If SCO is serious about Open Source, where does that leave older Unix code? I have the Lions Book, as many out there do. It makes no sense to me that if I could find a suitable compiler (which I can, at Dennis Ritchie's home page) that I have to pay $100 to get an Ancient Source Code license. Will the old stuff be made freely available? (What about current stuff, for that matter? How would an open-source SVR5/Unixware/whatever fit into SCO's business model?)/Brian
A few odds and ends about Microsoft's latest inroads into the PDA market...
I think it was in the Journal today that I read something about Microsoft finding out what Apple feels like. That's not really a fair comparison -- after all, Apple's situation was (largely through their own stubbornness) having the rug pulled out from under them. WinCE was never more than a reaction, and now that Palm owns the market I don't think there will be much for them to worry about.
I do think it might be in Palm's best interest to opensource PalmOS as a preemptive strike against MS attempting to force corporate hands (i.e. we can't guarantee that your Win2K machines will be able to HotSync properly after the next service pack release). I doubt their revenue stream is all that dependent on licensing anyway; they probably make the bulk of their money off of hardware.
At any rate, I think Microsoft is playing the same card they played to dilute the Network Computer revolution (a misguided idea to begin with, but bear with me). *Of course* users want extra functionality! *Of course* they want better integration!
You tell me: do you really need a PDA that can do everything a laptop can do? I don't know about anything else, but the day I can finally afford one what I'll be looking for is a smart notebook, not something I can use to replace my Mac and Linux box with on the road (that's what laptops are for).
I don't think I'd buy a WinCE machine (old or new) unless I could nuke WinCE and replace it anyway. That's just me, though.
I think a lot of people may not be aware that much of the Win16 API is actually standardized by ECMA (go on, go to ecma.ch and ask them for a current CD -- it's free).
Therein lies the key to your successfully functioning open source Windows community. Since I don't believe there is much risk of Windows forking too badly (the code is too big and probably too byzantine for anyone to really understand, I'm sure), Microsoft in principle hasn't much to worry about as far as losing control. An open source windows would most likely fork no more than once, and that into the hands of people who feel that they can beat Microsoft at their own game without destroying crosscompatibility.
More likely Microsoft winds up getting hit hard on the distribution level -- there will be a Win2xxx kernel, and there will be the Official Microsoft Distribution (you know it's official 'cos it's got the hologram), and then you'll have half a dozen other Windowses out there, binary-compatible but featuring vastly different mixes of high-level features. I could see someone slicing and dicing Mozilla into.ocxen, for example, and replacing the HTML-rendering components with Mozilla code.
I think it's probably reasonable to assume either way that even if Windows/Open Source forks in twelve different directions, anyone who does not maintain binary compatibility as Microsoft has defined in the Win32 developer's library will be dead on arrival.
The problem is that even after all is said and done, Microsoft still must be left with the ability to support itself. It's a reasonable thing to believe that they should be whipped, stripped and hung on high, but at the end of the day there has to be enough left to earn a profit. This is a truly grotesque situation, but to go beyond that would be excessive intrusion by the government.
A breakup can't be done particularly obviously, and structural remedies won't work because MS will crap on them. They can keep Office and their multimedia stuff, but the following packages should be opened up:
-Windows 2000 itself (to prevent unnecessary/predatory OS tying) -All network service packages (IIS, Exchange, and IE especially) (to prevent abuse of standards) -All APIs
I haven't read the whole debate, so I can't comment for certain on Tanenbaum's attitude, but... I don't see much sense of accusing ast of arrogance here. It may not be inaccurate (I don't know the man so I can't say), but it's irrelevant. Minix was created strictly as an academic system. If it does it for you, fine. If it doesn't, well, that's why it's under BSD license now. You can change it or use something else. But saying that ast got snowed for being arrogant and resistant to change misses a few points: -Linus was mostly learning low-level 386 assembler at that point, not trying to create a revolution. That didn't even begin to gather steam until the 2.0 kernel showed up. -Andrew Tanenbaum is a CS professor, and his intent with Minix was not to create a commercially viable OS but to create a functioning toy that does everything it has to to teach the student *but nothing more*. If I want to learn how Linux works, I have the Coriolis book, but that's only the very basics of the system. If I want to learn Minix, that's much easier -- everything is there, not that much more complicated than the Lions book I imagine. Ripping ast for his design is shortsighted as well -- there are those out there that might conveniently forget that Linux is hardly cutting-edge. It's a monolithic kernel, built with no principle in mind save "make it work and work well". Minix is a microkernel-based OS, designed to illustrate and teach as much as being used. Blew up in his face? Why, because a system designed for simplicity and elegance didn't take over the world? Slow down, buddy. The OS wars aren't a zero-sum game, and Minix was never even fighting in them in the first place./Brian
First off, many of you don't quite understand that Darwin *is* BSD. What NeXT/Rhapsody/MacOS X really is is a Mach-based BSD (similar in principle to Lites), with the Objective-C runtime developed by NeXT as part of the overall system. Darwin is that BSD system stripped to the basics: no GUI, no high-level APIs -- in other words, a vanilla BSD with some cool tricks up its sleeve.
The biggest problem with Darwin at this point is the almost-but-not-quite opensourceness of the APSL -- you can do what you want freely, but not behind closed doors because Apple wants a cut of your code. I don't think that's a big restriction in practice, but some find it philosophically limiting. Despite that, Apple's to be applauded for taking the free-ride approach to OS development; not only do they (in theory) make off like bandits, they also give their work back.
They're not free of reasons for criticism, however. They aren't making enough public. I want to see the wreckage of Copland put out there. I want to see Quicktime on Linux. Apple is dribbling things out there, but needs to show more of its cards and not worry about OS and service revenue (face it, that market will be shot to hell in two years anyway; why do you think there's a flat price for OS X Server to begin with?)
Darwin, IMHO, occupies a bit of a grey area -- if developers picked it up and ran with it, the BSD community would be that much stronger, and Apple would be able to generate some revenue by porting OS X to hell and back. But the APSL has to change, and Apple needs to put more on the line. Let's see Aqua go open-source. Let's see Apple cut loose on OS 7 through 9 -- the fact is that they won't be needing it anymore when OS X Client comes out, but a hell of a lot of non-G3 users will.
First off, using Oracle or Sybase sort of misses the point. The idea is that you *can* create something like this and go strictly open-source with it. Now I'm not all that familiar with the guts of Slashdot. But I know for a fact that the nice thing about opensource is that you can turn a pile of junk into something like a rather nice entry-level Beowulf or something of the sort, so cheap hardware for redundancy and high availability is no big deal. And I do suspect that a nice distributed-Slash project would be very helpful; abstracting the database-access portions away from MySQL dependency might be a good idea as well. But the simple fact is that Slashdot usually does work pretty well. IMHO, however, VA/Andover would probably be right to put some funding into creating a distributed version of Slash; the abstraction layer would be quite a helpful bit of hackery in and of itself./Brian
Stallman *is* a crackpot, IMHO, but not because he's wrong. It's because he's too radical for most people's tastes.
It seems to be an axiomatic thing about politics, most visibly demonstrated by the behavior of the US Republican party under Clinton: the more the moderates rule the roost, the more extreme the standard bearers of the wings become. Linus Torvalds has created a system that uses the GPL to its fullest advantage, yet repudiates the ivory-tower extremism of the FSF: give it away or sell it, just don't think you can own it. Companies like RedHat and SuSE are proving that this can work.
The problem is that this leaves Richard Stallman on the fringe, no longer in control of the philosophical movement he created. So he does the human thing: he backlashes. He tries to force the GNU/Linux issue. He rails against the corporatization of Linux, forgetting that commercial acceptance is critical to its future. He slams the open source movement because it doesn't do things the GNU way (check out his comments about APSL, for example). He has even demoted the Library GPL to Lesser GPL.
It would be wrong to say that rms doesn't have a point. But I'm not one of those people who agree with the phrase "extremism in defense of liberty is no vice" -- extremism is always a vice as it a) does not allow for the possibility of redefining a position if such proves to be necessary and b) tends to turn off those who you most have to reach -- those darned moderates again. Stallman is an extremist, and therefore (by *my* definition anyway, and that subjectivity should be very much acknowledged here) a crackpot.
In any case, I've often felt that rms likes to play the same embrace-and-extend game as Microsoft with no purpose other than to lock people into GNU. I've felt that way for a long time, ever since I read the part of the GCC manual that talks about the purpose of other compilers being to compile GCC. The creation of GUILE is another example; IMHO its only reasons for existence are that a) Stallman is not fond of tcl/tk and b) Stallman is a scheme junkie. Not because another tool was needed, but because *rms wants it that way*. The Apple boycott of years past is another thing -- the FSF was punishing AU/X and MacOS users for Apple's behavior, no matter that the Mac people might have as much to contribute as anyone else. In their own way, the FSF is no different from the commercial establishment they're fighting; it's like Steve Jobs vs. Bill Gates. The question is not whether one is worse than the other, it's a matter of who's holding the hammer at the moment.
Actually, just as an exercise I'd like to see someone create a diminished-GNU linux distro. I don't think it would be especially popular, since GNU programs tend to be the best in their class, but someone should try it on principle. The FSF to me has long resembled a child who comes to a party with a game idea, but then wants to take it and go home when other kids start adding rules, the kicker being that they've already given it away! (Though at least the gcc->egcs->gcc split-countersplit indicates they're smart enough to know when there's a better way sometimes...)
Do I think open-source is a good thing? Yer damn straight. It's made a penguin-lover out of this longtime Machead. Do I think rms should be honored for creating and managing the GNU project? Yes. The movement came before him and will outlast him, but he's the philosophical nexus. But do I think his behavior is a bit outlandish because of his unyielding philosophical positions?
Oh, yeah.
See, as I said, Linus has set the precedent for a kinder, gentler open source movement. IMHO it's time for rms to get in line and debate with some flexibility or just shut up and keep writing good software.
Actually, the jury's still out on OS X, which is what I suspect you're referring to.
To be honest, this is one thing I think should be brought up. Miguel has mentioned in the past that GNOME's user interface is ripped directly from Windows. I'd like to know whether he'd be interested in creating a GNOME UI guideline spec to create a cleaner interface.
I'd also like to suggest one particular thing for GTK/GNOME development: an easy way to turn off GTK tooltips. They were a blatant ripoff of the Mac Balloon Help when Microsoft cooked them up, and they don't even have the ability to be hidden once the user knows enough about the interface that they're more trouble than they're worth. I don't per se mind tool tips, but they're fairly obtrusive.
So yeah, that's my basic question. Will GNOME develop any formal UI standards of its own, or will it continue to be done in the Windows mold?
You're forgetting one thing. One can make a case that defeating CSS is protected under DMCA as reverse-engineering for interoperability (how can it be otherwise, is what I'd like to know). But DeCSS is essentially clean-room -- we still don't necessarily know the encryption algorithm, and without that you still can't burn a CSS-protected DVD. To be honest, DVD is, right now, too much. It is an incredible format -- there's no better way to watch a movie at home -- but production seems to be a bitch and the fact remains that the MPAA will fight to the death over the issue (I still think it's interesting that the DVD patent holders couldn't seem to care less about this whole flap). Wake up, movie people: with DeCSS on the scene, you're no better off than the RIAA vs. MP3. You've already lost -- give it up gracefully./Brian
Unlike many of you, I genuinely believe the real dmg isn't a troll, merely a non-geek who truly does not understand the open source culture. He seems like a fairly rational individual, clueless or misguided, but most likely not a troll.
Now, a comment or two about the Slackware spinoff to help the old/. karma rating...
IMHO this is the Right Thing for Slackware, being spun off. I've always had my doubts about whether being under Walnut Creek's umbrella was the best thing for it anyway; after all, Walnut Creek's biggest focus has been FreeBSD. I don't know that that helps Slackware, since it's a bit like Microsoft sublicensing Darwin from Apple, tacking on DCOM and Win32, and releasing it as Win2K for Macintosh. Slackware is second-class to cdrom.com (anyone wishing to say otherwise should be directed to the fact that cdrom.com runs on FreeBSD).
Spinning Slackware off will finally give it the strength to be a player again (face it, there's something nice about using the oldest still-extant distribution, but I use RedHat because it's better than my last Slack experience (v3.4)). I say more power to Volkerding &c. This situation can only help.
You come off as being a bit of a troll, but not so comprehensively so that you could be convicted in an American criminal court. So I'll treat your comments more as naive than flamebait.
I don't consider myself a guru at all, but I think I'm a somewhat sophisticated Linux user. I like the idea that if I do need to change something, what's in there is just an
$ emacs kernelfile.c &
away.
The main point that I need to make to the minority of hardcore GPL Linux zealots, is that the whole GPL thing is holding Linux back.
It might be worth pointing out that Linus himself isn't especially hardcore. The GPL was the best fit available at the time for Linux, given what he was out to do.
Linus Torvalds needs to produce a BINARY only release of Linux, and must take Linux back to the "closed source" model, if Linux is ever to compete with Microsoft on the desktop.
A bit of a history lesson might be in order here.
A lot of people have only come to know Linux after its commercial success; that's when I started using it myself (Slackware 3.6). But the commercial Linux explosion is a relatively recent phenomenon in the nine years or so that Linux has been around; originally, Linus created it in order to better understand i386 assembler and operating system design.
Linux is not inherently a commercial operating system. The many distributions out there (RedHat, SuSE, Slackware, etc.) are more or less commercial software in nature, but they are created in the same spirit (for the most part).
It's not enough to provide Joe Sixpack with a user friendly gui (KDE/Gnome) or a nice install tool. Joe Sixpack is quite simply not tech-savvy enough to be trusted with the source code to a complex system such as Linux. He probably can't even find the on-off switch on a bad day.
You're assuming that Joe Sixpack will want to muck around with the source code in the first place. Go pick up a copy of the Coriolis Group's Linux Kernel Code and Commentary; it's written in the same spirit as the old Lions book (a complete annotation of the source code for Unix V6). Granted, Joe Sixpack can't necessarily be trusted not to mangle something in there; not only is it all written in C (a programming language that most non-hackers would find rather cryptic and intimidating), it's written in nonstandard (GNU) C and therefore requires an extra level of sophistication above that. But a quick look through the book should indicate that Joe Sixpack would probably be smart enough not to get anywhere near it. (In any case, changing the source code won't do anything much unless you try to recompile anyway... all Linux distributions do ship ready to run, and you don't have to install the source if you don't want to!)
Anyone out there who has ever had to administrate the machines of a bunch of clueless users will be with me on this. How many times have you had to reinstall someones system because they screwed up their "autoexec.bat" file ? How many "config.sys" files have you seen trashed ? How many person-years of productivity have been lost because one of your users managed to fire up "regedit32.dll" and wreak havoc on the registry ?
How many times have you sat by waiting months or years for a bug in your system to get fixed? How many times have you heard of a "service release" including a few stealth "features" that wind up making your computing experience just a little less pleasant (ask a number of PowerMac G3 owners about Apple's little ROM update last year for example)? How many times have you seen a development schedule slip behind for years with no sign of some major, mission-critical feature becoming available? (Okay, Linux does this all the time, but at least you can get the beta kernels, and they're usually almost as stable as the production code. Closed-source vendors, on the other hand, will withhold for a long time on things like this. Another MacOS example: you know anyone running the Golden Master of Copland?)
Now imagine, in a Windows system there are only about 10 files the user can damage. Imagine now that the user has the whole source to the operating system ? Multiply the cluelessness of the average PC user by the number of opportunities for him/her to be clueless, and you will come up with a very large number indeed.
But it's a question of how badly the system can get tangled. A clueless user can make a mess of anything, true. But as I pointed out, what user is going to go mucking around in/etc and do anything except look?
In any case, "10 files that the user can damage" is a canard. I assume one of those files is the Registry. My understanding of the Registry is that it's like a master database of settings from around the system. Well, that makes the Registry akin to, well, every config file on the system mashed into one gigantic tarball. Sounds to me that the Unix method is better -- you trash one file, you trash one file, and damage is compartmentalized. You trash the Registry... think Armageddon.
As I said, the word is compartmentalization. There's lots more opportunity (at least in theory) to make a mess, but it's also a lot easier to contain the damage.
The user will never admit they screwed it up either. For Linux to succeed, Linus must take the source code back under his own copyright, and only make it available to certified Linux professionals, who can be trusted not to screw things up. To make sure they are serious about Linux, a nominal fee should be charged for the source code license (say about $1000).
And what purpose will that serve?
Okay, it is well within Linus's right to do that. The GPL is big on acknowledging ownership, and he can withdraw it for, say, Linux 2.4.
But then what happens? And this is where you betray your ignorance of the culture...
The source will not close. It will fork. There will be a closed Linux and an open Linux, since one of the big points of opening source in the first place is to make sure that it won't become obsolete when the formal project ends.
Your marketing model may make sense *to you*, but to anyone who understands Open Source it's nonsensical on its face.
Some people may whinge and say this is against the spirit of "open source", however, for Linux to survive it must adapt to its market, and in the consumer market place, the "closed source" model rules the roost. An examination of Microsoft (MSFT) stock price will demonstrate this clearly.
Here I need to grit my teeth and remind you that I am giving you the benefit of the doubt here. Closed Source has traditionally ruled the roost, yes, but I don't think a case can be made any longer that this is how it's going to be, forever and ever, world without end, amen. Linux is just too available. It was designed that way, and not too many people will be willing to give it up just because a marketing expert might believe that it's better off closed.
I think the point you're missing is the one I made above. Linux per se is not, never will be, never even can be a commercial product. It is developed according to the needs of its users, not the needs of Red Hat, VA Linux, SuSE, TurboLinux, etc. That will never change.
Truthfully, I don't think you really get the idea of Open Source. In addition to the Linux book I mentioned above, I suggest you hook up with a copy of the Lions Book (full title: Lions' Commentary on Sixth Edition Unix) from Peer-to-Peer Communications. If any one thing can be said to be the root of the Open Source movement it's this book, written by a former CS professor from Australia, which created the Unix culture as we know it today almost single-handedly (in 1976!). (And there's even more before that, but I'm not too familiar with the rest of that history...)
Oh, and in the future... if you want to avoid looking like a troll, log in nice and proper. The term "Anonymous Coward" is there for a reason.
Well... if it's so easy to do in perl, someone really ought to do it ASAP before the patent lawyers show up... relevant reference: http://www.assurdo.com Look for the info on PerlFS./Brian
Okay, I've seen a lot of complaints that Netscape is a bugfest, that IE is a better-quality browser, that Netscape is too slow to properly render a nude image of Natalie Portman pouring grits down her pants (pause so the humor-impaired can fire up the flamethrowers).
Those of you who have gone to great lengths to point that out have forgotten something: The only funky thing about NPL is that Netscape can use the code however it wants. But if you don't like it, you're free to fork. It's part of being open-source.
Another thing that's being forgotten is that Mozilla != Netscape, at least not 100%. It's a rather different product that happens to spring from the same original source base, and given its essential open-sourceness should be much more stable by the 6.1 release (if not earlier) than any previous version. Don't judge Mozilla based on its predecessors, then; the rules have changed.
Slashdot Mirror
If SCO is serious about Open Source, where does that leave older Unix code? I have the Lions Book, as many out there do. It makes no sense to me that if I could find a suitable compiler (which I can, at Dennis Ritchie's home page) that I have to pay $100 to get an Ancient Source Code license. Will the old stuff be made freely available? (What about current stuff, for that matter? How would an open-source SVR5/Unixware/whatever fit into SCO's business model?) /Brian
A few odds and ends about Microsoft's latest inroads into the PDA market...
I think it was in the Journal today that I read something about Microsoft finding out what Apple feels like. That's not really a fair comparison -- after all, Apple's situation was (largely through their own stubbornness) having the rug pulled out from under them. WinCE was never more than a reaction, and now that Palm owns the market I don't think there will be much for them to worry about.
I do think it might be in Palm's best interest to opensource PalmOS as a preemptive strike against MS attempting to force corporate hands (i.e. we can't guarantee that your Win2K machines will be able to HotSync properly after the next service pack release). I doubt their revenue stream is all that dependent on licensing anyway; they probably make the bulk of their money off of hardware.
At any rate, I think Microsoft is playing the same card they played to dilute the Network Computer revolution (a misguided idea to begin with, but bear with me). *Of course* users want extra functionality! *Of course* they want better integration!
You tell me: do you really need a PDA that can do everything a laptop can do? I don't know about anything else, but the day I can finally afford one what I'll be looking for is a smart notebook, not something I can use to replace my Mac and Linux box with on the road (that's what laptops are for).
I don't think I'd buy a WinCE machine (old or new) unless I could nuke WinCE and replace it anyway. That's just me, though.
/Brian
I think a lot of people may not be aware that much of the Win16 API is actually standardized by ECMA (go on, go to ecma.ch and ask them for a current CD -- it's free).
.ocxen, for example, and replacing the HTML-rendering components with Mozilla code.
Therein lies the key to your successfully functioning open source Windows community. Since I don't believe there is much risk of Windows forking too badly (the code is too big and probably too byzantine for anyone to really understand, I'm sure), Microsoft in principle hasn't much to worry about as far as losing control. An open source windows would most likely fork no more than once, and that into the hands of people who feel that they can beat Microsoft at their own game without destroying crosscompatibility.
More likely Microsoft winds up getting hit hard on the distribution level -- there will be a Win2xxx kernel, and there will be the Official Microsoft Distribution (you know it's official 'cos it's got the hologram), and then you'll have half a dozen other Windowses out there, binary-compatible but featuring vastly different mixes of high-level features. I could see someone slicing and dicing Mozilla into
I think it's probably reasonable to assume either way that even if Windows/Open Source forks in twelve different directions, anyone who does not maintain binary compatibility as Microsoft has defined in the Win32 developer's library will be dead on arrival.
The problem is that even after all is said and done, Microsoft still must be left with the ability to support itself. It's a reasonable thing to believe that they should be whipped, stripped and hung on high, but at the end of the day there has to be enough left to earn a profit. This is a truly grotesque situation, but to go beyond that would be excessive intrusion by the government.
A breakup can't be done particularly obviously, and structural remedies won't work because MS will crap on them. They can keep Office and their multimedia stuff, but the following packages should be opened up:
-Windows 2000 itself (to prevent unnecessary/predatory OS tying)
-All network service packages (IIS, Exchange, and IE especially) (to prevent abuse of standards)
-All APIs
Anyone got a problem with that?
/Brian
I haven't read the whole debate, so I can't comment for certain on Tanenbaum's attitude, but... I don't see much sense of accusing ast of arrogance here. It may not be inaccurate (I don't know the man so I can't say), but it's irrelevant. Minix was created strictly as an academic system. If it does it for you, fine. If it doesn't, well, that's why it's under BSD license now. You can change it or use something else. But saying that ast got snowed for being arrogant and resistant to change misses a few points: -Linus was mostly learning low-level 386 assembler at that point, not trying to create a revolution. That didn't even begin to gather steam until the 2.0 kernel showed up. -Andrew Tanenbaum is a CS professor, and his intent with Minix was not to create a commercially viable OS but to create a functioning toy that does everything it has to to teach the student *but nothing more*. If I want to learn how Linux works, I have the Coriolis book, but that's only the very basics of the system. If I want to learn Minix, that's much easier -- everything is there, not that much more complicated than the Lions book I imagine. Ripping ast for his design is shortsighted as well -- there are those out there that might conveniently forget that Linux is hardly cutting-edge. It's a monolithic kernel, built with no principle in mind save "make it work and work well". Minix is a microkernel-based OS, designed to illustrate and teach as much as being used. Blew up in his face? Why, because a system designed for simplicity and elegance didn't take over the world? Slow down, buddy. The OS wars aren't a zero-sum game, and Minix was never even fighting in them in the first place. /Brian
First off, many of you don't quite understand that Darwin *is* BSD. What NeXT/Rhapsody/MacOS X really is is a Mach-based BSD (similar in principle to Lites), with the Objective-C runtime developed by NeXT as part of the overall system. Darwin is that BSD system stripped to the basics: no GUI, no high-level APIs -- in other words, a vanilla BSD with some cool tricks up its sleeve.
The biggest problem with Darwin at this point is the almost-but-not-quite opensourceness of the APSL -- you can do what you want freely, but not behind closed doors because Apple wants a cut of your code. I don't think that's a big restriction in practice, but some find it philosophically limiting. Despite that, Apple's to be applauded for taking the free-ride approach to OS development; not only do they (in theory) make off like bandits, they also give their work back.
They're not free of reasons for criticism, however. They aren't making enough public. I want to see the wreckage of Copland put out there. I want to see Quicktime on Linux. Apple is dribbling things out there, but needs to show more of its cards and not worry about OS and service revenue (face it, that market will be shot to hell in two years anyway; why do you think there's a flat price for OS X Server to begin with?)
Darwin, IMHO, occupies a bit of a grey area -- if developers picked it up and ran with it, the BSD community would be that much stronger, and Apple would be able to generate some revenue by porting OS X to hell and back. But the APSL has to change, and Apple needs to put more on the line. Let's see Aqua go open-source. Let's see Apple cut loose on OS 7 through 9 -- the fact is that they won't be needing it anymore when OS X Client comes out, but a hell of a lot of non-G3 users will.
Just thinking...
/Brian
First off, using Oracle or Sybase sort of misses the point. The idea is that you *can* create something like this and go strictly open-source with it. Now I'm not all that familiar with the guts of Slashdot. But I know for a fact that the nice thing about opensource is that you can turn a pile of junk into something like a rather nice entry-level Beowulf or something of the sort, so cheap hardware for redundancy and high availability is no big deal. And I do suspect that a nice distributed-Slash project would be very helpful; abstracting the database-access portions away from MySQL dependency might be a good idea as well. But the simple fact is that Slashdot usually does work pretty well. IMHO, however, VA/Andover would probably be right to put some funding into creating a distributed version of Slash; the abstraction layer would be quite a helpful bit of hackery in and of itself. /Brian
Stallman *is* a crackpot, IMHO, but not because he's wrong. It's because he's too radical for most people's tastes.
It seems to be an axiomatic thing about politics, most visibly demonstrated by the behavior of the US Republican party under Clinton: the more the moderates rule the roost, the more extreme the standard bearers of the wings become. Linus Torvalds has created a system that uses the GPL to its fullest advantage, yet repudiates the ivory-tower extremism of the FSF: give it away or sell it, just don't think you can own it. Companies like RedHat and SuSE are proving that this can work.
The problem is that this leaves Richard Stallman on the fringe, no longer in control of the philosophical movement he created. So he does the human thing: he backlashes. He tries to force the GNU/Linux issue. He rails against the corporatization of Linux, forgetting that commercial acceptance is critical to its future. He slams the open source movement because it doesn't do things the GNU way (check out his comments about APSL, for example). He has even demoted the Library GPL to Lesser GPL.
It would be wrong to say that rms doesn't have a point. But I'm not one of those people who agree with the phrase "extremism in defense of liberty is no vice" -- extremism is always a vice as it a) does not allow for the possibility of redefining a position if such proves to be necessary and b) tends to turn off those who you most have to reach -- those darned moderates again. Stallman is an extremist, and therefore (by *my* definition anyway, and that subjectivity should be very much acknowledged here) a crackpot.
In any case, I've often felt that rms likes to play the same embrace-and-extend game as Microsoft with no purpose other than to lock people into GNU. I've felt that way for a long time, ever since I read the part of the GCC manual that talks about the purpose of other compilers being to compile GCC. The creation of GUILE is another example; IMHO its only reasons for existence are that a) Stallman is not fond of tcl/tk and b) Stallman is a scheme junkie. Not because another tool was needed, but because *rms wants it that way*. The Apple boycott of years past is another thing -- the FSF was punishing AU/X and MacOS users for Apple's behavior, no matter that the Mac people might have as much to contribute as anyone else. In their own way, the FSF is no different from the commercial establishment they're fighting; it's like Steve Jobs vs. Bill Gates. The question is not whether one is worse than the other, it's a matter of who's holding the hammer at the moment.
Actually, just as an exercise I'd like to see someone create a diminished-GNU linux distro. I don't think it would be especially popular, since GNU programs tend to be the best in their class, but someone should try it on principle. The FSF to me has long resembled a child who comes to a party with a game idea, but then wants to take it and go home when other kids start adding rules, the kicker being that they've already given it away! (Though at least the gcc->egcs->gcc split-countersplit indicates they're smart enough to know when there's a better way sometimes...)
Do I think open-source is a good thing? Yer damn straight. It's made a penguin-lover out of this longtime Machead. Do I think rms should be honored for creating and managing the GNU project? Yes. The movement came before him and will outlast him, but he's the philosophical nexus. But do I think his behavior is a bit outlandish because of his unyielding philosophical positions?
Oh, yeah.
See, as I said, Linus has set the precedent for a kinder, gentler open source movement. IMHO it's time for rms to get in line and debate with some flexibility or just shut up and keep writing good software.
/Brian
Actually, the jury's still out on OS X, which is what I suspect you're referring to.
To be honest, this is one thing I think should be brought up. Miguel has mentioned in the past that GNOME's user interface is ripped directly from Windows. I'd like to know whether he'd be interested in creating a GNOME UI guideline spec to create a cleaner interface.
I'd also like to suggest one particular thing for GTK/GNOME development: an easy way to turn off GTK tooltips. They were a blatant ripoff of the Mac Balloon Help when Microsoft cooked them up, and they don't even have the ability to be hidden once the user knows enough about the interface that they're more trouble than they're worth. I don't per se mind tool tips, but they're fairly obtrusive.
So yeah, that's my basic question. Will GNOME develop any formal UI standards of its own, or will it continue to be done in the Windows mold?
/Brian
You're forgetting one thing. One can make a case that defeating CSS is protected under DMCA as reverse-engineering for interoperability (how can it be otherwise, is what I'd like to know). But DeCSS is essentially clean-room -- we still don't necessarily know the encryption algorithm, and without that you still can't burn a CSS-protected DVD. To be honest, DVD is, right now, too much. It is an incredible format -- there's no better way to watch a movie at home -- but production seems to be a bitch and the fact remains that the MPAA will fight to the death over the issue (I still think it's interesting that the DVD patent holders couldn't seem to care less about this whole flap). Wake up, movie people: with DeCSS on the scene, you're no better off than the RIAA vs. MP3. You've already lost -- give it up gracefully. /Brian
This ain't dmg, folks.
/. karma rating...
Unlike many of you, I genuinely believe the real dmg isn't a troll, merely a non-geek who truly does not understand the open source culture. He seems like a fairly rational individual, clueless or misguided, but most likely not a troll.
This guy? Troll. Whiner. Flamebaiter. Whatever. Ignore him, there's enough flamage
Now, a comment or two about the Slackware spinoff to help the old
IMHO this is the Right Thing for Slackware, being spun off. I've always had my doubts about whether being under Walnut Creek's umbrella was the best thing for it anyway; after all, Walnut Creek's biggest focus has been FreeBSD. I don't know that that helps Slackware, since it's a bit like Microsoft sublicensing Darwin from Apple, tacking on DCOM and Win32, and releasing it as Win2K for Macintosh. Slackware is second-class to cdrom.com (anyone wishing to say otherwise should be directed to the fact that cdrom.com runs on FreeBSD).
Spinning Slackware off will finally give it the strength to be a player again (face it, there's something nice about using the oldest still-extant distribution, but I use RedHat because it's better than my last Slack experience (v3.4)). I say more power to Volkerding &c. This situation can only help.
/Brian
Allow me to add my voice to the chorus...
/etc and do anything except look?
You come off as being a bit of a troll, but not so comprehensively so that you could be convicted in an American criminal court. So I'll treat your comments more as naive than flamebait.
I don't consider myself a guru at all, but I think I'm a somewhat sophisticated Linux user. I like the idea that if I do need to change something, what's in there is just an
$ emacs kernelfile.c &
away.
The main point that I need to make to the minority of hardcore GPL Linux
zealots, is that the whole GPL thing is holding Linux back.
It might be worth pointing out that Linus himself isn't especially hardcore. The GPL was the best fit available at the time for Linux, given what he was out to do.
Linus Torvalds needs to produce a BINARY only release of Linux, and must
take Linux back to the "closed source" model, if Linux is ever to compete with
Microsoft on the desktop.
A bit of a history lesson might be in order here.
A lot of people have only come to know Linux after its commercial success; that's when I started using it myself (Slackware 3.6). But the commercial Linux explosion is a relatively recent phenomenon in the nine years or so that Linux has been around; originally, Linus created it in order to better understand i386 assembler and operating system design.
Linux is not inherently a commercial operating system. The many distributions out there (RedHat, SuSE, Slackware, etc.) are more or less commercial software in nature, but they are created in the same spirit (for the most part).
It's not enough to provide Joe Sixpack with a user friendly gui (KDE/Gnome)
or a nice install tool. Joe Sixpack is quite simply not tech-savvy enough to be
trusted with the source code to a complex system such as Linux. He probably
can't even find the on-off switch on a bad day.
You're assuming that Joe Sixpack will want to muck around with the source code in the first place. Go pick up a copy of the Coriolis Group's Linux Kernel Code and Commentary; it's written in the same spirit as the old Lions book (a complete annotation of the source code for Unix V6). Granted, Joe Sixpack can't necessarily be trusted not to mangle something in there; not only is it all written in C (a programming language that most non-hackers would find rather cryptic and intimidating), it's written in nonstandard (GNU) C and therefore requires an extra level of sophistication above that. But a quick look through the book should indicate that Joe Sixpack would probably be smart enough not to get anywhere near it. (In any case, changing the source code won't do anything much unless you try to recompile anyway... all Linux distributions do ship ready to run, and you don't have to install the source if you don't want to!)
Anyone out there who has ever had to administrate the machines of a bunch of
clueless users will be with me on this. How many times have you had to
reinstall someones system because they screwed up their "autoexec.bat" file ?
How many "config.sys" files have you seen trashed ? How many person-years
of productivity have been lost because one of your users managed to fire up
"regedit32.dll" and wreak havoc on the registry ?
How many times have you sat by waiting months or years for a bug in your system to get fixed? How many times have you heard of a "service release" including a few stealth "features" that wind up making your computing experience just a little less pleasant (ask a number of PowerMac G3 owners about Apple's little ROM update last year for example)? How many times have you seen a development schedule slip behind for years with no sign of some major, mission-critical feature becoming available? (Okay, Linux does this all the time, but at least you can get the beta kernels, and they're usually almost as stable as the production code. Closed-source vendors, on the other hand, will withhold for a long time on things like this. Another MacOS example: you know anyone running the Golden Master of Copland?)
Now imagine, in a Windows system there are only about 10 files the user can
damage. Imagine now that the user has the whole source to the operating
system ? Multiply the cluelessness of the average PC user by the number of
opportunities for him/her to be clueless, and you will come up with a very
large number indeed.
But it's a question of how badly the system can get tangled. A clueless user can make a mess of anything, true. But as I pointed out, what user is going to go mucking around in
In any case, "10 files that the user can damage" is a canard. I assume one of those files is the Registry. My understanding of the Registry is that it's like a master database of settings from around the system. Well, that makes the Registry akin to, well, every config file on the system mashed into one gigantic tarball. Sounds to me that the Unix method is better -- you trash one file, you trash one file, and damage is compartmentalized. You trash the Registry... think Armageddon.
As I said, the word is compartmentalization. There's lots more opportunity (at least in theory) to make a mess, but it's also a lot easier to contain the damage.
The user will never admit they screwed it up either. For Linux to succeed,
Linus must take the source code back under his own copyright, and only make
it available to certified Linux professionals, who can be trusted not to screw
things up. To make sure they are serious about Linux, a nominal fee should be
charged for the source code license (say about $1000).
And what purpose will that serve?
Okay, it is well within Linus's right to do that. The GPL is big on acknowledging ownership, and he can withdraw it for, say, Linux 2.4.
But then what happens? And this is where you betray your ignorance of the culture...
The source will not close. It will fork. There will be a closed Linux and an open Linux, since one of the big points of opening source in the first place is to make sure that it won't become obsolete when the formal project ends.
Your marketing model may make sense *to you*, but to anyone who understands Open Source it's nonsensical on its face.
Some people may whinge and say this is against the spirit of "open source",
however, for Linux to survive it must adapt to its market, and in the consumer
market place, the "closed source" model rules the roost. An examination of
Microsoft (MSFT) stock price will demonstrate this clearly.
Here I need to grit my teeth and remind you that I am giving you the benefit of the doubt here. Closed Source has traditionally ruled the roost, yes, but I don't think a case can be made any longer that this is how it's going to be, forever and ever, world without end, amen. Linux is just too available. It was designed that way, and not too many people will be willing to give it up just because a marketing expert might believe that it's better off closed.
I think the point you're missing is the one I made above. Linux per se is not, never will be, never even can be a commercial product. It is developed according to the needs of its users, not the needs of Red Hat, VA Linux, SuSE, TurboLinux, etc. That will never change.
Truthfully, I don't think you really get the idea of Open Source. In addition to the Linux book I mentioned above, I suggest you hook up with a copy of the Lions Book (full title: Lions' Commentary on Sixth Edition Unix) from Peer-to-Peer Communications. If any one thing can be said to be the root of the Open Source movement it's this book, written by a former CS professor from Australia, which created the Unix culture as we know it today almost single-handedly (in 1976!). (And there's even more before that, but I'm not too familiar with the rest of that history...)
Oh, and in the future... if you want to avoid looking like a troll, log in nice and proper. The term "Anonymous Coward" is there for a reason.
/Brian
Well... if it's so easy to do in perl, someone really ought to do it ASAP before the patent lawyers show up... relevant reference: http://www.assurdo.com Look for the info on PerlFS. /Brian
Okay, I've seen a lot of complaints that Netscape is a bugfest, that IE is a better-quality browser, that Netscape is too slow to properly render a nude image of Natalie Portman pouring grits down her pants (pause so the humor-impaired can fire up the flamethrowers).
Those of you who have gone to great lengths to point that out have forgotten something: The only funky thing about NPL is that Netscape can use the code however it wants. But if you don't like it, you're free to fork. It's part of being open-source.
Another thing that's being forgotten is that Mozilla != Netscape, at least not 100%. It's a rather different product that happens to spring from the same original source base, and given its essential open-sourceness should be much more stable by the 6.1 release (if not earlier) than any previous version. Don't judge Mozilla based on its predecessors, then; the rules have changed.
/Brian