Even though there are a lot of happy people using Apple’s iPhone very happily, there’s also a group of people who are not so happy, most likely because of Apple’s rather strict policies regarding applications and developers. While most of these people would just jailbreak the thing, some take it a step further – by installing another operating system. Yes, Linux now runs on the iPhone (1st gen/2nd gen, and the 1st gen iPod Touch).
The port is obviously not complete, since a lot of the hardware on the iPhone has not yet been reverse engineered to understand how they work. Consequently, there’s a whole list of things currently not working: write support for the NAND, wireless networking, the touchscreen, sound, the accelerometer, and baseband support. What it does have is a framebuffer driver, a serial driver, a serial-over-USB-driver, and things like interrupts, MMU, clock, and so on. It boots using openiboot, a custom boot loader, which allows you to choose between booting into the default iPhone operating system, or Linux. The Linux 2.6 installation runs Busybox, and uses an USB keyboard (attached to the computer the iPhone is connected to) for input.
iPhone Linux Demonstration Video from planetbeing on Vimeo.
Poject lead ‘planetbeing’ explains that “the current userland we’re using, in the interest of expedience, is a Busybox installation created with buildroot, but glibc works fine as well, and we’re going to build a more permanent userland solution.” The team is asking for help with this effort, and especially Android hackers are welcomed (hint, hint). To get started, there are instructions, and download mirrors can be found in the blog announcement and its comments.
I know there is still a long way to go, but I home this opens the way to porting Android. I might hang on the my iPod Touch a bit longer and see where this goes.
Google has only allowed their SDK to be released for developers to create apps, not the OS. Though I agree I would prefer the option of Google’s Android OS compared to running a Linux distro on it.
Edited 2008-11-30 01:04 UTC
If it’s open source, then you can do whatever you want.
Except break the licence.
The entire Android sourcecode was released some weeks ago
google.com –> “android” “source” –> http://source.android.com/
I’m glad to see Linux getting ported to yet another device
Reminds me of putting Linux onto the Ipaq. Why no touch screen or X Windows though, isn’t it an ARM?
Just give it time. If there’s one thing that the Linux community has demonstrated without doubt is that they’re persistent. It’ll get there.
I’m definitely watching this project and hope it is a huge success!
The persistence is based upon billions of dollars invested by corporations who pay for the bulk of the heavy lifting.
Don’t get overly titillated about this as major corporations aren’t going to touch this one with a 100 m pole.
Yeah, because corporate Linux backers are the ones behind ports to the iPods and other hardware with no enterprise use.
Grow up.
The persistence is based upon billions of dollars invested by corporations who pay for the bulk of the heavy lifting.
I just HAVE to butt in, sorry for that, but…that’s a seriously naive and ill-informed statement right there. Do you believe Linux development have been funded by corporations all the way from the beginning? No, it hasn’t. All the “heavy lifting” has been done by people who freely donated their time, especially so in the beginning, and all out of enthusiasm. I hope you know the meaning of that word.
Don’t get overly titillated about this as major corporations aren’t going to touch this one with a 100 m pole.
I seriously doubt that f.ex. Damn Small Linux or Puppy Linux are that popular in major corporations either and they are both doing fine. Not having financial support from a major corporation does not mean people will stop working on something. This work, Linux on iPhone, is done by people who love Linux and just like to tinker around. I don’t think they’ll stop doing that anytime soon.
What processor the thing is running has very little to do with the display working right and having X running. You have to have drivers for that hardware working… itll happen, in due time.
As if “Apple’s rather strict policies regarding applications and developers” have ANY THING to do with porting Linux to the device. Just tell me how you arrived at that conclusion.
Ah, I detect sarcasm
.
There’s any number of things I’d love to say on this topic. But, knowing everyone here, I’m going to get flamed if I dare to blaspheme… oh, what the heck I’ll say it anyway.
Why? While people in the Linux community are porting it from device to device to device… and yet to more devices, major issues remain unsolved. No stable API. No stable ABI. No standardization across distributions–and please, don’t give me that “LSB will solve everything” baloney. No standard way of installing packages–we have how many different package managers and repositories now, with their own little quirks and dependencies? I guess that’s the great thing about Linux, everyone can just blame everyone else since “that’s not what we’re working with.” The kernel folks can blame the driver manufacturers while ignoring the real problem–if you had a stable API and ABI whether the driver is proprietary or open source wouldn’t matter worth a damn. The driver would work. Come on guys, Solaris has gotten this right for years now. I can take a driver for Solaris 10 and install it, in binary form without access to the source code, on any of the builds of Opensolaris (what will become Solaris 11) and guess what, the driver works. The desktop developers can point the finger at any number of stacks under them in the chain, saying it’s not their problem, meanwhile everyone underneath them is pointing right back, saying they have to fix it on their ends since it works fine and they’re just not using the correct functions. And then meanwhile we have people porting Linux to their freaking toasters, and who knows what else.
I guess my point is, why is it being ported to the iPhone when these major issues remain completely unacknowledged? You guys want Linux to be the operating system of choice, and I respect that. I wouldn’t necessarily object to this happening either. But please, stop worrying about every little device and get some of the baseline together. Then get Linux ported on to every bloody device on the market, if you want. You’ll be in a much better position to do it, that’s for sure, and you might actually achieve your goal of everything just working out of the box.
I don’t expect anyone in the Linux community to listen… but it felt really, really good to say that!
So, essentially, you are telling the folks who port Linux to the iPhone: Stop doing that! Fix my issues first!
But, ya know, those folks don’t give a shit about your issues. Because they are not issues to them. The issue they currently resolve is to bring Linux to the iPhone. That’s what they are interested in, that’s what they are doing.
Why blame them? Because they just don’t have any feelings about the driver API? Because that part just already works for them?
And guess what: It works for me, too. I don’t need any proprietary driver to work with any awesome stable driver API. All I need is a stable userspace API and good quality open source drivers, both already provided.
If it doesn’t work for you: Too bad. Fix it or leave it. There is nobody paid for painfully maintaining a stable driver API in Linux, so nobody does it. People can live with that, and surprisingly well so. It is wrong to claim that Linux has so many issues. Look at all these thousands of use-cases and systems it runs on already.
Ah, the old “works for me, so screw you” response the Linux community is so fond of. You do realize that it is those like you who make Linux users look snobbish and rude, right?
I am not saying “fix my issues first.” In fact, most of the open source drivers work for me, too, and I’m able to get around what does not. But the approach linux is taking is dead-ended and is leading to horrible fragmentation. Further, it is not developer friendly. note, I said developer-friendly, not hacker-friendly. Everything is hacker-friendly, regardless of what it is, it can be hacked, and bravo for them. But, if you want application developers and driver developers to pick up your platform, you need a consistent base. Until it can be guaranteed that an application developed for Linux will run on Linux, and not break with the next minor point release of glibc or the kernel, then you’re on your way to having a platform, and not a mishmash of parts.
I realize that the current way is fine for open source. But guess how many people actually care about that? Look at Linux’s market share and that should tell you. But having a stable base is fine for everybody, not just open source, and that, I would think, would be the best of all worlds.
No. You are saying that everyone should want what you want. Read your own posts in this thread. Clearly, it is you who is being snobbish and rude in this instance.
Edited 2008-11-30 00:27 UTC
No… The problem here is plain and clear the Linux user.
Here we have YET ANOTHER platform where Linux runs half backed without any real value whatsoever…
Where is the “Scratch one’s Itch”?
Eric Raymond says:
Every good work of software starts by scratching a developer’s personal itch.
The personal itch reference here is to solve a problem, solve an issue that they have.
http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar…
So tell me what the freaken problem here is? Answer there is none other than, “hey I have time to kill with no real reason on why I would do that…”
Apache scratched a real itch a real problem. Perl, PHP, MySQL all scratched itches, real problems that people were having.
I disagree. You are saying that if these guys didn’t get a Linux kernel running on iPhone hardware, then the Linux kernel developers would implement a policy of maintaining a stable internal driver abi, and there would be one package manager to rule them all instead of yum, apt, and yast. That is a completely nonsensical assertion.
Actually, Linux is not running “half-baked” here. It is just barely running at all because that is the state of this very young project.
I use OpenWRT every day in my work to extend the capabilities of WRT54GL routers to handle the needs that my clients have. Standard WRT54GLs do not have the capabilities required. The OpenWRT firmware is rock solid stable and far *better* than the firmware that the routers come with. But there was a time that the OpenWRT project was at this stage, just getting the kernel to boot, and I’m sure someone must have made a post similar to this one in his blog or on the OpenWRT site. And now the results of that project are invaluable to me and many others.
See the above paragraph.
Eric Raymond is (in my opinion, of course) a self-important windbag whose quotes generally aren’t worth the bandwidth used to propagate them.
The “freaken” problem is that you and Darknexus are presuming to dictate what other people should and shouldn’t do for you with their time, talents, and efforts. And while we’re on the subject… what have you done for me lately?
Edited 2008-12-01 23:00 UTC
Brava, Sir, Brava! To top it off now you have Janet Jackson stuck in my head. Brilliant!
A little undue sarcasm because I agree with you.
WTF are you talking about? The author of the code most certainly had an itch, just because its not your itch doesn’t mean his is not valid. Tell me what possible value did Pong have? Someone decided it was their itch and it so happened that others had that itch as well now we have a multi-billion dollar industry out of that itch, that doesn’t negate the that fact that the game was developed on what was then a super computer that was supposed to be used for scientific and educational purposes. Who the hell are you to judge someone on how they spend their time. The saying goes scratch YOUR OWN itch, not scratch the other guys itch. To some Linux not running on everything possible under the sun is a problem, an itch if you will.
The nerve of some people, the sense of entitlement displayed nowadays astounds me to no end. You want something as “useful” as Apache code it you damn self. in the meantime I’m going to work on my game which may hold absolutely no interest to anyone but myself. Why? My back is itchy and I’m gonna scratch that bitch to till it bleeds. I could give two shits about Apache.
Edited 2008-12-01 22:53 UTC
I disagree. I like choice. Besides, there really is only a few package managers: rpm, dpkg, source, and maybe a couple of others. Applications like Apt, Synaptic, Yum, YaST, URPMI, and others only manipulate the package managers and do some dependency resolution. The real package managers are rpm, dpkg…
But, back to the choice thing. I think homogenation is bad for Linux. I use PCLinuxOS, but if it goes a different direction than I want it to, I’ll move somewhere else that gives me what I want. To homogenate distros so that they’re standardized is just saying that I’ll take what they give me and I’ll like it. No, I won’t. I like certain package management applications and I like having a nice configuration application like the Mandriva Control Center or YaST. If I had to use Ubuntu for the rest of my life, I’d buy a Mac. It’s that simple. I’m all for FOSS, but I’m not a die hard and will use something else if that works better for me. Right now, Linux is everything that I need. But, this kind of crap could change that…
I am not the Linux community. In fact I believe there is no single person in the world who may speak for “the Linux community”.
You have to realize that what you are talking about is exactly not the thing most of the Linux developers are interested in. The are absolutely _not_ interested in the people who “don’t care”. The people who use it, the enthusiasts, these are people who _do_ care and these are the important ones as well – they bring the whole thing forward, not your Grandma.
This is not a company who needs mass revenues to survive. This is not a religion who needs to educate everybody else. This is _exactly_ a bunch of “works for me (or I will eventually fix it)” people, and that’s nothing rude.
The open-source ecosystem is in very good health ever since. Old school market rules don’t apply to it. This may not be how you like things to work, but it is just as it is and it doesn’t have anything to do with the people involved or their attitude.
It just doesn’t make so much sense for people to work on issues they are not interested in themselves without payment. That’s why some things get done and others don’t. The amazing thing now is how many things actually get done.
I also don’t believe in this “best of both worlds” thing. I don’t think you can get that, there are too many contrary paradigms. This will take a lot of time — but the beauty about this thing is, that it will have all the time it takes. It won’t suffer an early death, as companies and proprietary systems do.
Although I do believe what you say is true about the linux kernel hackers and their intentions and goals “works for me”, unfortunately, based on the entire business world and IT professionals, these linux hackers have chosen the wrong path. Pretty much every review I have read is about “linux” on the desktop and consistency as darknexus has stated. Based on the type of behaviour of the kernel hackers, I can assure everyone these reviewers and IT professionals wanting this type of consistency have a voided hope. Where do I come? Well, “it almost always works for me” and personally I have no problem using Linux and although in general I am a third party software developer, unfortunately I am not as advanced as these kernel hackers are — in order for me to do something about consistency. If I could help, I would help. This is where people like me and others fall apart. I want to help but I can’t, others can help but they won’t.
Back on topic, that linux on the iPhone is good as a personal exercise and train yourself, get more experience and learn about linux but I don’t see it how it would be useful and why (besided learning) anyone would want to install it. As a training excercise it is good but that’s about it.
Edited 2008-11-30 01:48 UTC
If you want to put it like that, yes. The reason why open source software is used by so many now is that quite a few people can say “Works for me, so screw you”, and developers such as these just churn away until more and more people are able to say that.
That’s the only way it can work.
Yes, you are.
On the contrary. Open source code has brought things together. Old Unix could never work the way Linux systems do today.
We have a great deal of binary compatibility within Linux platforms today which many just don’t seem to realise. Maybe no one cares when things work? You can grab many old Loki games as well as Motif applications from years ago and have them just run. The kernel already is binary compatible with applications. How do you think we went from 2.4 to 2.6? I can’t think of very many occasions, if any, where a new version of glibc has caused any problems. As demand gets greater, compatibility will be extended to where it is needed.
I agree with your sentiments, but that process takes time.
Edited 2008-12-01 23:14 UTC
Exactly, just like with BeOS and Haiku which share a clean and stable DDI.
I’ve been ranting about it for years…
And the “this will help hw vendors evade the GPL” reason is totally ridiculous.
It just impairs people developing external drivers like OSS4.
Well, while Linux does not have a stable *internal* API for drivers, it does have something that both Haiku and BeOS lack: relevance.
Troll grade: D-
1) Seriously that is difficult to read without it being put into paragraphs
2) This is a flame…a fun flame but nonetheless a flame
I suspect you are being deliberately naive about what the “Linux Community is” your on OSNEWS for gods sake they even hate each other, and there are articles upon articles about this, but generally the answer is fulfill their needs not yours, or MONEY, and this has a filer down effect to the rest of us.
Stable API…seriously I don’t care I’ve never met a user who does, but if your talking about Linux standards you understand the PROs and CONs, pick a side, I’m personally quite happy with evolution picking standards especially when evolution brings me a better product.
Personally I would love a device that is supported under linux as an MP3 player. I’ve been hungry for one for sometime I was gearing up for an Openmoko but untill the GPS is fixed I’ll pass. Something running rockbox would be ideal, but much like this project it will lag behind the proper release.
Now I can say why I want Linux on an Portable Media Player/Mobile Phone if you don’t then you should look seriously at one of the major advantages of the platform, That is evolution if you want the latest greatest version you do so at the click of a button. You are not tied to the hardware. Look at Vista users on there 2 year old RTM OS and have to buy a machine to get it…only a real idiot would get a retail copy of 64bit Ultimate, or those who continually upgrade their ipods. Rather than buying an item for the properties in the hardware.
The bottom line is my Linux experience continues to improve dramatically while Vista users have to wait years, and then experience massive regressions.
What’s the connection – why are those things mutually-exclusive? It’s not as if there’s some sort of central Linux authority that has said “you developers must work on an iPhone port of Linux, instead of standardizing package management or creating a stable driver API/ABI.”
Developers are also not interchangeable. You can’t just assume that a developer working on Linux on the iPhone would have the necessary abilities/specialty to fix the problems that you described.
@darknexus: No, I won’t flame you, and many of your points are valid, but you’re missing one thing. Linux is where it is BEACUSE of its unfriendliness to proprietary, binary-only drivers. The most compatibility is on the source code level. If you want the whole thing (kernel and programs) to run on another architecture, the code just needs to be recompiled, and sometimes tweaked (ported) as well. But it’s doable.
And think for how many architectures the proprietary drivers would be supported? Would they support PPC, CELL, z/Architecture and ARM processors? I understand YOU may not need it nor use it, but there are folks that do. And they may become pretty angry if their hardware of choice doesn’t work because of lack of availability of proprietary drivers.
You’re 100% right. I used Solaris 8, 9 and 10 on original Sun Blades and Ultras and there’s no problem with this. But such a driver works only on architectures supported by the OS, unfortunately (which means “less than Linux”). I believe that Sun and/or OpenSolaris have access to the driver’s source code, that’s why it can be recompiled by them. However this still is a problem with proprietary drivers from commercial vendors.
I think you’re slightly missing my point here. I’m aware of the relative ease of porting Linux to other architectures due to it being open source, and this is great for what it is. But by no stretch can we say being unfriendly to proprietary drivers is a good thing. Proprietary drivers would work on the architectures the manufacturer supports, that’s true and no one is going to argue differently.
And how is that different from the current situation? There need not be a proprietary driver for which there already is an open source equivalent that works well enough for people. But there is a lot of hardware that there is _no_ driver support for. Isn’t manufacturer support better than no support at all? Let’s make this clear: being unfriendly to proprietary drivers is not going to force most companies to open their specs and/or driver code. It’s simply going to drive them away from your platform. There are exceptions, of course, and some companies have started to open their specs, but I can count them on one hand and the only reason they’ve done this is they have a vested interest in getting themselves into the Linux realm due to netbooks and other such devices. You seem to be under the impression that Linux be moved entirely to proprietary drivers, which is not at all what I was trying to say. The quality of the open source drivers varies wildly from excellent to downright awful. If Linux were more driver-friendly to manufacturers, then you could have a choice, and isn’t that what we all want?
Further, providing stable interfaces would ease the process for manufacturers at writing drivers for various platforms. Provided they don’t need to access any machine instructions directly of the target host, they could simply recompile the driver for whatever architecture is needed, and let the kernel do the lower-level work.
No, it’s not. That is precisely my point, I can take a binary driver (the binary file, no source code or access to it) and drop it into a Solaris system of higher version (10 to 11, or what is now Opensolaris, comes to mind). These drivers still work, no recompiling, No source code access. Drop them into place, run a few devfsadm commands if necessary, and load the module. This sort of thing does not work in Linux, and it is precisely because of a stable API and ABI that it works well in Solaris. You’re absolutely right that the architecture of the driver and the target machine must be the same, however, a driver compiled for x86 will not work on SPARC, for example. Often though, the driver simply needs a quick recompile by the manufacturer to work fine and, in this case, Solaris’s limited architecture support becomes a boon rather than a bane. There is only three architectures that need be supported, not fifty.
To recap: I’m not saying we should abandon open source drivers for proprietary ones. There are drawbacks to both approaches, and proprietary drivers would only work on the architectures the manufacturers decide to support. But in many cases this would give us more choice than we currently have, and the option to perhaps use a better-quality driver is one we *should* have. Further, this would automatically make open source drivers work better, as they wouldn’t need to be recompiled or tweaked every time the kernel internals change. This would be great for open source drivers that, for one reason or another, aren’t included in the mainline kernel tree. The API doesn’t even have to change if they don’t want it to, just freeze it and stabilize it, then stick to it for a while only changing it when a bug is found. Solaris proves this can be done quite well.
Linux does have a stable userspace API and ABI. You can still run statically linked binaries from ages ago in current systems. What it does not have is a stable *internal* API/ABI.
Maintaining a stable internal API/ABI for drivers would be a huge task for kernel developers, and it would prevent the kernel from evolving freely.
In the end, whatever people think could be gained from making it easy to have out-of-tree drivers would be negated by the kernel becoming increasingly bloated and cumbersome, and kernel developers loosing interest in contributing.
Do you think the people doing these kind of tasks for the proprietary OSes do it for fun?
It’s funny how people complain that Windows is overburden with backward compatibility, but then want Linux to suffer from the same problems.
Yep, games and applications works really well after a couple of years.
Loki tried to bring real games to Linux.
They are all defunct today because of whatever technical reason.
Meanwhile the Windows, original versions plays quite well.
I’d make a more detailed comment, but just now I’m playing the 32 bit port of Rune that I bought from Loki Store back in 2001, on my Ubuntu 8.10 x86_64 quad core box running Intel’s latest X4500 video chipset on the latest Xorg 7.4, with Linux kernel 2.6.27, and Glibc 2.8, and I’ve become rather absorbed in my game.
Some of those lizard monsters are really tricky to kill. You’ve gotta hit ’em in just the right place on the neck with the short sword.
C’ya! 🙂
Edited 2008-11-30 18:22 UTC
The reason you don’t have a stable ABI in Linux is that the kernel folks do not want binary-biased driver integration. It isn’t about flexibility. They want you to have to compile your drivers with the distribution. It’s about ideology. They want the source code for all drivers to be available to everyone. Which means that driver support will always be second-rate in Linux. Imagine how much better things would be if you had a common ABI across distributions. You could use drivers from ANY distribution for your processor indiscriminately. But, no, the ABIs are constantly in flux, creating an environment which is brittle and unstable. Why? So Stallman can feel good about making sure the driver source code is available? So Torvalds can arbitrarily change the ABI when the whim suits him? C’mon, folks, let’s be honest here: this is RIDICULOUS. Somebody should let the grown-ups take over. Oh, and regarding LSB, it’s a joke. LSB has had nearly a decade to improve this situation, and it has only taken marginal steps. Everybody claims that “the next version of LSB will make everything better”. That’s like guerillas in the southern hemisphere claiming that the “next government” will stop poverty and end oppression (i.e. don’t hold your breath). Flame me all you want. You know that I’m right.
Agreed, 100%. I doubt it’s got anything to do with Stallman feeling good, though, as the kernel has binary firmware blobs in certain drivers. He’s already said the Linux kernel, as it stands now, does not hold with his ideas of free software. Personally, I think the man’s crazy, but that’s neither here nor there. So, we can rule Stallman out as a reason for this situation.
Its a shame because on this one point Linus and Stallman are in 100% agreement its only you that doesn’t. Seriously you should keep out of these discussions if you don’t have even the basic understanding of them.
Maybe some kernel developers do want to prevent binary drivers for ideological reasons, but for the vast majority that’s just a (welcome) side-effect.
Do you have any idea how hard it is to maintain a stable ABI? It isn’t only the case of avoiding changes to the interfaces (API), it is a full time job of making sure unrelated changes don’t change symbols around thus breaking it.
So, on one hand they would be restrained in how much they can change the kernel insides to accomodate new features, fix bugs, increase performance or just plain refactoring to make the whole thing better. On another hand, someone would have to police the entire thing to make sure nothing breaks. All this while keeping the whole environment interesting enough to attract contributions. One cannot have the cake and eat it too.
There’s an interesting paper from Ulrich Drepper about maintaining ABI stability for shared librares which should be read before making claims about ABI stability *inside* the kernel, which is harder.
http://people.redhat.com/drepper/dsohowto.pdf
Edited 2008-11-30 21:46 UTC
exactly… the sad thing is, nobody realizes that making kernel APIs stable (for at least a reasonable assured time-/version- span duration, say… 3 years? ) and encouraging HW manufacture to open their drivers or at least their specs, are different and in fact orthogonal problems…
… and that it’s not that easy, or reasonable, or “normal” for a HW maker, to even want to care for the users of an OS, that make less than 1% of the market that company caters to (not that they are really obliged… if i state a graphics card of mine specifically is a “DX10 gaming card for windows”, if someone manages to use it on maya under linux good for him – but support for unintended use is not due)
nor, in case they care, to entrust some random developer (totally uninvolved with the design of a HW product; and, for what they may assume, uninvolved with the development of similar devices, or drivers for them, or driver development in general) with the IP related to their product, just to let him (try to) write a driver for that device…
the most suitable ones to write device drivers are those working for the manufacturer (and for as overly exceptionally complex a device can be, as in the case of a modern vga, one will just not suffice)
this is a simple but often overlooked fact that does not deny the obvious, is that peer review of code will be important for system level integration, auditing and debugging (but only at that point)
or, you’d have a chance to use a device (e.g. a webcam) as soon as a driver for it (for as broken as it may be) is made available, without the need to draw in a newer kernel
being able to resolve problems caused by the new driver just uninstalling /updating it, instead of reverting the whole system to the previously used kernel, and to avoid the possible regressions or new vulnerabilities inherent to a less mature kernel, may be invaluable…
it’s ironical, even more so when they attempt to pass a political decison for a techinical one, pointing at GCC’s incompatiblity with itself across major versions as a major cause of ABI instability in the kernel, as if it were something impossible to obtain… (of course ABI incompatibility between GCC ports for different architectures is understandable and somewhat obvious – though i’ve often seen it mentioned to justify the above (!)) …
i used to have faith in the LSB but then i stumbled upon this: http://www.linuxfoundation.org/en/LSB:PM:Best_Effort_Dynamic_Linkin…
<< Right now, the LSB requires a different dynamic linker than the rest of the system. This linker is often not provided at all on non-LSB systems, and cannot be guaranteed to be available even on distros that can be LSB-compliant (if the LSB environment is not installed).>>
when a spec which is expected to be a common denominator and something all (compliant) linux distributions would be based from an “architectural” perspective, but (by words of its very proponents) turns out to be an optional side by side “environment”, loses a lot of credibility in my book…
The LSB is better than no LSB…. Say what you want about it, I’m glad that at least we have something.
It’s kinda like the United Nations: Kinda ineffective, but I wouldn’t want a world without it…
This port wasn’t done by the Linux community, it wasn’t done for the Linux community and it wasn’t done to advance the Linux community. It was done for kicks by some people who think this sort of thing is a fun way to kill time. Or are you the sort of person who looks down on people who have hobbies, since the world would be better off if they instead donated their time to charity.
There really is no “Linux community” and any argument based around the assumption that there is is inherently flawed.
My guess is because a couple of people wanted to know if it could be done and thought it sounded a bit like fun. Have you never done something for fun without worrying about the greater social ramifications? How (and why) should we stop people working on things they find fun and force them to work on things you find important?
Perhaps the people who did this (for fun remember) didn’t think that tackling those issues sounded like fun.
“We guys” want all kinds of things and have nothing that could even approach a cohesive agenda. Some want Linux to become a dominant force in the OS market and knock MS off the throne, while others want a small easy to hack OS that runs on their toaster. Neither is more ‘right’. Most of us simply want an OS the fulfill our personal or professional needs, and if that happens to be Linux then great. If not, then there are other OS’s out there.
At its heart Linux isn’t a company or organization or charity or social movement or anything else even resembling any of those. It’s a bunch of source code that people can pick up and use if it fits their need or simply ignored if it doesn’t fit their need. Linux doesn’t care one way of the other.
Question: If Sun’s strategy is so great compared to the Linux strategy… how come hardware support for Solaris/OpenSolaris is so poor compared to that of Linux? Solaris is about the same age as Linux (1992), and OpenSolaris has been out for a few years.
Despite any theoretical arguments you might have, the Linux strategy has actually *worked* better than the Solaris strategy that you are advocating.
Edited 2008-12-01 02:42 UTC
Actually, strategy doesn’t come into play here at all.
Why does Solaris have less drivers? Simple answer: because until recently, there was no vested interest on running Solaris on any hardware but that which Sun endorsed. It is only recently that interest has really arisen in Solaris on general-purpose machines other than SUN workstations and servers. Remember, Solaris was designed as a corporate os for servers and, to some extent, workstations. On the server side, supporting every bit of hardware is less important than supporting a few hardware combinations to their fullest extent. Desktop interest in Solaris has only come about within the past few years.
Regardless of that fact Linux supports more hardware NOW. All those who have complained about Linux so far seem to want Linux to support more hardware NOW. If that is the case they better not hold their breath that Solaris will get there anytime soon. Solaris is going to have issues with hardware for years to come, Stable ABI or not. It would be more likely that some enterprising hacker will write a driver for the hardware you are bitching about than for you to see a Solaris driver written by the manufacturer.
If the stable internal kernel abi of OpenSolaris was going to encourage manufacturers (or third parties) to target it, that would have happened by now. Three and a half years of OpenSolaris is plenty long to observe any such effect there, and we should have seen lots of *new and current* devices supported by OpenSolaris that Linux doesn’t support. We don’t see that.
Regarding “not new” devices, OpenSolaris might gradually accumulate additional drivers, which require less maintenance, in that category. But that’s the category where Linux especially shines, anyway, so the bar is set pretty high there.
Edited 2008-12-01 21:05 UTC
There are a lot of clueless people on this thread. Now I may be one of them, but last time I checked, MOST OS’s don’t offer a stable kernel API/ABI. You can’t run Windows 3.11 drivers on Vista and vice versa. Why would you expect that type of thing on Linux? Not to mention, the main reason for unstable behavior in Windows is drivers. Why would you want to bring that on Linux? Most companies do NOT spend the time on programming to make sure there stuff is rock solid. They program it until its good enough and then they ship it. I don’t want close source drivers in Linux and neither do the people writing it. But you are free to implement whatever you want in your kernel…
As for package management, again, you are chasing ghosts. There is one universal package for Linux. Its called source code. And it works on every system. Some distros make it easier, some dont.
Out of curiosity, does anyone in the know care to comment on FreeBSD’s policy on internal kernel api/abi?
Edited 2008-12-01 23:07 UTC
Linux uptake has absolutely nothing to do with a stable ABI. It’s just one of the many idiotic reasons that come up every time Linux’s success on the Desktop comes up. If that really was an issue Linux would not be doing so well in server marketshare. This argument belongs with the “X sucks”, “multipls distros suck”, and “multiple desktops suck” arguments. Multiple distros hasn’t caused any issues in ther server market either.
As for your stated goal of “everything working out of the box”, I don’t think anyone has any illusions of that ever being the case. It’s not even the case for Windows. Not only that it makes no difference whatsoever. People don’t install their own OS. Linux won’t gain desktop marketshare without preinstallations and people won’t switch until they have a compelling reason to. Some of use have already found that reason, others have not. If we can get a “killer app” that only works on Linux, or works much better on Linux than Windows then maybe Linux will start to take away a measureable percentage of Microsoft’s desktop marketshare.
Yes and while linux is “getting there” mac osx is bringing out massively parallelised APIs (new stuff in snow leopard) which people are using. Creating another generation of locked in users. Hell Microsoft even will have time to update all their APIs and lock in everyone again. Who knows maybe even a new Active Directory replacement?
There are a crapload of linux phones already. What we really need is a standard way of setting up 2.1, 4.1,5.1,7.1 headphones etc. In Linux. It’s 2008. Microsoft had this stuff down graphically TEN YEARS AGO if not longer. Hell Joystick configuration is still a rat’s nest on Linux. Microsoft nailed that stuff down back in 1995. s Sure the drivers weren’t always stable but for 99% of stuff start -> settings -> control panel -> game controllers.
The problem with Linux is it ISN’T getting there. For 8 years I’ve used Linux, for 8 years we’ve complained about this and 8 years later it’s still NOT WORKING. Fix the House before buying a yacht IMHO.
The iPhone is a shiny toy, in 2-3 years apple will make a 4G version with a new encrypted firmware that breaks all the work these guys have done (look at ipod classic (6G) still not running rockbox) We still won’t have 5.1 audio easily configured, and joysticks will still be f–ked on Linux.
Hell just today I had to rewrite part of a plugin for mupen64 just to get my usb nintendo 64 adapter to work properly because the calibration was f–ked and none of the calibration tools would work. Yes that’s both jscal and jscalibrator. Hell I found some kernel bugs in Force Feedback drivers a few months ago. There needs to be a hell of a lot more checking in the kernel. I can’t code for crap, I don’t even understand how the syntax for inheritance works properly in C. Most users of linux can’t even understand the tiny bit of code I know they’d be even more screwed.
So excuse me for not being happy about another new platform the OS runs on. I use Linux daily but man I wish developers who know howto and have the ability to write drivers would focus on core stuff first. This glamorous shit brings nothing to the platform worth having. Mark my words. In 6 months they won’t have all the animation crap the iphone has, they probably won’t have half the hardware working, and they’ll have blown 6 months of coding work that could have gone to something useful. Leave breaking the iphone to run whatever music/apps people want to the crackers/jailbreakers/pirates. They’ll unlock all the stuff people want. A free software OS on the iPhone isn’t what most people want anyway. They want a phone that works with their service provider and makes/receives calls properly.
Wish I could mod you up. Too bad the zealots buried your comment.
That’s something the Linux distros should be tackling. Not something that should be the responsibility of hobbyist kernel hackers.
Most people don’t have a hobby to primarily make other people happy, they do it to amuse themselves. Why should people do stuff they find boring and make you happy, rather than doing stuff they find fun and make themselves happy?
Of course they won’t. Nobody’s even expects that. Making a prediction like that is like predicting the tides.
Do you consider all time that people spend on things that have no immediate general social or economic value as “blown”? Why can’t people have fun? Why do people have to donate their free time work on things you consider useful?
I don’t know about you, but I spend a lot of my time doing things that are of no use to hardly anybody and will never bring me any financial gain, simple because I enjoy it. Does this make me a bad person who’s wasting his life?
You’re absolutely right: it is silly to get angry at someone just for doing what they feel like instead of what you think they should do. And stopping projects like this will not make problem X on Linux get fixed any faster.
I think the real reason for the parent poster’s annoyance is the feeling that the Linux desktop is being neglected (which really isn’t true, even if it feels like it sometimes), combined with a sense of powerlessness for a non-hacker to change that, combined with the fact that shiny stuff like iPhone ports get all the headlines in the geek news while useful stuff like driver development gets largely ignored (which btw is just the way things are, no changing that unfortunately).
What is important to remember in regards to Linux is how new the whole open-source model is for most players in the industry. It will take another long while before hardware companies get comfortable with the idea of releasing all their drivers as open-source. I do believe however that they will eventually come around.
I love how people flame a free operating system. STFU or GTFO. Seriously nobody is twisting your arm to use linux. It’s a matter of choice. Don’t like something about it? Fix it or move on. I’m tired of people saying linux needs this or that or linux should be like this or that. All the source is out there for a reason. Stop bitching and use it. Build your own distro hell fork the kernel if you want. It’s all about freedom and personally I think its cool that linux is running on the iphone. Is it pratical? Not really, but the fact that it can be done is enough for most people.
Let me explain, for the record, what flaming is. Better yet, I’ll give an example:
Linux sucks! I f*****g hate it, it sucks so bad… the FOSS community are stupid morons… etc. Clear enough?
I do not flame, actually. I state my opinion in a clear, concise way and without, I might add, telling people to STFU.
Do you seriously believe that your response is favorable to Linux’s image? Telling people to STFU? Thank you, you’ve made one of my points for me. Couldn’t have done it better myself.
And while no one is twisting my arm to use Linux, I can’t help but notice that every time anyone brings up a point against it, no matter how trivial, they get flamed by people like you and are often called “corporate slaves,” or “naive,” or other such terms. What do you call that, then? Is that not a form of arm twisting, specifically I’d call that the guilt trip.
There, THAT should make them feel guilty.
Put up or shut up. You can use Linux for free (legally). You can also buy a boxed distro with support. You can donate to developers without corporate backing.
Bitching is free (as in beer). Doing something useful? That’s work!
The last few threads of comments to this story are amazing in that they point out the contradictory nature of FOSS advocacy.
The FOSS advocates don’t want stuff like stable ABIs and APIs, prefer their software to be flexible and source-oriented, and gently admonish the beggaring would-be FOSS users to “STFU or GTFO”.
OK, so should the same attitude be returned to beggaring FOSS users who clamor for, say, Linux/X11 ports or open-sourcing of proprietary vendors’ software applications (and openly damn the vendors if they don’t, alot like the reaction of Mac users in similar straits)? Should we actually encourage the proprietary vendors to not port (or open-source, or even open-spec) their applications to Linux/X11 just to spite the users of Linux/X11 systems and let them stick with their flexibility?
I think such a request would do the Linux/X11 users a favor.
Edited 2008-11-30 20:16 UTC
Yeah becus it works so well with the stabel driver abi.
There is not one os with a stabel driver abi except windows that are well supported.
macosx wont find any drivers except for hardware that apple sells
beos hardly any drivers there
the list can go on and on about operating systems with a stable abi for drivers but still no drivers.
if linux had ben dependent on a stable abi and manufactures making the drivers it would be a really small os today with almost no hw support.
Just about every external device has Mac drivers. It’s true that OS X is optimized for Apple internal hardware (the hardware they use, it’s not built by them), but to say that drivers are only available for Apple devices is pure FUD.
Oh, and fixed the spelling in your subject line.
no really fud.
sure apple support many usb devices but so do almost all os.
there are loads of usb devices that dosent wok with osx.
mp3 players webcams wireless and tv capture.
Maybe his post was FUD, but there is truth there. Both Windows and OSX support less hardware than Linux out of the box. Stable ABI or no, the fact of the matter is that Linux supports out of the box more hardware, more platforms, than almost any other OS. Almost all of the drivers are open source, they get regurlarly updated, some of the drivers have been written by interested hackers by reverse engineering their hardware of interest and the kernel is rock solid.
Maybe I’m not getting hat you are saying because that looks like a success to me.
Surely the interesting issue is not whether they should be doing it – they do not care one way or the other what people think about this, they are doing it because they feel like it, as some people take up square dancing or birdwatching.
The question is, what effect will it have on Apple and iPhone if it matures into a viable alternative? And what will Apple’s reaction be? Will they try to lock the hardware to their own OS?
And secondarily, what will the Mac fraternity find to say in their defence if they do indeed try to lock the hardware? Which I confidently expect they will
(a) by a clause in the Eula, so using a condition on sale, which will provide that by opening the box you agree not to run any OS other than the one supplied with it
(b) by some sort of hardware code to prevent the installation or booting of an alternative OS.
One can hear it now: they had to do this to save their phone hardware business, and to continue to offer those of their customers who do not want to run a different OS the benefits of seamless integration….
Come back Yogi Berra, where are you now that we really need you!
I think your vision of the future is a little skewed.
The iPhone’s main strength is its software. Most people who buy it are happy to trade a little freedom for smooth operation. Even if an iPhone-Linux were developed to same usability and aesthetic standards of the real iPhone OS (which seems rather unlikely), I can’t imagine anyone outside of a select geek community would feel the need to mess around with it.
Apple would be wise to let this thing slide and pretend it doesn’t exist. Chances are low that it will be installed on even 1% of all iPhones — so let the geeks have their fun.
Edited 2008-12-01 13:39 UTC
I very much doubt that Apple will say or acknowledge this project. Just like they did with the osx86 project before Psystar came and fscked it up for everyone (not that it will stop anyone from making their own hackintoshes). Apple most likely doesn’t care what you do to iPhone as long as you payed for it. They weren’t doing all that much about Jailbroken devices in the beginning but had to do something because the community was basically using common security exploits that Apple had to plug in-order to gain enterprise aceptance and its still gets Jailbroken within days of official FW releases. Even Sony does a better job at locking down the PSP, and the only reason to get a PSP is to hack it (because there are no games for the device).
As long as someone is not selling iPhones with Linux installed (China I’m looking at you) I’m sure Apple will act like business as usual. They made the money off of you at the beginning if you bought an iPhone). Where they may get a little pushy is if someone ports the iPhone OS to another device. i’m ure they woudl rain down hell on that person if they tried to sell.
Apple knows that the enthusiast market will do its own thing, regardless of legality, and they know that the only people interested in going through the hassle of installing an alternative OS on a device with an already spectacular OS is not their target audience. They generally leave them alone. They didn’t bother the pearpc folks and they didn’t bother the osx86 folks so I don’t think this will be an issue.
They got Linux working on the original Xbox a long time ago. Still no hardware accelerated graphics. I doubt they will ever get it for the iPhone too. Don’t know who makes the graphics chip on the iPhone but it would require their cooperation. Just like the xbox is an Intel/NVidia box, the graphics chip is custom and the NVidia Linux driver won’t work.
The graphics ship in the iphone should be supported since I think there are other Linux phones which use thesame chipset that are already supported. The G1 I think uses the same chipset if not a more advanced one. Don’t quote me on that though. The PS3 and Xbox are totally different since the chipset is by nVidia and they aren’t planning to release non-binary drivers to any of their hardware anytime soon(or ever). Linux on the PS3 does support Cell though and that can be used as a GPU in its own right.
Absolutely ,but its not the same now as the world then. There are not at least 2 phone stacks that I can think of that are functional. Its not like Nvidia graphics haven’t now been somewhat reverse engineered. It is a pity its not an AMD chipset, Linux has an awful lot more developers now than it has then. Its not even the same sort of device. As for the PS3 stuff…Its complicated although it does really need a TV-Tuner built in which is only a matter of time.