The GPL does not take away any freedom, it only grants it. But it places some restrictions on that grant.
Under copyright law, the only thing you may not do with a copyrighted work or derived work is redistribute it. GPL, in addition to all the freedoms you automatically get from copyright law, also gives you the freedom to redistribute the original or derived work, even charging a fee if you want. It merely places the limitation on this freedom that you must give others the same freedom.
It's sort of like the fundamental rights to life, liberty and property - if you don't respect those of others, you forfeit yours.
Maybe I'm misparsing your message, but it sounds like you are saying that Michael Tiemann's comments about infrastructure vs. applications are a very bad thing. However, you work for a proprietary software company yourself (not that this is a shameful thing, I do too - but it seems unfair to hold Cygnus to a higher standard).
As for Red Hat being controlled by outside investors, I think the founders are being _very_ careful to sell only small minority stakes. Both RH and the investors are looking at these investments as more meaningful in strategic and political terms than in strictly financial terms.
Incidentally, I think the Cygnus pendulum is swinging back towards free software somewhat - I remember there were a few years when they did their best to expunge their web site and marketing materials of any reference to the fact that their code was free. Now they play up the open source angle in almost every press release.
But the whole internet angle is only a new twist on a problem with a long history, which is that women who are raped cannot effectively press charges unless they are basically the purest most chaste thing on the planet. Any hint that a woman ever actually wanted sex in her life, and this is the way the whole damn justice system will treat her.
It is sad that these kinds of backwards attitudes are still around.
Free/Open GNU/Linux has lots of good spokespeople!
on
ESR Wants to Retire
·
· Score: 2
I think he admitted in this essay that his judgement is not as sharp as it once was, given the job he has to do and the stresses that come with it.
It's true that he's responded to some things that may have been well-considered differences of opinion as personal attacks. But he's also taken a lot of flak that was just mean-spirited personal attacks, before he did anything that deserved much disagreement. I know things like that can get to wear on you - after a while it gets hard to tell the difference.
Think about what his essay said - do you think you could do his job better? Free software would have probably still broken through without him at some point, but would it have done it as fast or as thoroughly as it has.
Although direct democracy moderation might sound like a good idea in theory, it would probably not work well in practice - imagine a KDE vs. GNOME flamewar in an article's comments if everyone were able to moderate. The scores would have as bad a signal-to-noise ratio as the comments. So a specific group of moderators provides a useful buffer.
But moderators must have accountability too, and if there are a lot of them, Rob will have a hard time managing this himself.
So how about this suggestion:
* Make article scores floating point.
* Give each moderator a weight, which is the amount by which he can change an article's score when moderating. Initialize all moderators to 1.0 and make 1.0 the maximum (or pick other suitable parameters).
* Let everyone vote on whether or not they like the way an article was scored. This would feed back to the weight of the moderators who moderated it towards it's current score.
Thus, ultimately moderators who consistently score articles up or down for bad reasons will have their weight lowered until their moderation does not really affect anything.
This sort of applies the principles of representative democracy to moderation: moderation will be insulated from the momentary whims of the masses, but ultimately in a long-term sense, moderation is under the control of the entire community.
GMC is going to get rewritten. It does have some internal problems.
gnome-terminal (and the zvt library it is based on) works fine. People have complained about it but I have not seen a substantial explanation of the problem yet.
You mean that the FSF created the C library? Wow, and all this time I thought it was the ANSI C committee who formalized that.
I bet a pretty small proportion of the GNU C library is an implementation of the ANSI standard. A lot of the other stuff is POSIX, BSD, SVR4 and GNU extensions, various "traditional" Unix stuff, I18N support etc. If you think writing all that stuff to be thread-safe, efficient, multi-architecture, multi-platform, and conformant to more standards than you can shake a stick at is easy, you're wrong.
Stallman's GPL is hurting programmers now.
on
Feature:Free Linux
·
· Score: 1
"It absolutely does prevent those things. The best a programmer can do, once he or she has released a product under the GPL, is make money doing peripheral tasks such as making CD-ROMs or consulting. He or she cannot make money from the software itself; it's available for free!"
Nothing in the GPL prevents you from also offering your code under other licenses at the same time (assuming all the GPL'd code is yours). If it's not all yours, you have no business whining that you have the way to license your code anyway you want, so how dare these other people license their code in any way they want, such that if you use it you must follow certain conditions! Why, the very gall!
We're seeing this now. Be, a would-be challenger to Microsoft, can't get a foothold; vendors are looking at installing Linux instead.
Be seems to think they benefit from Linux, since by gaining so much market share mindshare for a non-Microsoft product, it has opened the technology world to the possibility that non-Microsoft products might be better for certain tasks. I agree to some degree; I think at least that their odds would not be any better if Linux never existed.
But even if they are wrong and you're right, let's say FreeBSD (supported primarily by admitted CD-ROM foundry Walnut Creek) had been the one to make it big instead of Linux. Would that make Be's position any stronger?
Of course, you can pretend that this isn't happening if you'd like, but you'll only be ignoring the mounting evidence.
I personally don't mind that this is happening. Really, what you are arguing is that a critical mass of GPL software will doom most proprietary software to oblivion. The only way this could be stopped, according to your theory as I understand it, is for the mass of people working on GPL software to give up and go home. That's not going to happen. So businesses today need to deal with the market reality that GPL code is out there, and find a business model that deals with that fact. Asking GPL programmers to just go home, or to relicense their code under X-like licenses, is not going to change anything.
Not GNU utilities, Linux/GNU utilities ;)
on
Feature:Free Linux
·
· Score: 1
Popular though Linux is today, I bet there are still more total non-Linux installations of GNU utilities and tools than total Linux ones. Pretty much every Unix shop has GNU stuff installed. A very high proportion of embedded developers use GNU tools.
In fact, I think one of the biggest technical wins of Linux in the early days was that you didn't have to install all the GNU utilities and tools on it by hand. Every time I have to deal with an arbitrary limit, or type "gzip -dc somefile.tar.gz | tar xvf -" instead of "tar zxvf somefile.tar.gz", or deal with multiple differently broken vendor C++ compilers in a multiplatform environment, I am reminded of this.
Sorry, I get sick of people flaming about the same points over and over again when they are untrue. Since when has courtesy been mandatory at Slashdot High School?:-)
The FSF takes the position that programming examples in manuals are covered by the fair use exception and thus a specific exception for them would be redundant. I'm not sure if I agree with this, I think an explicit exception would be good. But in any case, they don't GPL their docs and don't even particularly try to encourage others to use any specific documentation license, AFAIK.
The difference between an idea and an expression of an idea is in fact fairly clear as far as copyright law is concerned. If I hold copyright on a paper about a particular scientific theory, that gives me no rights whatsoever in the theory itself; I may not preclude others from applying, writing about, or extending my theory.
When the distinction between an idea and expression of an idea is unclear to the point that the idea could not be expressed without effectively copying a copyrighted text, the copyright holder has no standing (at least in US court) to sue over even a literal copy.
So there is no way someone could copyright the concept of socket programming in Perl.
The only form of intellectual property protection that covers an idea rather than an expression is patent law, which the FSF opposes applying to the software domain at all. So no fear of finding patented manuals for FSF software either.
gazillions of free compilers - most of which suck
on
Feature:Free Linux
·
· Score: 1
Yes, but for the most part they all suck - not very portable, poor support for even one language standard let alone many, rather crappy optimizations, etc. If you have ever worked in a Unix or embedded software shop, you'd know why the world's proprietary compiler vendors all consider GCC their biggest threat.
GCC was designed from the start to support multiple target architectures, multiple hosts, and a wide array of optimizations. Designing a framework for an optimizing compiler so that you can compile something from the same code base that will optimize well for many architectures, and can be extended in the future to architectures and languages totally different from anything around when you started, is very impressive.
Trust me, extending the average undergrad project compiler to do good optimization and output code for multiple targets would not really be feasible.
RMS may foam at the mouth in person sometimes, but at least his writings are well-reasoned and level-headed - you have to admit that much at least even if you don't agree with his premises. Everything I've seen of TC's public writings has been vicious flamage. Now, maybe that's not a representative sample, but that's all I've personally personally seen.
Actually, the name of the Solaris kernel is SunOS. The name of the operating system as a whole is Solaris. You don't see it called by just the kernel name any more, do you. Hmmm.
Not even any of the free BSD distributions are 100% free of FSF software. Removing the GNU stuff would take a damn lot of work in addition to being pointless and counterproductive. Heck, the first thing I do on a BSD box is install the GNU utilities, just like on any other non-Linux Unix system.
As a separate point, the FSF is responsible for the exsitence and/or freedom of a lot of code they don't hold copyright to. For instance, everything owned by the X Consortium would not have been free if it weren't for the work of the FSF. The same goes for many other entries on that list.
And finally, shouldn't that be the Daemon Penguin? The BSD mascot is not a "Demon".
It's the Fed's fault? Nobody put a gun to people's heads and said "speculate!", just as no-one's saying "do really bad derivatives deals!" today;
People are greedy by nature. Regulation attempts to legislate against the greed. A free market makes greed naturally self-regulating. Considering this particular example: investors naturally want to maximize their profits. A free market in credit, however, naturally limits the damage high-risk but potentially high-yield investments can do to the economy as a whole. Interest rates, the price of the credit needed to speculate on other people's money, and incidentally also the return on lower-risk investments such as bonds, will naturally and gradually rise in such situations, making speculation less profitable relative to safer investments.
The Fed destroyed the natural information signals like this that propagate through the economy by keeping interest rates too low (incidentally in the course of trying to help England prop up its currency at an unsustainably high level). That the ultimate result of an economy having key feedback paths destroyed this way was disaster is unsurprising. The Fed does a better job of regulating interest rates more intelligently today, but the same could happen again.
Think of it as negative feedback cycles, the same way the human body regulates itself. Do you really think your body would run better if some central agent had to control all aspects of its metabolism?
a corrupted super-free market wouldn't be any better than what Russia has now.
To me, a fully free market specifically means a 100% separation of economics and state, i.e. that the government has no power, as established constitutionally, to affect influence or regulate the economy. Under such a system there is no meaningful influence for gangsters to buy or sell. Of course, with enough systematic corruption, constitutional limitations can be made irrelevant. Eternal vigilance is the price of liberty.
The US economy is only as good as the consensus opinion of it in Bed-Stuy and Appalachia; my opinion of Eastern Bloc economies would be the same, if I knew where to look. Please don't give me travelogues or macroeconomic statistics - I want to know the opinions of those at the receiving end of what's being dished out. I don't care about some amorphous "they".
My personal relatives appeared to be materially better off. People in general seem to be doing a much better job of creating their own opportunities instead of waiting for the state to take care of them, and that to me is even more important. The very worst off people are probably doing quite poorly, but that was just as true under Communism. However, I don't agree with your premise that the success of an economy is best measured by the material conditions of the worst off. Any socioeconomic system will have _some_ people in it who are suffering. This cannot be changed, and attempts to do so are Quixotic. The best you can do is maximize everyone's opportunity to improve his or her own condition.
I can definitely say that for the average person, Poland has improved a great deal under "shock treatment". You don't have to wait for hours in line to get bread any more (literally - many people from better off countries can't believe this, but I experienced it). You don't have monthly rations for how much meat and sugar you can buy. You are free to start your own small business, and many people have. You are free to change jobs when you want, not when the state permits you. Of course, a lot of people can't get used to the idea that they are responsible for their own economic welfare.
I can't explain this in detail in the small space of a slashdot post, but in brief, the economy is full of long-term private investors, and a large number of long-term investors seeking to maximize profits will regulate interest rates more effectively through market forces than any central planner, because no central planner can have possibly enough knowledge of the whole economy.
Alan Greenspan is perhaps the most effective central banker this or any other country has ever seen, and he believes his job should be eliminated because no one can regulate as well as a free market.
First of all, I am not a conservative by any stretch of the imagination. If you read my post you would know that.
Second, I presented some facts supporting the theory that the great depression was caused by government intervention rather than the free market. But instead of actually presenting contrary arguments, you chose to attack my use of the phrase "revisionism" to describe the other theory. Which one of us is "cacnelling out any chane of a rational argument"?
Wether this effort is sincere or not, I bet even the very rumors of it will spur many other application ports to Linux. Imagine software engineers everywhere saying to their bosses "Microsoft has a Linux strategy, why don't we?"
Eric has been part of the hacker community for a long time, he was part of it before Linux even existed. I think the reason so many people resent him now is that he's trying to act as the salesman/marketeer of the hacker culture, and most hackers instinctively hate salesmen and marketing.
I've been suspicious of the idea myself, but I think it's naiive to believe that free software can take over the world just by being better, without any sales or marketing. Look at how many inferior products have come to dominate their industry segments (x86 architecture, PC system architecture, Microsoft anything) by force of superior marketing. And if we're going to have marketing than it should be led by one of our own.
In fact, perhaps it's best that the marketing be done by someone who's known more for his analysis of the community on a social level than for hacking code.
Calling Eric a suit is just too out there to even think about.
Slashdot Mirror
The GPL does not take away any freedom, it only grants it. But it places some restrictions on that grant.
Under copyright law, the only thing you may not do with a copyrighted work or derived work is redistribute it. GPL, in addition to all the freedoms you automatically get from copyright law, also gives you the freedom to redistribute the original or derived work, even charging a fee if you want. It merely places the limitation on this freedom that you must give others the same freedom.
It's sort of like the fundamental rights to life, liberty and property - if you don't respect those of others, you forfeit yours.
Actually, last I checked, Perl's memory management was jut simple refcounting, not even smart enough to avoid leaking circular garbage.
Maybe I'm misparsing your message, but it sounds like you are saying that Michael Tiemann's comments about infrastructure vs. applications are a very bad thing. However, you work for a proprietary software company yourself (not that this is a shameful thing, I do too - but it seems unfair to hold Cygnus to a higher standard).
As for Red Hat being controlled by outside investors, I think the founders are being _very_ careful to sell only small minority stakes. Both RH and the investors are looking at these investments as more meaningful in strategic and political terms than in strictly financial terms.
Incidentally, I think the Cygnus pendulum is swinging back towards free software somewhat - I remember there were a few years when they did their best to expunge their web site and marketing materials of any reference to the fact that their code was free. Now they play up the open source angle in almost every press release.
This detective was obviously an asshole.
But the whole internet angle is only a new twist on a problem with a long history, which is that women who are raped cannot effectively press charges unless they are basically the purest most chaste thing on the planet. Any hint that a woman ever actually wanted sex in her life, and this is the way the whole damn justice system will treat her.
It is sad that these kinds of backwards attitudes are still around.
I think he admitted in this essay that his judgement is not as sharp as it once was, given the job he has to do and the stresses that come with it.
It's true that he's responded to some things that may have been well-considered differences of opinion as personal attacks. But he's also taken a lot of flak that was just mean-spirited personal attacks, before he did anything that deserved much disagreement. I know things like that can get to wear on you - after a while it gets hard to tell the difference.
Think about what his essay said - do you think you could do his job better? Free software would have probably still broken through without him at some point, but would it have done it as fast or as thoroughly as it has.
Although direct democracy moderation might sound like a good idea in theory, it would probably not work well in practice - imagine a KDE vs. GNOME flamewar in an article's comments if everyone were able to moderate. The scores would have as bad a signal-to-noise ratio as the comments. So a specific group of moderators provides a useful buffer.
But moderators must have accountability too, and if there are a lot of them, Rob will have a hard time managing this himself.
So how about this suggestion:
* Make article scores floating point.
* Give each moderator a weight, which is the amount by which he can change an article's score
when moderating. Initialize all moderators to 1.0 and make 1.0 the maximum (or pick other suitable parameters).
* Let everyone vote on whether or not they like the way an article was scored. This would feed back to the weight of the moderators who moderated it towards it's current score.
Thus, ultimately moderators who consistently score articles up or down for bad reasons will have their weight lowered until their moderation does not really affect anything.
This sort of applies the principles of representative democracy to moderation: moderation will be insulated from the momentary whims of the masses, but ultimately in a long-term sense, moderation is under the control of the entire community.
GMC is going to get rewritten. It does have some internal problems.
gnome-terminal (and the zvt library it is based on) works fine. People have complained about it but I have not seen a substantial explanation of the problem yet.
If the projects did merge, I think GDE would be a much better name than KNOME. Blech, KNOME just looks ugly.
You mean that the FSF created the C library? Wow, and all this time I thought it was the ANSI C committee who formalized that.
I bet a pretty small proportion of the GNU C library is an implementation of the ANSI standard. A lot of the other stuff is POSIX, BSD, SVR4 and GNU extensions, various "traditional" Unix stuff, I18N support etc. If you think writing all that stuff to be thread-safe, efficient, multi-architecture, multi-platform, and conformant to more standards than you can shake a stick at is easy, you're wrong.
"It absolutely does prevent those things. The best a programmer can do, once he or she has released a product under the GPL, is make money doing peripheral tasks such as making CD-ROMs or consulting. He or she cannot make money from the software itself; it's available for free!"
Nothing in the GPL prevents you from also offering your code under other licenses at the same time (assuming all the GPL'd code is yours). If it's not all yours, you have no business whining that you have the way to license your code anyway you want, so how dare these other people license their code in any way they want, such that if you use it you must follow certain conditions! Why, the very gall!
We're seeing this now. Be, a would-be challenger to Microsoft, can't get a foothold; vendors are looking at installing Linux instead.
Be seems to think they benefit from Linux, since by gaining so much market share mindshare for a non-Microsoft product, it has opened the technology world to the possibility that non-Microsoft products might be better for certain tasks. I agree to some degree; I think at least that their odds would not be any better if Linux never existed.
But even if they are wrong and you're right, let's say FreeBSD (supported primarily by admitted CD-ROM foundry Walnut Creek) had been the one to make it big instead of Linux. Would that make Be's position any stronger?
Of course, you can pretend that this isn't happening if you'd like, but you'll only be ignoring the mounting evidence.
I personally don't mind that this is happening. Really, what you are arguing is that a critical mass of GPL software will doom most proprietary software to oblivion. The only way this could be stopped, according to your theory as I understand it, is for the mass of people working on GPL software to give up and go home. That's not going to happen. So businesses today need to deal with the market reality that GPL code is out there, and find a business model that deals with that fact. Asking GPL programmers to just go home, or to relicense their code under X-like licenses, is not going to change anything.
Popular though Linux is today, I bet there are still more total non-Linux installations of GNU utilities and tools than total Linux ones. Pretty much every Unix shop has GNU stuff installed. A very high proportion of embedded developers use GNU tools.
In fact, I think one of the biggest technical wins of Linux in the early days was that you didn't have to install all the GNU utilities and tools on it by hand. Every time I have to deal with an arbitrary limit, or type "gzip -dc somefile.tar.gz | tar xvf -" instead of "tar zxvf somefile.tar.gz", or deal with multiple differently broken vendor C++ compilers in a multiplatform environment, I am reminded of this.
Sorry, I get sick of people flaming about the same points over and over again when they are untrue. Since when has courtesy been mandatory at Slashdot High School? :-)
The FSF takes the position that programming examples in manuals are covered by the fair use exception and thus a specific exception for them would be redundant. I'm not sure if I agree with this, I think an explicit exception would be good. But in any case, they don't GPL their docs and don't even particularly try to encourage others to use any specific documentation license, AFAIK.
The difference between an idea and an expression of an idea is in fact fairly clear as far as copyright law is concerned. If I hold copyright on a paper about a particular scientific theory, that gives me no rights whatsoever in the theory itself; I may not preclude others from applying, writing about, or extending my theory.
When the distinction between an idea and expression of an idea is unclear to the point that the idea could not be expressed without effectively copying a copyrighted text, the copyright holder has no standing (at least in US court) to sue over even a literal copy.
So there is no way someone could copyright the concept of socket programming in Perl.
The only form of intellectual property protection that covers an idea rather than an expression is patent law, which the FSF opposes applying to the software domain at all. So no fear of finding patented manuals for FSF software either.
Yes, but for the most part they all suck - not very portable, poor support for even one language standard let alone many, rather crappy optimizations, etc. If you have ever worked in a Unix or embedded software shop, you'd know why the world's proprietary compiler vendors all consider GCC their biggest threat.
GCC was designed from the start to support multiple target architectures, multiple hosts, and a wide array of optimizations. Designing a framework for an optimizing compiler so that you can compile something from the same code base that will optimize well for many architectures, and can be extended in the future to architectures and languages totally different from anything around when you started, is very impressive.
Trust me, extending the average undergrad project compiler to do good optimization and output code for multiple targets would not really be feasible.
Interesting. Perhaps that would be the reason the FSF does not use GPL for it's documentation, but rather a different, much simpler license.
Perhaps that is also why copyright covers expressions of ideas rather than ideas themselves.
Clues are good things.
RMS may foam at the mouth in person sometimes, but at least his writings are well-reasoned and level-headed - you have to admit that much at least even if you don't agree with his premises. Everything I've seen of TC's public writings has been vicious flamage. Now, maybe that's not a representative sample, but that's all I've personally personally seen.
Actually, the name of the Solaris kernel is SunOS. The name of the operating system as a whole is Solaris. You don't see it called by just the kernel name any more, do you. Hmmm.
Not even any of the free BSD distributions are 100% free of FSF software. Removing the GNU stuff would take a damn lot of work in addition to being pointless and counterproductive. Heck, the first thing I do on a BSD box is install the GNU utilities, just like on any other non-Linux Unix system.
As a separate point, the FSF is responsible for the exsitence and/or freedom of a lot of code they don't hold copyright to. For instance, everything owned by the X Consortium would not have been free if it weren't for the work of the FSF. The same goes for many other entries on that list.
And finally, shouldn't that be the Daemon Penguin? The BSD mascot is not a "Demon".
New proprietary compiler? Where'd you hear rumors like that?
It's the Fed's fault? Nobody put a gun to people's heads and said "speculate!", just as no-one's saying "do really bad derivatives deals!" today;
People are greedy by nature. Regulation attempts to legislate against the greed. A free market makes greed naturally self-regulating. Considering this particular example: investors naturally want to maximize their profits. A free market in credit, however, naturally limits the damage high-risk but potentially high-yield investments can do to the economy as a whole. Interest rates, the price of the credit needed to speculate on other people's money, and incidentally also the return on lower-risk investments such as bonds, will naturally and gradually rise in such situations, making speculation less profitable relative to safer investments.
The Fed destroyed the natural information signals like this that propagate through the economy by keeping interest rates too low (incidentally in the course of trying to help England prop up its currency at an unsustainably high level). That the ultimate result of an economy having key feedback paths destroyed this way was disaster is unsurprising. The Fed does a better job of regulating interest rates more intelligently today, but the same could happen again.
Think of it as negative feedback cycles, the same way the human body regulates itself. Do you really think your body would run better if some central agent had to control all aspects of its metabolism?
a corrupted super-free market wouldn't be any better than what Russia has now.
To me, a fully free market specifically means a 100% separation of economics and state, i.e. that the government has no power, as established constitutionally, to affect influence or regulate the economy. Under such a system there is no meaningful influence for gangsters to buy or sell. Of course, with enough systematic corruption, constitutional limitations can be made irrelevant. Eternal vigilance is the price of liberty.
The US economy is only as good as the consensus opinion of it in Bed-Stuy and Appalachia; my opinion of Eastern Bloc economies would be the same, if I knew where to look. Please don't give me travelogues or macroeconomic statistics - I want to know the opinions of those at the receiving end of what's being dished out. I don't care about some amorphous "they".
My personal relatives appeared to be materially better off. People in general seem to be doing a much better job of creating their own opportunities instead of waiting for the state to take care of them, and that to me is even more important. The very worst off people are probably doing quite poorly, but that was just as true under Communism. However, I don't agree with your premise that the success of an economy is best measured by the material conditions of the worst off. Any socioeconomic system will have _some_ people in it who are suffering. This cannot be changed, and attempts to do so are Quixotic. The best you can do is maximize everyone's opportunity to improve his or her own condition.
I can definitely say that for the average person, Poland has improved a great deal under "shock treatment". You don't have to wait for hours in line to get bread any more (literally - many people from better off countries can't believe this, but I experienced it). You don't have monthly rations for how much meat and sugar you can buy. You are free to start your own small business, and many people have. You are free to change jobs when you want, not when the state permits you. Of course, a lot of people can't get used to the idea that they are responsible for their own economic welfare.I can't explain this in detail in the small space
of a slashdot post, but in brief, the economy is full of long-term private investors, and a large number of long-term investors seeking to maximize profits will regulate interest rates more effectively through market forces than any central planner, because no central planner can have possibly enough knowledge of the whole economy.
Alan Greenspan is perhaps the most effective central banker this or any other country has ever seen, and he believes his job should be eliminated because no one can regulate as well as a free market.
First of all, I am not a conservative by any stretch of the imagination. If you read my post you would know that.
Second, I presented some facts supporting the theory that the great depression was caused by government intervention rather than the free market. But instead of actually presenting contrary arguments, you chose to attack my use of the phrase "revisionism" to describe the other theory. Which one of us is "cacnelling out any chane of a rational argument"?
The effect that Microsoft hopes to see and the effect it _will_ see are two very different things.
Note however, that most third-party software out there is _not_ an office suite. Why would that logic apply to such software.
Wether this effort is sincere or not, I bet even the very rumors of it will spur many other application ports to Linux. Imagine software engineers everywhere saying to their bosses "Microsoft has a Linux strategy, why don't we?"
Eric has been part of the hacker community for a long time, he was part of it before Linux even existed. I think the reason so many people resent him now is that he's trying to act as the salesman/marketeer of the hacker culture, and most hackers instinctively hate salesmen and marketing.
I've been suspicious of the idea myself, but I think it's naiive to believe that free software can take over the world just by being better, without any sales or marketing. Look at how many inferior products have come to dominate their industry segments (x86 architecture, PC system architecture, Microsoft anything) by force of superior marketing. And if we're going to have marketing than it should be led by one of our own.
In fact, perhaps it's best that the marketing be done by someone who's known more for his analysis of the community on a social level than for hacking code.
Calling Eric a suit is just too out there to even think about.