, though I do my development work and some testing on my Powerbook, my stable test server is an Ubuntu box. It's just easier to keep up-to-date, what with apt-get and all.
Natural selection at its finest!
So much for OSS "community" stepping up to the plate. What, is it only if you're taking on Microsoft that you guys give a damn about a project? And it's not a shock that many of you OSS devs were mooching off of OpenDarwin's servers to host your insignificant little projects, while contributing nothing to the OpenDarwin project itself.
The idea of improving GnuStep so it is binary compatible with the OS-X api is so you can run OS-X binaries.
The fact that I can't think of a single OS-X binary I'd want to run for which there isn't a corresponding windows binary might has something to do with why no-one wants to improve GnuStep in this way.
How we know is more important than what we know.
Look, not to get too preachy, but that's the problem with FreeBSD and projects with other, equally open licenses, like MIT, etc. As I said in my not-so-tongue-in-cheek blog to the audienceless ether years ago...
These licenses [X11, BSD, MIT] don't do enough to protect the contributions of the people that made the code -- they essentially enable legalized plagiarism. It's certainly one's right to make code that's this unregulated, but these licenses are from nearly overly altruistic motavations.
I'm using OS X right now. I'm happy FreeBSD enabled its creation. I'm posting from Safari. I'm happy Konq's code helped Apple build this very fast, mature browser. Without totally free and open licenses like the ones I wrote about, above, we wouldn't have this OS X.
Yet at the same time, this happiness doesn't change that I wish Apple would have partnered with GNU/Linux. We'd see a very different OS X and a very different collaboration (some would argue a "collaboration" would be a new thing, and I believe I agree) between Apple and the GNU hacker community today. Linux has not yet come close to hitting the tipping point on the desktop for the typical semi-technical user. With Apple's help, it would be much closer. With BSD's sabotage -- the license -- that help and the FreeBSD code has been thrown into the closed system of consumerist capitalism.
It's all 0s and 1s. Or it's not.
So, what were it's goals?
Dave
I write a blog now, you should be afraid.
You could use Linux and use wine or cross over to run the win version of iTunes.
I use gtk+ instead of Qt because the licensing. Either way there's gtkmm that uses c++ anyway so your argument there is moot.
"You're everywhere. You're omnivorous."
> I mean, what Mac apps would you want to run on Linux that you can't find a windows version to run on Wine or some windows emulator.
OsiriX Medical Imaging Software: http://homepage.mac.com/rossetantoine/osirix/
The code is written for MacOS X, using Cocoa I think. It is free software (GPL), but requires proprietary libraries to work.
On the converse side, frhed, a great free (GPLed) Windows hexeditor could simply not run on Linux/BSD if it weren't for the existence of a free compatability layer (wine/winelib).
Yes, Linux/BSD needs GNUStep to become a better compatability layer for OS X apps, if only for source compatability with the GPLed/etc OS X applications that exist. Yes, re-implementation is important for Linux/BSD - witness GNU Classpath and all that free java, witness osflash.org, witness wine/winelib, witness octave (matlab thing). Free sofware/open source *cannot* afford to ignore proprietary APIs and to some extent ABIs (mainly to bring people closer to Linux/BSD).
For those who want to pay for MacOS X, there is always Mac-On-Linux, but obviously that still ties you to non-free software (as well as Linux).
No, see, the reason why Qt owns over gtk+ is simply the fact that Qt is internally a work of art, and gtk+ is internally a dog's breakfast.
How we know is more important than what we know.
In my experience, the 'achievement and goals' mentality seem to stem from a corporate-culture, while open source projects generally have a vision but the vision can change over time and the 'goals' change with the community. In general what I've noticed is that projects that concern themselves with the community rather than the goals tend to stay afloat longer than those that have a directive. If opendarwin became a place where osx-focused oss projects were housed / referenced, then why call that a failure? Perhaps it wasn't what the creators of the project had in mind initially, but it seems pretty unfortunate to loose this resource in the community.
It seems to me that this is a fundamental problem and it shows the total failure of the "Open Source" cooperate with proprietary approach over the "loony" free software people who want only to cooperate with those who share.
Analysed from the Open Source point of view, inevitably Open Darwin and it's "bazzar" development should have quickly overtaken Apple. Soon it would have become the main source of Darwin code.
Analysed from the Free Software point of view, Open Darwin had at it's heart a cancer of Apple's proprietry code and proprietary control. The death of the project was inevitable. Apple would always be losing control, the more code moved into the free parts of OS X. Anybody sensible could see that contributing to such a project where inclusion was controlled by could only advance Apple's agenda and would be largely used against users.
The lesson to be drawn is that the elimination of proprietary (and I DO NOT mean commercial) influence from Free Software projects should always be the first priority.
Think of the amount of effort that could have been spent improving Linux or OpenBSD to be better in every way as OS X and has now been lost to Darwin. Think of the extra effort that can now be made in future due to the collapse of this project. Try to apply those efforts so that the work of OpenDarwin has not been entirely in vain.
Yes perhaps, but Qt does stand a chance against mono or (c# + gtk + U). The way I see it the dog has ate its breakfast and is now running past its former competition...
Regards,
Yes perhaps, but Qt does not stand a chance against mono or (c# + gtk + U). The way I see it the dog has ate its breakfast and is now running past its former competition...
Regards,
of the fittest. Darwin was right yet once more then..
<before>now</before>
You should now find you have more than 5000 packaes instead of 1800 to choose from and the latest version oof PERL, Ruby, KDE etc. are all there. You will have to update all your old packages to use them though, with Fink you can either choose stable or unstable, not a mixture. Having said that I have over 1000 unstable Fink packages installed on this mac aand they work fine.
Happy finking.
Losing is the second.
But when you do it while claiming to be someone famous they come to take you away, ho, ho, he he, ha, ha!
KFG
Whats with mod for funny - I was being serious. Its how I listen to my music :S
You can wrap a turd in tin foil but you'll never make a jewel.
How we know is more important than what we know.
Having used both Mac and Linux. I've come to this conclution. Linux is powerful. Mac is powerful. But lets face it guys when was the last time you had to stay up all night getting an freakin program installed on a Mac?...that wasn't open source "cough". Linux works well as a server not a desktop and even then a simple sudo yum update on a production box can blow away the config files you spent 4 days setting up don't say you've never had any "simple" problems with linux also there not even one server management package that comes close to Apple's on Linux. Here is the way i look at it say you want a barn. You have two options you can buy one that is already made and start using it the way it was designed start being productive or you can build one which you may not know exactly how to so there's alot of stuff to learn and figure out before you're ever able to use it. There's an infinite amount of ways you can build it which has its benifits and disadvantages. Main disadvantges being time and energy. Apple has its "way" of doing things, because of this it is easier to control what you are able to do. When you have control you have stability. Now some of you may say blaaaa thats why we use OSS. Everytime i've gone to do something on a mac there's hardley ever a better way to do it than their way. WHICH WORKS!! at any rate some how i'll suffer through my Quad G5 with 8gigs of ram and my 30" Cinema Display with my working NVIDIA graphics card drivers......ohh pooo i wish i could have a fancy linux box :-P
Pile of wood or Amish construction!!
Seeing as you replied twice, here's my second reply: heard of Qt#? Most Linux distros won't even bundle mono from irrational fears of microsoft anyway.
How we know is more important than what we know.
http://www.musicpd.org/ is wonderful.
...and that is all I have to say about that.
http://jessta.id.au
They haven't failed. They've just found a way that doesn't work and leads to death. All part of the natural consequences of evolution.
But some are asking, what will happen to the Open Darwin Project's adorable mascot? With the project coming to a close, the outlook for all is bleak.
Oh! This just in: lovable OpenDarwin mascot Hexley has signed a deal with Diz Nay Studios, and will be starring in a series of cartoon platypus porn films, presumably in order to pay for a much-rumored PPC addiction.
Our hearts go out to poor Hexley in this dark time.
(mplayer fan myself, though)
molmod.com - computing tips from a molecular modeling
Uhm... You're mistaken. Some of Apple's open-sourced code:
And of course, there's more, in addition to all the other existing open source components which they use and contribute to.
There's even more which they don't release, and you can like that or not (it's a business decision to them), but you can't claim that they don't release code.
The "find" command is more flexible if you're looking to recursively list your mp3 files thay may be in directory hirarchies of varying depth... find . -name "*.mp3" > playlist
The threat of hacking to allow OSX to run on white box computers?
It's already hacked, and working stunningly well: I've used OSX on a Dell C640 laptop and it's that good that I'll buy a MacBook Pro when it goes Merom. Everything I want is on OSX, and it really does appear that Apple has taken the greatest ideas from OSS and polished them to the point of near perfection.
Part of me does wonder if Apple doesn't mind people running OSX on unsupported boxes to get them hooked to the point that they go out and buy the apple hardware. I now know many, many people who have done this: dip toes in OSX, find all is wonderful, buy Apple.
Only bit I don't like is the development environment, as I'm not convinced I want to learn Objective C, but the CARBON APIs may be the way to go there I suppose.
But back to the point - the white box argument for not opening source doesn't fly IMHO, as OSX is already on vanilla boxes near you right now.
Even though you didn't reply to me...
> Most Linux distros won't even bundle mono from irrational fears of microsoft anyway.
novell - check
opensuse - check
fedora core - check
debian - check
ubuntu - check
red hat enterprise linux desktop - uh... not yet.
mandriva - check
g3nt00 - ch3ck
so what _is_ this "Most Linux distros" thing you're smoking?
Why not Amarok?
Not a flame, just curious. I actually prefer it to iTunes...
For the last couple of years, I have stumbled upon OpenDarwin once every 2-3 months and thought I'd really love to install and use it at some point. However, I always looked at the hardware support list and decided it probably wasn't ready.
But there was always the dream..
And now it's gone..
Will program for karma.
Unfortunately that particular large ape lives on....
Neerrrrrrrrrrd. ;)
'If you're flammable and have legs, you are never blocking a fire exit.'
The pkgsrc project www.pkgsrc.org supports Mac OS X. The packages it contains are much more up to date than either Fink or DarwinPorts, and can also be used on a number of other Unix like operating systems. I bought a Mac at the beginning of the year, and intended to wipe the disk to install NetBSD. I ended up dual booting it because I found I liked Mac OS X so much, especially when I can use pkgsrc on it.
This is really a shame. Great projects such as ncutil were built for darwin and are very useful within an OS X environment.
Sun really has the right idea, with an open source operating system. Darwin should stay open source; open source is a great idea.
A partition has been setup for people to beg for the open darwin project to be kept alive. I am signing!
Any one else fell like groveling?
Qt is internally a work of art
This is precisely why GNUStep gets no traction: the Linux crowd actually believes that a cross-platform abortion like Qt is acceptable. Of course, this isn't surprising for a community that still hasn't admitted to itself what an abomination X11 is.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
Qt, a toolkit written in a language that Trolltech finds so deficient that they extended it, a language that has such a baroque library that Trolltech wrote their own. As for Glib and GTK+, I have been through a fair bit of the code, and it is a work of art. I'm not saying that the Qt code isn't a work of art, but your criticism of Glib/GTK+ is bullshit.
If anyone had been keeping up with Rob Braun's musings about Open Darwin and Apple's behavior with the OS community, this decision was simply not a matter of 'if' but 'when.' The following links below illustrate that this wasn't a spur-of-the-moment decision but rather the final straw:
A Brief History of Apple's Open Source Efforts
WebKit and Apple's Open Source Efforts
Those are just for starters. And to top it all off where Braun gets to the meat of the matter:
Why Darwin Failed
It doesn't take a degree in rocket science to figure out that the holdouts on the Darwin project have finally had it with Apple.
In a nutshell: Apple have never let anyone touch their code which is a twisted beige box-grade edition of FreeBSD. If something burps no one can help outside of Cupertino. Worse, Apple deliberately makes it nearly impossible to report bugs and allow for patches to be made. This extension of Jobs' secrecy policy is why some holes remain wide open while the rest of the *nix world have patched them a long time ago.
With OpenDarwin shutting down not too long after Apple closed down OSx86, Apple execs selling Apple shares all over the place, and the exodus of two former NeXT gurus, it isn't hard to see what path Apple and OS X are heading down.
Go ahead and mod me as a troll for preaching against the Gospel of Steve, but if key players both at Apple and in the developer community do not believe in OS X (or are giving up on it entirely), how can the rest of us do so?
This is precisely why GNUStep gets no traction: the Linux crowd actually believes that a cross-platform abortion like Qt is acceptable. Of course, this isn't surprising for a community that still hasn't admitted to itself what an abomination X11 is.
*snort* - and people say that members of the linux community are rude!
There are shills on slashdot. Apparently, I'm one of them.
In addition to the use of find(1) suggested by another poster, you can also eliminate your use of a temporary file:
find -name '*.mp3' | mplayer -shuffle -playlist -
I'm not sure I can take an article seriously when the author does not know the difference between then and than.
Look at OS X's memory protection, demand-paging, preemptive multitasking, and GNU toolchain, present in other systems for decades before Apple got around to adopting them. And Apple didn't even write it themselves; they had to buy from a third party.
The point being, who cares? Basically you're saying that coming late to the party is unfashionable. I don't give a rat's ass about that, and neither do you -- or you would share the complaint I feigned above.
Cute Toy Story reference, but sid == unstable and is meant for Debian developers only. I use testing (etch) and haven't had the sort of drama you allude to, which I can't say of OS X.
(By the way, your saying "mac make computers" instead of "Apple makes computers" makes me suspicious of your Mac experience.)
Okay, my turn to call bullshit. From Terminal, ssh to a Debian box and run nano on a file at least several screens long, and page down. It doesn't update correctly. Manually setting TERM=vt100 works around this, but that's not "interconnecting flawlessly" in my book. Where's sshfs? Where's svn?
Bingo. Nail, meet hammer. I demand an open system, and that is why I'm switching.
In general, it is safe and legal to kill your children. -- POSIX Programmer's Guide
> There just was no need for OpenDarwin without Aqua.
Is it possible to run Darwin with Etoile / Gnustep? I think so.
I mean, if aqua misses why not recreate it.
And sure, you could run KDE on Darwin, so what's the problem?
I also am a bit dissapointed with fink, and am considering moving to Darwin ports. I was worried that it too would dissappear and am now relieved.
Thanks.
GNUstep is binary compatible with Cocoa, as long as you're on Mac OS X then you can configure GNUstep with the apple-apple-apple library combo and it'll build against the Cocoa.framework. GNUstep is also to a large extent source code compatible with Cocoa on other platforms, not surprising really as OpenStep was the original "write once, run everywhere" platform a couple of years before Java hit. If you want to run Mac OS X binaries on a non-Mac platform you've got a bit more work ahead of you, but take a look at NetBSD's COMPAT_DARWIN as a good starting point.
though not Free Software because it doesnt' seem to like GPL stuff much, like many corporations
Let's see what happens at WWDC. The writing's on the wall about Mach at Apple - it's slow and Avie left. Apple is keeping xnu for x86 secret and dropping weird hints about waiting and seeing.
So, Apple has to pick a new kernel. The allure of the NT kernel is hard to argue against (and others have given good arguments for it), given driver support, on a practical basis, but Apple has to realize getting into bed with Microsoft never benefits any company in the long term. So, what else is there?
Open Solaris? Not bad, but Sun is running it. BSD? Probably easiest to shoehorn in but not likely to be featureful enough for Apple. Linux? It has the momentum, the clout among the geeks, IT, and even Wall Street to a certain extent, and has all the features Apple wants, plus forkability. It's where Apple's former Alpha Geeks have gone.
The GPL is still something Apple doesn't embrace fully, but they managed to swallow their pride with WebKit and it's worked out well, so perhaps they've warmed. Let's just say I won't be surprised to see a version of linux running launchd with an IOKit layer at WWDC.
My next laptop is going to be a linux machine; I'm just not sure yet if it will be a Mac.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
I use OSX daily.... to ssh into linux and bsd boxes where all my real work happens,
and will continue to do so as long as ssh doesn't break.
The day Apple fucks things up so I can't ssh will be the day you find my box on Craigslist.
"the tiny bit of code sharing they have done."
... you can cavil about the usefulness of these packages if you like, but if Microsoft had released one percent of their comparable code it'd be a miracle. Hell, Microsoft's gotten proportionally more milage out of a smidgen of effort.
When you see reactions like this from the open source community it's no wonder Apple's gotten a mite miserly over recent months. Back when Rhapsody was first brought to the table Apple wasn't planning on even shipping the "BSD subsystem" beyond what was absolutely necessary to run the OS. I'm not talking about source code here, I'm talking about binaries. You were going to have to download your awk and sed and vi and grep and maybe even eventually the shell from Apple IF you wanted it. UNIX hackers the world over, inside and outside Apple, responded in shock... and someone managed to convince the Steve to tear down the wall.
From their point of view they've bent over backwards, going far beyond the requirements of the FOSS licenses on the software they've used. What have they gotten in return? They've got slander, libel, words you never heard in the Bible. They responded to the complaints about Safari by setting up a new repository for Webkit. They've released a whole boatload of packages that were completely developed inside NeXT and Apple, including netinfo, HFS, launchd,
Can you download the source to Microsoft's HTML control (based on NCSA Mosaic), NTFS, or even Microsoft's BSD-based software like the Interix and TCP userland? No. Yet even with the tiny crack of support for Open Source that Interix and the carefully selected open source honeypots^Wreleases at Sourceforge represent Microsoft's gotten tremendous traction. Even the FSF backed down on killing GCC on Interix (even as they knocked back Dave Korn and U/WIN!).
It seems like the best way to get good press from the open source community is to be a bastard. It doesn't pay to be the nice guy.
Sun got the same message, didn't they? Open up Java a bit too much, Microsoft turns it into a honeypot. Pull back, you get slapped, and the FOSS community jumps aboard the dotnet bandwagon with Mono. They're probably saying "thank god we never Open Sourced NeWS...".
Case-sensitive filesystems are stupid. If you tell someone "go bring up the 'college' document" do you think they hear any case-sensitivity? If they see "college", "College", or "COLLEGE", don't you think they'll assume they found the right one? How would he know you said "college" and not "COLLEGE" - they sound the same to me!
Case-preserving non-case-sensitive filesystems are absolutely the right way to go for any consumer-targeted OS. Only (some) geeks would say otherwise. If you're relying on case to differentiate files, may I suggest that you choose better names?!
Constitutionally Correct
Leopard will be Universal (i.e., won't have two trees like 10.4.x does)
:)
Interesting, and Informative. That's worth +2 at least.
it seems to be more of a full system (with GNOME and WindowMaker)
Now if they were going ahead and building on GNUstep, I'd be interested. But sticking Windowmaker on top of yet another Gnome clone isn't very interesting.
what Mac apps would you want to run on Linux that you can't find a windows version to run on Wine or some windows emulator.
Running apps under Wine is about as insecure as running apps on Windows, because it's the Win32 subsystem and all the COM-derived code like ActiveX that are the source of Windows security woes. All you're doing by running Win32 on a Linux kernel instead of an NT kernel is bringing all the things that make Windows suck so badly over to the UNIX world, in a less convenient and lower performance package. You're better off just using VMware or dual-booting.
OS X apps, on the other hand, are UNIX native at the lowest level, and even ports of Windows apps to OS X generally benefit from that. On Linux or a non-Mach-infected BSD they'd likely run *faster* than on OSX.
I say they'll muddle on for another year.
blah.
Well, basically, it's like this: the people who know enough to work on it are, for the most part just using Mac OS X, and most of the Linux crowd can't really tell the difference between GNUStep and Gnome (ie, they actually believe Gnome is good enough).
God, you're such a cynical bastard. Alas, I agree with you 100%.
C is syntactically and semantically poorly suited to object oriented programming, and it doesn't matter how much radioactive spider venom you pump into it it's never going to start swinging from building to building like some comic-book super hero. Even wrapping a powered exoskeleton around it is iffy, but at least there's some realistic hope that you can implement something that'll make the transition from the funny pages to the front page that way.
C++, of course, is the spider-man standin, and Objective C is our metaphorical iron man. I'd rather program in Javascript than either of them... at least Javascript is built around the object model from the start in a way that even Java (the "new spider man") hasn't managed.
What we really need is to for someone rip the dregs of Xerox PARC's ugly-sister user interface out of Smalltalk or Squeak and just use Objective C as a bridge to an open-source programming language that doesn't suck, but alas the Smalltalk crowd's got this horrible baby-duck fixation on the worst parts of the platform...
Even for an AC, this gets my vote for dumbest post of the week. Both OT and inane, and extra credit for throwing in Craigslist. Well done!
Their dreams being that they would create an open source project, and everybody else would do the work for them.
They have contributed patches and fixes to GCC, they have released their WebKit, based on KHTML, as open source which was recently adopted by Nokia for their web browser, and they host the source for several of their own in house projects, such as Bonjour, Darwin (though out of date), Quicktime/Darwin Streaming Server, and others.
So they have full time developers working on open source projects (GCC), release their own source, and contribute to existing projects. Are you just pissed that Apple hasn't adopted Linux and moved it forward?
GPL Deconstructed
How could this be made easier?
http://developer.apple.com/bugreporter/index.html
Sorry to nitpick, and I realize that this isn't contributing much to the discussion, but for the record it's "Mac OS X" written just like that, with the capital letters and spaces as written. It's usually abbreviated to "OS X" but never MAC or OSX.
I believe part of the confusion over this is that it used to say "MacOS 8" on the startup screen, but since then Apple has clarified the way it is to be written.
The good news is that darwinports, while related to OpenDarwin, is a separate project, and James Berry indicated on the mailing list that porting efforts could continue on.
http://tinyurl.com/4ny52
Same here, and that's after using iTunes since version 2 in OS 9. Though I just started using it about three weeks ago, so maybe past versions have sucked hard? I suggest the GP should give it another shot.
Thanks for taking the time to post that, I really appreciate it. I'm currently working through Cohen's Mac XCode 2 book, and have yet to write any code but was leaning towards carbon - your post does help me to rethink that! I hadn't realised about GnuStep, either, which means I'm not wasting effort.
Yes, correct, I was a bit too extreme sayng that Apple does not release any piece of code. However, as it wwas very nicely explained above in other posts, If you had to recompile the OSX yourself with the bits of source that actually are available, you would not go further than the login screen.
Yes... I saw that post right after I commented.
I'll give that a try!
Thanks!
xar!
The question here is: What is Mac OS X?
You can compile your own PPC Darwin. You can boot that and log into it. There's no Intel version of Darwin (yet - maybe Apple will never release it, maybe they will do it after 10.5), but again, what is Mac OS X? The relevant userland parts are from BSD, anyway. You can combine FreeBSD or OpenBSD with some of Apple's code and KDE to get something that shares a lot of the functionality with Mac OS X.
There are some things you can't get: The most obvious are Cocoa, Carbon, the Finder and the Core compenents (like Core Graphics). There's a simple reason for that: Apple relies on these parts for its competitive advantage. If you want Apple to exist, you must accept that they will keep this proprietary. If they were to open these things, they'd lose their competitive advantage and would be an iPod-only company within few years.
Most Mac users buy Macs for two reasons:
Opening the Finder and stuff like Core Graphics would mean that Linux would have a superior file manager and UI within months. Opening Carbon and Cocoa would mean that Mac apps would run under Linux within months.
Blaming Apple for not opening these parts is hardly fair. They need these parts to remain proprietary if they want to keep existing as a computer manufacturer and OS developer.
"Bundle" does not mean "make available to download", it means "put on the install CD".
How we know is more important than what we know.
Linus has declared that he does not believe the GPL covers binary modules and he will not sue anyone distributing them. But there is no special exception in the Linux GPL, and the laywers who wrote the GPL disagree with Linus, and any of the other kernel authors could sue ATI and NVIDIA at any time.
meh- karma to burn. If you're not sick of GNAA shenanigans, you haven't used the internet long enough.
Temporary files!? Why not just mplayer -shuffle -playlist ?
Temporary files!? Why not just mplayer -shuffle -playlist <(ls *.mp3)? Or even ls *.mp3 | mplayer -shuffle -playlist /dev/stdin (if mplayer doesn't support the - convention)?
Whoops, typo.
On OS X some of the best performing multi platform software is written using QT. They have great reviews from ordinary mac users too. I see Mono guys shipped .NET for OS X but I haven't seen a single software requires it.
.NET
.NET for OS X or Linux? I mean the usage. I see there is .NET for OS X true, downloadable, perfectly packaged. Where is the software using it?
0 9/17/226241
QT programs are complete commercial success. A good example is Skype. If you are against QT because they require money from closed source developers who sell commercial software please say it openly.
If they will be crushed, I won't believe until one of these guys who actually codes real life, billion dollar stuff switch to
http://www.trolltech.com/customers
As we speak about Apple here, these guys gave up PowerPC because they see future as portable. Guess who already has a working product which works on every portable device out there? Trolltech.
I see c# people speaks about how great thing it will be and how excellent it performs but in reality I got Opera Mini on my phone, a free application which uses 107kb of space. It is the J2ME 2.0 version with all bells and whistles. I got Skype which gets good reviews from the most OS X community.
Where is the
Perhaps people got their own reservations because of stories like this?
http://developers.slashdot.org/article.pl?sid=05/
What, exactly, is acceptable?
Qt is straining so much at C++'s limitations that the authors felt the need to extend C++ to implement it. OpenStep takes this one step further and grafts an entire second-generation* object oriented system onto C to try to make C a viable platform for its development. They're not exactly comparable either as OpenStep also covers a wide range of functions outside of the widget toolset.
Both, quite honestly, suck in many ways, but they get the job done, and most of the religious wars to do with both have to do with what the developers are used to.
Meanwhile, pretty much the entire GNU/Linux community does, contrary to your comment, accept that X11 is an abomination. It's just the alternatives so far come up short. Most Windowing systems don't even run over a network unless you're willing to take your entire desktop with you. So we plod on with X11 as the least awful of the terrible windowing systems, waiting for the day someone will come out with something actually good. Word to the wise: that ugly, PDF-inspired (WTF?), mess that is Quartz is NOT even remotely a decent design.
* First generation == SmallTalk and other "complete" OO designs. Second generation == systems grafted onto existing procedural unmanaged languages in an attempt to fix them. Thid generation == Java, C#, OO implementations that are complete and don't compromise in an attempt to be compatible with legacy code, that nonetheless use familiar syntax.
Quartz is NOT even remotely a decent design.
Show me something better. That's Ok, I'll wait.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
when I mention Linux they respond with "But OS X is Linux" [...] but they say "same thing" it really pisses me off(especially as a Linux user).
:)
How is it that something Apple isn't doing (claiming OS X is Linux) is pissing you off with Apple? There's lots of people who confuse BSD and Linux, but that doesn't make me pissed off with Linux Torvalds.
I cannt imagine what the BSD developers/users feel.
Clearly, because this BSD developer/user isn't pissed off with Apple at all. Some Linux users, though, really need a free whack with a clue stick.
I spent 20 years supporting a team of PhD engineers and programmers, and most of them were [non-geeks]. Even a lot of the ones who thought they were, who pulled stupid stunts that ended up with me having to disinfect their computers while they railed at me about how they were special and should be allowed to ignore the security policies that would have kept them from getting infected.
IBM and HP made System V into incompatible proprietary variants, not BSD.
GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
I have often wondered why apple's open offerings didn't get traction on other platforms and then I read the APSL - Apple Public Source License.
It has nice protections from Patent infringement law suits -- but only for Apple. I haven't read the license word for word, but from a quick glance Apple is keeping most of the protections for themselves, not any other contributors.
There: Something at a specific location.
Their: Owned by someone.
Please make sure your english compiles.
In fact it's the FSF that convinced the CSRG to remove the attribution clause from the BSD license.
Ain't that ironic?
I wish Apple would have partnered with GNU/Linux.
That wouldn't have happened. If the choice had been Linux, BeOS, cutting a deal with AT&T over the System V license, or finishing Copland... Linux wouldn't have made the short list. The license is a deal breaker for anything but an embedded system.
the radioactive spider venom "just works".
Only in comic books, Objective C works in real life.
I'd say that their web browser is a strategically more important component to Apple and its userbase than some unix userland utils.
Mach is a LONG way from a complete operating system, and there's most of a BSD kernel there in kernel space alongside Mach: the "BSD single server" for operating system services.
If they made the kernel GPL, they'd have to make many of their kernel modules - and most of the really critical ones - GPL as well, which would have brought Quartz and everything built on top of it into the GPL. That was just NOT going to happen.
I don't find GPL software stays in the commons any better than BSD software. I have just as much trouble finding old GPL software as old BSD software.
What keeps open source software in the commons is active development by the commons. Lose that, and the software rots... falls behind current APIs... and eventually vanishes.
The only difference with BSD software is that with BSD software there can be proprietary forks. If the open source model works (and it does, for an awful lot of stuff) the proprietary fork has to play catch-up. Where the open-source model doesn't work, the version in the commons is going to tend to be a poor cousin to the commercial version anyway... and end up falling out of active development.
Like happened here, and would have happened eventually no matter what Apple did, because the real problem was the lack of interest from the open source community for a slower and less capable cousin of FreeBSD, OpenBSD, NetBSD, Red Hat, Suse, Debian, Gentoo, Solaris, or Lites...
It's like Microsoft opensourcing the NT kernel and keeping Win32, DirectX, COM, .NET*, etc closed.
That would be tremendously useful. The NT kernel is a really interesting system that has all kinds of potential that Microsoft's thrown away for fear they'll distract attention from their "crown jewels"... the Win32 subsystem. An open source NT kernel would allow for a real open UNIX-compatible subsystem, rather than having to choose between staying in the awful Win32 environment or depend on Microsoft's goodwill with Interix. It would allow for GOOD ports of filesystems like UFS or XFS to NT, and better interoperability with open systems. It would also let people do to Microsoft what OSx86 did to Apple, and wiping out all their DRM and license validation in one fell swoop... which is the real reason it won't happen.
Yes, it's nice that Apple "just works".
It's nicer that Apple actually has commercial applications available.
If red Hat was willing to give up the input from the open source community to the extent that Apple has, they pretty much *could* fork off and go it alone. There have been forks in GPLed projects in the past, and there will be in the future.
It's not the GPL that binds Linux together, it's the open source community. It's not the GPL that makes Open Source work, it's Open Source that makes the GPL work.
So why was nobody interested in Open Darwin? Because it's Apple's product.
Hardly.
I'm not interested in Open darwin because it's got no point.
Darwin is a lower performance kernel than FreeBSD, supports fewer drivers than FreeBSD, and can't do much of anything that FreeBSD can't do better other than provide a modicum of compatibility with OS X for server applications... and OS X is a poor server platform.
People don't run operating systems for the sake of running operating systems, they run them for the sake of doing something with them. What can yo do with Open Darwin you can't do better elsewhere?
What, is it only if you're taking on Microsoft that you guys give a damn about a project?
No, it's only if I've got a reason to use a project that I give a damn about the project.
As a platform for understanding and working with OS X, developing tools to work with OS X, darwin's got a point.
As an operating system, it's a sluggish poor relative to FreeBSD, and it doesn't do anything I can imagine wanting to do that FreeBSD doesn't do better.
The only valuable thing they were doing was providing a place for "the insignificant little projects" you're complaining about. The core itself is pretty much worthless beyond that.
Your other two bullet points are really all variants of this:
Over time, more APIs will have to go Open Source as they become commodities that people aren't going to pay for.
People don't pay for APIs, they don't even pay for operating systems, they pay for having their problems solved. Apple provides a way to run commercial software that doesn't involve the toxic swamp of Windows, and that's why people are still buying Macs.
And the same thing with open source. You can't *push* something into the open source community. Software succeeds as open source if it provides something that people need and are willing to pay for in time spent hacking on it. Darwin as a collection of open source projects that build on the source code Apple's releasing and do things with it that people actually need was a success. Darwin as an open source OS based on the core of OS X had no market. Why? Because nobody needs another free UNIX variant that provides no new and interesting capabilities. Darwin without the rest of OS X is just FreeBSD writ slower.
Open Source code from Apple is not the improvement playground of Linux, but a utilitarian structure that is only changed when necessary, and only by Apple.
The same is true of Red Hat, really, and yet they're the most popular Linux version in the real world.