"Yes, I know it's basically impossible to write non-trivial bug free software, and generally impossible to verify that software is bug-free."
I've wondered about this for a long time. At some level, computers are deterministic machines, and software is a concrete set of instructions. I know there exist some tools to use logical proof languages to create provably correct software. I think the trick becomes at that point deciding exactly what you want the software to do.
Why not break software down into a series of components. For example, a spreadsheet is composed of graphical widgets, IO routines, algorithms, and other stuff I'm forgetting. Each of those broad categories has many, many pieces that go into making up the whole, and those pieces are in turn built on other pieces. At some level though, things much break down into some set of logical operations which cannot be reduced further - each element unique, and well defined in behavior. If these elements can be built from the binary/assembly level in a provable manner, from those elements more complicated elements could be created using boundary condition studies (for example, regardless of input, an IO operation only would accept data formatted according to specification y, and in all other cases cleanly fails.) From there build on the proven properties of each more complex component until useful software is arrived at.
It isn't reasonable to expect someone to sit at a blank terminal and create perfect software. BUT, mightn't it be possible to build a system from the most basic components up, proving behavior in stages like building mathematical proofs one on top of the other? Just as mathematical proofs can build on one another, couldn't basic OS and software ideas be implimented one on top of the other? I don't know, I am not an expert. And I do see that this would essentially involve recreating every piece of useful software in the world from the ground up, starting with the operating system. And I know the design questions might need to be addressed in a different way from the coding proof issues. But is there a fundamental reason such a building blocks approach couldn't work?
One way to look at it might be as follows: the last decades have allowed us to find out what we want to do with software, and I think that is fairly well understood. Now that we know what we want to do, perhaps the time has come to do it in a robust manner. One software program can be duplicated millions of times all over the world - if it can be engineered in such a way as to guarantee with the force of a logical proof that it will behave as designed and flesh out the designs with flowcharting, diagramming, and maybe proofs if appropriate, wouldn't that be worth doing? Open source could be the perfect medium for such methods to take hold - commercial companies would never make the effort unless it was easy, which would require a support structure that won't come into being from any probable commercial entity. Perhaps open source can lead the charge.
'You're missing the point. If you're going to do something, do it the best you possibly can. When did people stop caring about quality?'
It's a question of allocation of effort. When someone is volunteering their spare time to work on a project such as an open source program, the expectations of rigor are only those which the programmer cares to self impose.
Is you every piece of clothing of the highest quality cloth and craftsmanship? Is all your furniture hardwood and are all your floors high quality material designed to last for fifty years? Do you only use high end Sun hardware for it's well build and designed characteristics? I'm guessing probably not - if so you are virtually unique. In most cases it's a question of how to allocate resources to achieve an objective. Grade A work is very difficult. Bully to you if you can do it all the time, but most of us mortals aren't built like that.
'This goes complete against what the open-source fanatics have been spouting for years.. "Open source is just as good, if not better, than closed-source".. ad infinitum, ad nauseum.'
How does it go against anything? It might be as good or better. You still don't get to berate people for it unless you pay for the privilage.
'What kind of hypocrisy does it take to claim on one hand "Open source is better than closed source", and on the other "Well, if it doesn't work, you didn't pay for it anyway..." ??? '
No hypocrisy at all. When people say "open source is better than closed source" they typically refer to the development model. I'll add open source is better for ME, because I want at least the potential of complete control over my computer, but I suspect I'm an exception.
No one is (or at least should be) saying all open source programs are more capable than their counterparts in the commercial world. That would be folly. But the way of doing things IS better, so long as control is not the issue for the developer and the user is willing to assume responsibility for making things work the way they want them to. Granted that is not a responsibility many users would care to assume, but for those of us who do care to assume it there is no better way than open source. Remember, the word "better" means different things to different people. To you, it appears to mean technical quality. To me, it is more than that - it is openness, philosophy, community, and the freedom to not worry about licensing issues. The technical proficiency of the software, whatever it may be, is a consequence of these other traits. Personally I find it a much nicer environment in which to spend my time, and so long as minimal functionality requirements are met technical merit becomes secondary. You have a different outlook, and so open source is not for you. I suggest you avoid it if at all possible, because the priorities there are completely different from yours.
'My whole point here is that developers should be releasing the best code they can create and owning up to (and fixing) their mistakes -- and it doesn't matter in this instance if you release your code as open source, or include it in a proprietary piece of work.'
Ah, but it does. Quality of code is a consequence of effort put forth - being paid for proprietary code places an additional burden on the programmer, because they presumably have more resources to commit to the problem.
'Shirking responsibility for creating good code (even if that code was "free") is not a trait I would admire in another developer.'
You don't understand. There is no "shirking" in the case of open source because no responsibility is implied. I understand full well when I use open source software I am dependant on the will of the developers to continue, and I have no power to force them to do anything. That is the extra burden I carry in return for being able to use the code and have rights to the source code without cost. I knowingly and willingly make that bargin - I do not want to hold the programmer's nose to the grind stone! That is the surest way to make them unhappy and abandon the project. Open source quality isn't bad in my experience, and if you don't like it there are plenty of commercial places you can obtain software on your terms.
Hmm. I really hope they do have thumbnail and bookmark support, and continue to add features. Xpdf is a nice renderer, but the interface IMHO is not exactly a nice one. If gpdf can become the full equal of Acrobat Reader I'll be one very happy camper.
A new GTK file selector. FINALLY. I can't wait to use the new one - the old one was one of the great warts of the free desktop world, IMHO.
But they have decided to remove the text entry box??? Eeep. I guess having the Ctrl-l shortcut to get one is OK (after all, it will most likely be geeks that want direct text on a file open) but thats one they need to document WELL.
On the whole though, it might be a good thing. I guess we'll have to wait and see. But text box or not, it can hardly be worse than the old one.
"Spoken like a true loser/script kiddie/whatever -- completely unprofessional."
Here's a hint - not everyone is professional at everything they want to do. Professionals get paid.
"Software developers should take pride in their work, and shouldn't release things that they aren't proud to claim."
Again, a proper attitude for the commercial world - NOT the hobby world.
"And the "and if they relied on it oh well, their fault...". What kind of an attitude is this? Is this the basic open-source "it's free software, so don't complain if it doesn't work" attitude? Or is this guy just a bigger loser than most?"
Excuse me - why should you have any guarantee of quality if you haven't paid me for my work? If you use free software you accept as your part of the deal that the software is not bound by commercial standards. That may or may not mean anything - some free software is very, very good - but it does mean you have no right to complain. You PAY for the right to complain. Please feel free to buy commercial software and complain about it's quality AFTER YOU HAVE PAID FOR THE PRIVILAGE. But don't use someone's hard work without compensation and then complain about the quality of the work. You have no right.
"If this idiot's opinions are indicative of how the open source community views their work, then the world is right to avoid it as much as possible"
If the world has your attitude, I hope it does. The open source community seems to be doing just fine on its own.
Tried to email, got an inbox full message. If you do start a lobbying effort, please try to get an article on slashdot about it. That will give it enough publicity to get some real momentum going.
"Ballmer said governments that abandon Microsoft are more interested in making a political statement than using the best and most affordable software."
What constitutes the "best" software? Most features? Maybe. Best stability? Maybe. Best security? Maybe.
For some situations and groups, the best software is software the furthers goals like avoiding dependance on a foreign company with a bad track record for business practices and near monopoly control. Like, say, foreign governments. Which are, after all, political institutions. Why wouldn't they make political statements?
Microsoft gets it, all right. They will do their best to make decisions other than for immediate $$ spent look silly, but for some in this world there really is more to it than that. Microsoft knows to fear thinking like this, because it cannot be controlled.
"The people who are making business decisions based on where are the applications, what is the value, what is the lowest cost of ownership, we're not losing them."
At, but there again value and cost of ownership are not always strictly a matter of $$. Frankly, it's a pretty cold world when that is true, and it's one of the things I dislike about the US. In any case, to solve the chicken egg problem of applications first or users first, the users typically have to take the plunge.
Ballmer can chuckle all he likes. What he isn't mentioning is that first adapters always, ALWAYS, have a hard time. Did we make fun of the first people who bought those really expensive first generation DVD players? Do universities shrink away from paying Peoplesoft $$$$$$$$$ for rather unimpressive systems that still need lots of tweaking? (I'm still convinced if a couple of them had hired GNU enterprise with that $$$ everyone would have been better off, but that's another post.) Change is tough. But for each person or group that makes the change, things are ironed out and it gets easier next time around. And as things get easier, a proven track record emerges, and the trail is paved, more people start to go down it.
So sure, Munich is chopping down trees to make a road through the forest right now. But the next time around someone else will have an example to follow, and will also do some more road clearing.
I'm quite sure if Munich had made the decision to switch over to Macintosh, they'd be facing many of the same problems. To a certain extent change is just hard, period. But the thinking here is long term, not short term. The Media reports short term, Microsoft laughs in the short term. But I'm a lot more interested in the long term, when Munich can look at the next upgrade cycle prices for Windows and laugh in their face.
"maybe they'll say that they'll accept the terms of GPL until such time as it is declared invalid (effectively squirming out of this particular mess)."
Nope. Can't do that, or they'd be distributing the Linux kernel, including their own IP, under a GPL they have temporarily accepted as valid, and so their whole IBM/license thing falls apart. (Of course, IANAL. Reality may vary from logic.)
If they were going to try to break the GPL into the public domain, they should have tried that FIRST. I suspect this particular consequence of their rhetoric did not occur to them until now.
Another example of why pissing off the community you live in (in SCO's case, the Unix world) is a Bad Thing.
Absolute numbers are fine, but what about normalizing it for the total number of BSD, Linux, and Windows servers in use in this study? That's the more meaningful number. Then, what constitutes a successful attack?
Also, a useful study would look at how machines are maintained, password policies, etc.
Now before I come off sounding like a Linux apologist, it is quite possible there are some serious weaknesses that need to be addressed. If so, I hope they give us full info on the attacks so we can fix the problems. But these numbers as they stand don't tell us a darn thing.
If a dedicated admin configures Selinux and heavy duty firewalls, and puts Klingon password policies in place, I'd personally still be confident to match that system against anything out there. Default Redhat installs, on the other hand, are something else again. So again we need more info. It's all in how things are set up and maintained. The question actually being asked here - which OS is strongest, all other things being equal - is a really really tough one to answer. There are many other issues that must be addressed first.
So, as far as any useful information is concerned, this article doesn't appear to have any. What if the Linux machines simply had the best intrusion detection in place? (I'm not saying they did, but it's a fair question.) Need More Information!
I'm starting to realise that comparing Fresco to X is a fairly pointless exercise. Fresco is concerned first, last, and only with Getting It Right. This does not lend itself to a rapid practical implimentation.
I hesitate to define Fresco as a failure. It hasn't replaced X, true, and probably won't for a very long time, but it does do interesting and new things. That's always a gutsy thing to do, and always to be applauded. They have explored a lot of options and written some working code, which is much further than most such projects get. Let's consider it windowing system research first, and a project to replace X as a remote second. It makes more sense in that light.
might very well be Mac on Linux, ironically enough. Check out this:
http://www.maconlinux.org/sshots/pic12.jpg
Running multiple versions of MacOS in parallel. Think about the possibilities for software developers. Having multiple environments immediately available for testing.
Then of course there's the ability to run all those Mac apps when needed and still have the Linux desktop to go to when they aren't needed.
Mac on Linux is what the open source world should try and create for the Windows world. Think of the possibilities if you could run Windows at work in a Window - be able to do all the windows specific stuff at need, but have Linux goodness in which to work as well. If a phb strolls in, just flip your desktop over to full screen windows. Then for the rest of the time go stealth with Xpde, good enough to fool a casual glance. Maybe some rootless window hack could even be figured out.
Of course, if your boss says you Must Use Windows, there's not a whole lot you can do. But perhaps this would be an acceptible compromise.
"When people writing code for business are optimizing for speed and redundancy mainly in the parallel sense (i.e., a failsafe swap to a sister server), how RELEVANT is that to blocks of code written never, ever, ever, ever, ever to fail on tested but "outdated" hardware?"
Not sure. Well written code is not to be scorned, whatever else might have happened in computers - Maxima was written decades ago and much of that code is quite useful today. (And that isn't high quality NASA code either!) What I'm hoping for, however, is the systems they use to develop that software. "Perfect" software in the sense of security I'd argue is currently the #1 concern for the internet, because there are more than enough people willing to be the one to knock over the dominio stack. Whatever software NASA uses to develop and document their "perfect" code, I'd like to get my hands on it. Who knows, maybe it will help everyone.
If this isn't a call to take a closer look at the possibility of more widely using tools like Z and B to develop important software, I don't know what is.
Yes, they're difficult. Yes, they aren't likely to eliminate all bugs. BUT. They provide a much better chance (as I understand it - I'm not an expert) that what is designed is what actually gets implimented. That shifts the burden onto the design, but that's OK - that burden was always there. It just means that the design gets properly implimented, which is all that can reasonably be asked of the coding process.
Currently, again as I understand it, the life of a software program in development is a constant struggle by the developers to cope with ever changing demands of customers. I think if people want matters to improve the customers are going to have to come to grips with reality, take the time to sit down and think things through, and make all critical design decisions BEFORE the development process begins. More expensive up front? You bet. That's why I think companies should look at cooperative effort for this type of thing. Distribute the cost of developing one really good program across an industry. A lot of the same core functionality can likely be shared between businesses - if they all pay for one proper design and implimentation of an open program up front, and they all get copies of the logic and proof code with rights to extend as they see fit, they all benefit. They can also open up the more general parts of the package to the world at large under GPL, and anyone could contribute who can generate valid B and Z designs/proofs. Sort of an "academic" open source code development forum - peer review and all. The companies get the benefit of all new development - if they are using it internally they can extend the GPL code for themselves, so long as they don't distribute it. If they do distribute it, they can so so under GPL for everyone to enhance. A plugin based model can also allow them to develop components to the system they can sell as commercial software, if they wish.
Whether this would work/appeal with corporate thinking I have no idea - many of those folks seem to view cooperation like the plague. But it might allow a higher grade of software to be developed and universally used, and I have a hard time imagining how that could be a bad thing for anyone.
"78. However, as is widely reported and as IBM executives knew, or should have known, a significant flaw of Linux is the inability and/or unwillingness of the Linux process manager, Linus Torvalds, to identify the intellectual property origins of contributed source code that comes in from those many different software developers. If source code is code copied from protected UNIX code, there is no way for Linus Torvalds to identify that fact."
Um. If source code was copied from protected UNIX code, how the @#$@%@# would Linus know about it? He doesn't have access to the protected source code - it's protected! The only way to know is if the owners of the protected source code make the claim and are able to back it up! How can Torvalds be faulted for not being clarvoyant? Do they mean identify it after the fact? AFAIK no one can say yet that the origins of code X can't be identified. SCO hasn't even let us TRY - they won't tell us what they want identified!
If what they are actually saying is that open source shouldn't be allowed to proceed simply because it doesn't have massive paperwork assigning every bit of code to some source, they've been hitting the crack again. Email archives, content management back trails anyone? And we can go further than that if we are really forced to - the FSF has been getting copyrights assigned to it for years just in case things come to that pass, and if it becomes utterly necessary that might become common practice.
What I'm hoping will come out of all this is a way that open source projects can set themselves up so that no one can sue them without them actually having done something wrong. (OK, OK - I know anyone can still bring the lawsuit. I mean create a situation where the project dispose of the suit in such a way that it doesn't cost the project or developers much of anything and discourages idiots like SCO from attempting it.) That would be useful, and if SCO is the start of a trend may become very necessary.
What's the incentive for Microsoft to stop their abuse? The abuse nets more money than any fine is likely to take away, and is the quickest way to make $$ back after the fine. This won't solve a darn thing.
The effective solutions (start multiple companies off with the Windows source code and have them compete, for example) are very radical, and I don't know if most of them are in the power of the EU. But if the US government is any example, the will to use them isn't there anyway, so Microsoft can write their check and go back to business as usual.
"Even if after everything we've heard from them
to date falls through, they may try to make
the claim in court that every OS in existence
is derived from SCO IP, and that being the
case Linux users STILL owe SCO money,
regardless of code."
Not gonna happen.
Yes. That's what I thought about some small company with silly claims suing IBM straight off for $3 billion, instead of picking a smaller fish to start with. Or about them taking the bluff to the point where they appear ready to sue on a scale the RIAA would envy. I'm hoping you're right, but nothing I have seen thus far indicates to me they wouldn't try it.
"No actually the details do matter. They matter in the court of law. Contrary to popular belief you can't just buy your way into winning every lawsuit."
The court of law isn't the battle SCO is worried about winning. They are trying to win the stock market game, get their name in front of the world, and hurt Linux/open source. The motivations for the latter are a matter of speculation, but clearly their actions and statements have that overtone. Their legal action is a long shot, and they must know that (barring incredible stupidity, which I suppose I can't rule out). So if they know that, what are their motivations? They aren't doing this for nothing.
"Yea, so what? MS wants linux destroyed as well and you how well that's worked for them."
Microsoft hasn't made ANY serious overt moves to destroy Linux. Not yet. They're still fighting regulatory stuff across the pond, and while in a practical sense the antitrust case in the US is dead they were ruled a monopoly. That's why so many people wonder if they are pulling SCO's puppet strings - they don't have to get down and dirty themselves, and thus avoid extra trouble. We're too complacient about Microsoft. SCO is the current issue, but once they're gone Microsoft won't just concede ground to Linux. They will fight, and do whatever it takes to win. If they can't win technically, they'll try to get the laws changed. The only thing I'm sure the WON'T do is nothing.
"Look, your just acting a bit hysterical here. Facts and law do matter. And contrary to your sky is falling ideas Linux isn't going to be destroyed by anyone."
Facts only matter in certain kinds of fights. Law matters, but it is slow and big money interests can lobby for new laws. I don't think SCO can stop Linux. But confining the analysis of this situation to SCO is using tunnel vision. Linux has and will have enemies, and they will learn from what SCO does. I don't have enough faith in the US political system to trust the lawmakers not to pass legislation that heavily favors commercial software, and if THAT happens SCO will be the least of our worries.
"You can't just dismiss that and say "well SCO will just think of some angle to spin this." My advice to you is relax a bit and ease off the coffee. "
I never said it wasn't important to the legal case. But the PR war may be lost by the time the courts get done with appeals processes. The legal system is slow - the software world moves at warp speed. DR-DOS won a decision against Microsoft, but by the time it did it was irrelevant.
"This case will resolve itself and there is no indication from any repudiable legal counsel or expert that SCO will be the end of Linux."
SCO may just be the beginning. Money and power defend themselves. Linux and open source threaten both. Ergo, we can expect trouble. It's as simple as that.
They might confess to past bad internal security in the interest of pursuing current licensing revenue. I'm not a legal person, so I don't know what all the various possibilities and ramifications are. My point is they aren't going to shut up just because they are in the wrong.
"How do you propose to fight them, if not by showing their claims to be incorrect?"
Essentially, in the specific case of SCO, starvation. SCO can't survive indefinitely if they don't get any revenue from anywhere, while paying their lawyers.
I'm not entirely convinced we will ever be rid of SCO, at least in spirit. The use of the legal system as a club against free products is something I have long dreaded, and I don't know any effective countermeasures short of changing the law itself.
I guess that would be my proposed way to fight - lobby for protection of free products against abuse of the legal system. I rather doubt a technical solution to a legal problem exists, so I suggest we look for legal solutions and defenses.
The letter to Congress from SCO a while back was the ultimate illustration of my concerns - there is nothing fundamentally that prevents Congress from saying open source software is illegal, because of security risk|unfair competition|what have you. When politically powerful corporations finally find out they can't FUD us to death or out compete us, they will try to use their political clout to get that accomplished. Never doubt it. And since open source folk don't mean much to re-election minded officials, I'd say our position is not the best.
SCO may be just the tip of the iceberg, a harbringer of the real war. We need to keep that in mind, so if that day comes we are mentally ready to meet it.
Slashdot Mirror
"Yes, I know it's basically impossible to write non-trivial bug free software, and generally impossible to verify that software is bug-free."
I've wondered about this for a long time. At some level, computers are deterministic machines, and software is a concrete set of instructions. I know there exist some tools to use logical proof languages to create provably correct software. I think the trick becomes at that point deciding exactly what you want the software to do.
Why not break software down into a series of components. For example, a spreadsheet is composed of graphical widgets, IO routines, algorithms, and other stuff I'm forgetting. Each of those broad categories has many, many pieces that go into making up the whole, and those pieces are in turn built on other pieces. At some level though, things much break down into some set of logical operations which cannot be reduced further - each element unique, and well defined in behavior. If these elements can be built from the binary/assembly level in a provable manner, from those elements more complicated elements could be created using boundary condition studies (for example, regardless of input, an IO operation only would accept data formatted according to specification y, and in all other cases cleanly fails.) From there build on the proven properties of each more complex component until useful software is arrived at.
It isn't reasonable to expect someone to sit at a blank terminal and create perfect software. BUT, mightn't it be possible to build a system from the most basic components up, proving behavior in stages like building mathematical proofs one on top of the other? Just as mathematical proofs can build on one another, couldn't basic OS and software ideas be implimented one on top of the other? I don't know, I am not an expert. And I do see that this would essentially involve recreating every piece of useful software in the world from the ground up, starting with the operating system. And I know the design questions might need to be addressed in a different way from the coding proof issues. But is there a fundamental reason such a building blocks approach couldn't work?
One way to look at it might be as follows: the last decades have allowed us to find out what we want to do with software, and I think that is fairly well understood. Now that we know what we want to do, perhaps the time has come to do it in a robust manner. One software program can be duplicated millions of times all over the world - if it can be engineered in such a way as to guarantee with the force of a logical proof that it will behave as designed and flesh out the designs with flowcharting, diagramming, and maybe proofs if appropriate, wouldn't that be worth doing? Open source could be the perfect medium for such methods to take hold - commercial companies would never make the effort unless it was easy, which would require a support structure that won't come into being from any probable commercial entity. Perhaps open source can lead the charge.
'You're missing the point. If you're going to do something, do it the best you possibly can. When did people stop caring about quality?'
It's a question of allocation of effort. When someone is volunteering their spare time to work on a project such as an open source program, the expectations of rigor are only those which the programmer cares to self impose.
Is you every piece of clothing of the highest quality cloth and craftsmanship? Is all your furniture hardwood and are all your floors high quality material designed to last for fifty years? Do you only use high end Sun hardware for it's well build and designed characteristics? I'm guessing probably not - if so you are virtually unique. In most cases it's a question of how to allocate resources to achieve an objective. Grade A work is very difficult. Bully to you if you can do it all the time, but most of us mortals aren't built like that.
'This goes complete against what the open-source fanatics have been spouting for years.. "Open source is just as good, if not better, than closed-source".. ad infinitum, ad nauseum.'
How does it go against anything? It might be as good or better. You still don't get to berate people for it unless you pay for the privilage.
'What kind of hypocrisy does it take to claim on one hand "Open source is better than closed source", and on the other "Well, if it doesn't work, you didn't pay for it anyway..." ??? '
No hypocrisy at all. When people say "open source is better than closed source" they typically refer to the development model. I'll add open source is better for ME, because I want at least the potential of complete control over my computer, but I suspect I'm an exception.
No one is (or at least should be) saying all open source programs are more capable than their counterparts in the commercial world. That would be folly. But the way of doing things IS better, so long as control is not the issue for the developer and the user is willing to assume responsibility for making things work the way they want them to. Granted that is not a responsibility many users would care to assume, but for those of us who do care to assume it there is no better way than open source. Remember, the word "better" means different things to different people. To you, it appears to mean technical quality. To me, it is more than that - it is openness, philosophy, community, and the freedom to not worry about licensing issues. The technical proficiency of the software, whatever it may be, is a consequence of these other traits. Personally I find it a much nicer environment in which to spend my time, and so long as minimal functionality requirements are met technical merit becomes secondary. You have a different outlook, and so open source is not for you. I suggest you avoid it if at all possible, because the priorities there are completely different from yours.
'My whole point here is that developers should be releasing the best code they can create and owning up to (and fixing) their mistakes -- and it doesn't matter in this instance if you release your code as open source, or include it in a proprietary piece of work.'
Ah, but it does. Quality of code is a consequence of effort put forth - being paid for proprietary code places an additional burden on the programmer, because they presumably have more resources to commit to the problem.
'Shirking responsibility for creating good code (even if that code was "free") is not a trait I would admire in another developer.'
You don't understand. There is no "shirking" in the case of open source because no responsibility is implied. I understand full well when I use open source software I am dependant on the will of the developers to continue, and I have no power to force them to do anything. That is the extra burden I carry in return for being able to use the code and have rights to the source code without cost. I knowingly and willingly make that bargin - I do not want to hold the programmer's nose to the grind stone! That is the surest way to make them unhappy and abandon the project. Open source quality isn't bad in my experience, and if you don't like it there are plenty of commercial places you can obtain software on your terms.
Hmm. I really hope they do have thumbnail and bookmark support, and continue to add features. Xpdf is a nice renderer, but the interface IMHO is not exactly a nice one. If gpdf can become the full equal of Acrobat Reader I'll be one very happy camper.
A new GTK file selector. FINALLY. I can't wait to use the new one - the old one was one of the great warts of the free desktop world, IMHO.
But they have decided to remove the text entry box??? Eeep. I guess having the Ctrl-l shortcut to get one is OK (after all, it will most likely be geeks that want direct text on a file open) but thats one they need to document WELL.
On the whole though, it might be a good thing. I guess we'll have to wait and see. But text box or not, it can hardly be worse than the old one.
as the Gnome desktop itself is the fact he's using the freedesktop xserver to run it. I had no idea it was so far advanced.
"Spoken like a true loser/script kiddie/whatever -- completely unprofessional."
Here's a hint - not everyone is professional at everything they want to do. Professionals get paid.
"Software developers should take pride in their work, and shouldn't release things that they aren't proud to claim."
Again, a proper attitude for the commercial world - NOT the hobby world.
"And the "and if they relied on it oh well, their fault...". What kind of an attitude is this? Is this the basic open-source "it's free software, so don't complain if it doesn't work" attitude? Or is this guy just a bigger loser than most?"
Excuse me - why should you have any guarantee of quality if you haven't paid me for my work? If you use free software you accept as your part of the deal that the software is not bound by commercial standards. That may or may not mean anything - some free software is very, very good - but it does mean you have no right to complain. You PAY for the right to complain. Please feel free to buy commercial software and complain about it's quality AFTER YOU HAVE PAID FOR THE PRIVILAGE. But don't use someone's hard work without compensation and then complain about the quality of the work. You have no right.
"If this idiot's opinions are indicative of how the open source community views their work, then the world is right to avoid it as much as possible"
If the world has your attitude, I hope it does. The open source community seems to be doing just fine on its own.
Unless they all do it/get legislation passed requiring it for "national security purposes."
These people know to play the game.
Tried to email, got an inbox full message. If you do start a lobbying effort, please try to get an article on slashdot about it. That will give it enough publicity to get some real momentum going.
"Ballmer said governments that abandon Microsoft are more interested in making a political statement than using the best and most affordable software."
What constitutes the "best" software? Most features? Maybe. Best stability? Maybe. Best security? Maybe.
For some situations and groups, the best software is software the furthers goals like avoiding dependance on a foreign company with a bad track record for business practices and near monopoly control. Like, say, foreign governments. Which are, after all, political institutions. Why wouldn't they make political statements?
Microsoft gets it, all right. They will do their best to make decisions other than for immediate $$ spent look silly, but for some in this world there really is more to it than that. Microsoft knows to fear thinking like this, because it cannot be controlled.
"The people who are making business decisions based on where are the applications, what is the value, what is the lowest cost of ownership, we're not losing them."
At, but there again value and cost of ownership are not always strictly a matter of $$. Frankly, it's a pretty cold world when that is true, and it's one of the things I dislike about the US. In any case, to solve the chicken egg problem of applications first or users first, the users typically have to take the plunge.
Ballmer can chuckle all he likes. What he isn't mentioning is that first adapters always, ALWAYS, have a hard time. Did we make fun of the first people who bought those really expensive first generation DVD players? Do universities shrink away from paying Peoplesoft $$$$$$$$$ for rather unimpressive systems that still need lots of tweaking? (I'm still convinced if a couple of them had hired GNU enterprise with that $$$ everyone would have been better off, but that's another post.) Change is tough. But for each person or group that makes the change, things are ironed out and it gets easier next time around. And as things get easier, a proven track record emerges, and the trail is paved, more people start to go down it.
So sure, Munich is chopping down trees to make a road through the forest right now. But the next time around someone else will have an example to follow, and will also do some more road clearing.
I'm quite sure if Munich had made the decision to switch over to Macintosh, they'd be facing many of the same problems. To a certain extent change is just hard, period. But the thinking here is long term, not short term. The Media reports short term, Microsoft laughs in the short term. But I'm a lot more interested in the long term, when Munich can look at the next upgrade cycle prices for Windows and laugh in their face.
"maybe they'll say that they'll accept the terms of GPL until such time as it is declared invalid (effectively squirming out of this particular mess)."
Nope. Can't do that, or they'd be distributing the Linux kernel, including their own IP, under a GPL they have temporarily accepted as valid, and so their whole IBM/license thing falls apart. (Of course, IANAL. Reality may vary from logic.)
If they were going to try to break the GPL into the public domain, they should have tried that FIRST. I suspect this particular consequence of their rhetoric did not occur to them until now.
Another example of why pissing off the community you live in (in SCO's case, the Unix world) is a Bad Thing.
Absolute numbers are fine, but what about normalizing it for the total number of BSD, Linux, and Windows servers in use in this study? That's the more meaningful number. Then, what constitutes a successful attack?
Also, a useful study would look at how machines are maintained, password policies, etc.
Now before I come off sounding like a Linux apologist, it is quite possible there are some serious weaknesses that need to be addressed. If so, I hope they give us full info on the attacks so we can fix the problems. But these numbers as they stand don't tell us a darn thing.
If a dedicated admin configures Selinux and heavy duty firewalls, and puts Klingon password policies in place, I'd personally still be confident to match that system against anything out there. Default Redhat installs, on the other hand, are something else again. So again we need more info. It's all in how things are set up and maintained. The question actually being asked here - which OS is strongest, all other things being equal - is a really really tough one to answer. There are many other issues that must be addressed first.
So, as far as any useful information is concerned, this article doesn't appear to have any. What if the Linux machines simply had the best intrusion detection in place? (I'm not saying they did, but it's a fair question.) Need More Information!
I'm starting to realise that comparing Fresco to X is a fairly pointless exercise. Fresco is concerned first, last, and only with Getting It Right. This does not lend itself to a rapid practical implimentation.
I hesitate to define Fresco as a failure. It hasn't replaced X, true, and probably won't for a very long time, but it does do interesting and new things. That's always a gutsy thing to do, and always to be applauded. They have explored a lot of options and written some working code, which is much further than most such projects get. Let's consider it windowing system research first, and a project to replace X as a remote second. It makes more sense in that light.
Looking more closely at their design choices, at the moment. So, if your looking for an X replacement, Fresco is still a long way off.
Really, I think the main problem is too few developers. If IBM or somebody would sponsor its development it might move faster.
might very well be Mac on Linux, ironically enough. Check out this:
http://www.maconlinux.org/sshots/pic12.jpg
Running multiple versions of MacOS in parallel. Think about the possibilities for software developers. Having multiple environments immediately available for testing.
Then of course there's the ability to run all those Mac apps when needed and still have the Linux desktop to go to when they aren't needed.
Mac on Linux is what the open source world should try and create for the Windows world. Think of the possibilities if you could run Windows at work in a Window - be able to do all the windows specific stuff at need, but have Linux goodness in which to work as well. If a phb strolls in, just flip your desktop over to full screen windows. Then for the rest of the time go stealth with Xpde, good enough to fool a casual glance. Maybe some rootless window hack could even be figured out.
Of course, if your boss says you Must Use Windows, there's not a whole lot you can do. But perhaps this would be an acceptible compromise.
Oh, not my bad. Gah. Need to sleep.
whoops. scorned. sorry
"When people writing code for business are optimizing for speed and redundancy mainly in the parallel sense (i.e., a failsafe swap to a sister server), how RELEVANT is that to blocks of code written never, ever, ever, ever, ever to fail on tested but "outdated" hardware?"
Not sure. Well written code is not to be scorned, whatever else might have happened in computers - Maxima was written decades ago and much of that code is quite useful today. (And that isn't high quality NASA code either!) What I'm hoping for, however, is the systems they use to develop that software. "Perfect" software in the sense of security I'd argue is currently the #1 concern for the internet, because there are more than enough people willing to be the one to knock over the dominio stack. Whatever software NASA uses to develop and document their "perfect" code, I'd like to get my hands on it. Who knows, maybe it will help everyone.
By B I mean the B method:
http://vl.fmnet.info/b/
If this isn't a call to take a closer look at the possibility of more widely using tools like Z and B to develop important software, I don't know what is.
Yes, they're difficult. Yes, they aren't likely to eliminate all bugs. BUT. They provide a much better chance (as I understand it - I'm not an expert) that what is designed is what actually gets implimented. That shifts the burden onto the design, but that's OK - that burden was always there. It just means that the design gets properly implimented, which is all that can reasonably be asked of the coding process.
Currently, again as I understand it, the life of a software program in development is a constant struggle by the developers to cope with ever changing demands of customers. I think if people want matters to improve the customers are going to have to come to grips with reality, take the time to sit down and think things through, and make all critical design decisions BEFORE the development process begins. More expensive up front? You bet. That's why I think companies should look at cooperative effort for this type of thing. Distribute the cost of developing one really good program across an industry. A lot of the same core functionality can likely be shared between businesses - if they all pay for one proper design and implimentation of an open program up front, and they all get copies of the logic and proof code with rights to extend as they see fit, they all benefit. They can also open up the more general parts of the package to the world at large under GPL, and anyone could contribute who can generate valid B and Z designs/proofs. Sort of an "academic" open source code development forum - peer review and all. The companies get the benefit of all new development - if they are using it internally they can extend the GPL code for themselves, so long as they don't distribute it. If they do distribute it, they can so so under GPL for everyone to enhance. A plugin based model can also allow them to develop components to the system they can sell as commercial software, if they wish.
Whether this would work/appeal with corporate thinking I have no idea - many of those folks seem to view cooperation like the plague. But it might allow a higher grade of software to be developed and universally used, and I have a hard time imagining how that could be a bad thing for anyone.
"78. However, as is widely reported and as IBM executives knew, or should have known, a significant flaw of Linux is the inability and/or unwillingness of the Linux process manager, Linus Torvalds, to identify the intellectual property origins of contributed source code that comes in from those many different software developers. If source code is code copied from protected UNIX code, there is no way for Linus Torvalds to identify that fact."
Um. If source code was copied from protected UNIX code, how the @#$@%@# would Linus know about it? He doesn't have access to the protected source code - it's protected! The only way to know is if the owners of the protected source code make the claim and are able to back it up! How can Torvalds be faulted for not being clarvoyant? Do they mean identify it after the fact? AFAIK no one can say yet that the origins of code X can't be identified. SCO hasn't even let us TRY - they won't tell us what they want identified!
If what they are actually saying is that open source shouldn't be allowed to proceed simply because it doesn't have massive paperwork assigning every bit of code to some source, they've been hitting the crack again. Email archives, content management back trails anyone? And we can go further than that if we are really forced to - the FSF has been getting copyrights assigned to it for years just in case things come to that pass, and if it becomes utterly necessary that might become common practice.
What I'm hoping will come out of all this is a way that open source projects can set themselves up so that no one can sue them without them actually having done something wrong. (OK, OK - I know anyone can still bring the lawsuit. I mean create a situation where the project dispose of the suit in such a way that it doesn't cost the project or developers much of anything and discourages idiots like SCO from attempting it.) That would be useful, and if SCO is the start of a trend may become very necessary.
What's the incentive for Microsoft to stop their abuse? The abuse nets more money than any fine is likely to take away, and is the quickest way to make $$ back after the fine. This won't solve a darn thing.
The effective solutions (start multiple companies off with the Windows source code and have them compete, for example) are very radical, and I don't know if most of them are in the power of the EU. But if the US government is any example, the will to use them isn't there anyway, so Microsoft can write their check and go back to business as usual.
"Even if after everything we've heard from them
to date falls through, they may try to make
the claim in court that every OS in existence
is derived from SCO IP, and that being the
case Linux users STILL owe SCO money,
regardless of code."
Not gonna happen.
Yes. That's what I thought about some small company with silly claims suing IBM straight off for $3 billion, instead of picking a smaller fish to start with. Or about them taking the bluff to the point where they appear ready to sue on a scale the RIAA would envy. I'm hoping you're right, but nothing I have seen thus far indicates to me they wouldn't try it.
"No actually the details do matter. They matter in the court of law. Contrary to popular belief you can't just buy your way into winning every lawsuit."
The court of law isn't the battle SCO is worried about winning. They are trying to win the stock market game, get their name in front of the world, and hurt Linux/open source. The motivations for the latter are a matter of speculation, but clearly their actions and statements have that overtone. Their legal action is a long shot, and they must know that (barring incredible stupidity, which I suppose I can't rule out). So if they know that, what are their motivations? They aren't doing this for nothing.
"Yea, so what? MS wants linux destroyed as well and you how well that's worked for them."
Microsoft hasn't made ANY serious overt moves to destroy Linux. Not yet. They're still fighting regulatory stuff across the pond, and while in a practical sense the antitrust case in the US is dead they were ruled a monopoly. That's why so many people wonder if they are pulling SCO's puppet strings - they don't have to get down and dirty themselves, and thus avoid extra trouble. We're too complacient about Microsoft. SCO is the current issue, but once they're gone Microsoft won't just concede ground to Linux. They will fight, and do whatever it takes to win. If they can't win technically, they'll try to get the laws changed. The only thing I'm sure the WON'T do is nothing.
"Look, your just acting a bit hysterical here. Facts and law do matter. And contrary to your sky is falling ideas Linux isn't going to be destroyed by anyone."
Facts only matter in certain kinds of fights. Law matters, but it is slow and big money interests can lobby for new laws. I don't think SCO can stop Linux. But confining the analysis of this situation to SCO is using tunnel vision. Linux has and will have enemies, and they will learn from what SCO does. I don't have enough faith in the US political system to trust the lawmakers not to pass legislation that heavily favors commercial software, and if THAT happens SCO will be the least of our worries.
"You can't just dismiss that and say "well SCO will just think of some angle to spin this." My advice to you is relax a bit and ease off the coffee. "
I never said it wasn't important to the legal case. But the PR war may be lost by the time the courts get done with appeals processes. The legal system is slow - the software world moves at warp speed. DR-DOS won a decision against Microsoft, but by the time it did it was irrelevant.
"This case will resolve itself and there is no indication from any repudiable legal counsel or expert that SCO will be the end of Linux."
SCO may just be the beginning. Money and power defend themselves. Linux and open source threaten both. Ergo, we can expect trouble. It's as simple as that.
They might confess to past bad internal security in the interest of pursuing current licensing revenue. I'm not a legal person, so I don't know what all the various possibilities and ramifications are. My point is they aren't going to shut up just because they are in the wrong.
"How do you propose to fight them, if not by showing their claims to be incorrect?"
Essentially, in the specific case of SCO, starvation. SCO can't survive indefinitely if they don't get any revenue from anywhere, while paying their lawyers.
I'm not entirely convinced we will ever be rid of SCO, at least in spirit. The use of the legal system as a club against free products is something I have long dreaded, and I don't know any effective countermeasures short of changing the law itself.
I guess that would be my proposed way to fight - lobby for protection of free products against abuse of the legal system. I rather doubt a technical solution to a legal problem exists, so I suggest we look for legal solutions and defenses.
The letter to Congress from SCO a while back was the ultimate illustration of my concerns - there is nothing fundamentally that prevents Congress from saying open source software is illegal, because of security risk|unfair competition|what have you. When politically powerful corporations finally find out they can't FUD us to death or out compete us, they will try to use their political clout to get that accomplished. Never doubt it. And since open source folk don't mean much to re-election minded officials, I'd say our position is not the best.
SCO may be just the tip of the iceberg, a harbringer of the real war. We need to keep that in mind, so if that day comes we are mentally ready to meet it.
If it's true, can all the /. folk who raised this possibility when the news broke claim part of the reward from SCO?