IMHO, saying (or implying) that the existence of many different Linux distributions is going to inevitably lead to fragmentation is FUD. And (again IMHO), this is what SCO's statement did --- or, at least, I think that this is how that statement would be read, and was meant to be read, by a vast majority of the intended audience.
Just because a statement is not pro-linux doesn't mean it's FUD. I wish people would start acknowledging some of the more serious problems with Linux instead of crying "FUD" all the time.
I agree. However, I don't think that this comes anywhere close to being an example of such behavior. The fact that some producers of non-free binary-only apps choose to distribute backwards-compatible versions for the convenience of their users is, at worst, a minor inconvenience for the producers of non-free binary-only apps. It certainly doesn't imply anything about the unity or disunity of the platform.
In the past, SCO and its representatives has made a number of statements about Linux (and free software in general) that many of us saw as FUD. In the most infamous example, these statements included:
"Linux at this moment can be considered more a play thing for IT students rather than a serious operating system..."
"The future of Linux is very uncertain... As there are such a large number of developers it is virtually impossible to predict what form Linux will take thus putting the future security of your business at risk."
"Currently there are over forty distributions of Linux... and as a result there is no single standard. Potentially, this means that software written for one system will not work on another."
Statements like these damaged SCO's credibility among the community that it now appears to be trying to embrace.
Do you/SCO still stand by these statements and opinions? If not, what changed your mind? Do you still assert that these statements were true when they were being made by SCO representatives --- or, in retrospect, do you admit that it was not accurate, but was just marketing FUD?
Well, the way I read the GPL, they could theoretically make the source available only in hardcopy mailed upon request and still be GPL compliant.
Those kind of tricks don't work. Part 3a of the GPL specifies that if you distribute a GPLed program in a binary format, you have to "Accompany it with the complete corresponding machine-readable source code, which must be distributed... on a medium customarily used for software interchange". I don't think that anyone could credibly argue that hard copy would meet those requirements.
Section 3 goes to say that the source code has to be in "the preferred for of the work for making modifications to it". This rules out distributing shrouded or otherwise-obfuscated sources.
Also...and I'm being totally serious here, what part of the GPL says it can't be linked with proprietary componets? I can't seem to find that in the Terms of Use....
It is in condition #2 of the GPL, which reads (in part): "These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it."
This is the (in)famous "viral" part of the GPL. Some people think it is a bug; I see it as a feature.
The combination of autoconf and automake is definitely the way to go.
The autoconf and automake docs can be a bit obscure; most people seem to learn how to use them by studying the configure.in and Makefile.am files from other projects.
There is also a brief discussion of autoconf and automake in Havoc Pennington's Gtk+/Gnome Application Development that might help you get started. This fine book is available on-line: here is the relevant section.
Actually, we should probably have a topical subgrouping, sort of like usenet. For instance, I think it'd be nice to know that linux stuff is under *.linux.comp and microsoft stuff under ms.comp.
OK, I'll bite --- where would Slashdot go in a usenet-style hierarchy? comp.linux.slashdot? sci.nanotechnology.slashdot? misc.legal.patents.slashdot? news.nerds.slashdot? soc.people.natalie-portman.slashdot?
This would lead to the bane of Usenet: massive cross-posting (or cross-URLing, in this case). Instead of everyone clamoring to get a.com, they would clamor to get their address into as many groups as possible.
We really need to link the namespace to geographical distribution - like the.us domain hierarchy.
I hear this argument from time to time, and I've never really understood it. It seems to me that trying to superimpose geography onto the net is both unnecessary and wrong-headed. The net is inherently a non-physical place: you never know quite where you are, and it doesn't matter.
How fine would the geographical gradiations be: By state? By city? The problems with this seem insurmountable. One example: If my company is in New York, but if my servers are located in New Jersey, are they at www.foo.com.ny.us or www.foo.com.nj.us? In this case, it seems like either answer is the wrong one.
In any case, this still wouldn't solve the inherent problem of domain name disputes, unless you used very fine geographic specifications and also added some sort of business type designation. Does anyone really want to type www.jadegarden.restaurant.chinese.upper-west-side. manhattan.nyc.ny.us into the location line of their browser?
On a minor note could someone create say an obfusicated version of the source code for release and then release the code so that it would take just a little longer to get ahold of?
No. Term #3 of the GPL specifies that "the source code for a work means the preferred form of the work for making modifications to it." So you are free to write your own obfuscated code by hand, but passing it through some obfuscation program is not allowed.
There is a general sense that, besides trying to increase shareholder value, Red Hat and VA are giving back to the community by employing GNOME hackers, kernel hackers, etc.
Red Hat and VA benefit from doing this, of course, but only in the sense that it benefits the free software community as a whole.
What sort of things are Corel doing along these lines? Have you hired any free software celebrities and given them the mandate to hack on anything want? What non-Corel development projects are you funding? Besides bigger and better graphical installers, what benefits will the average non-Corel Debian user derive from your involvement?
Given how much interesting real stuff is going on in science all of the time, you'd think that Wired would have no trouble finding stories to write. Instead, they seem to have consciously decided to focus on "fringe" stuff (cold fusion, anti-gravity, etc.), publishing long, credulous pieces that parrot various crackpot's claims that they are being "oppressed by the scientific establishment".
Now this article is nowhere near as bad as some of the stuff they've put out in the past, but it seems like further evidence of something I've long suspected: they are using science writers who both don't know and don't like science. Clearly it takes some goofy X-files type stuff to get their attention.
The irony is that the demographic reading Wired is probably much more interested in science, and in a much better position to understand and appreciate non-stupid science coverage, than your average readers.
However, given that the best money buys the best lawyer
Actually, legal positions at organizations like the ACLU and the Southern Poverty Law Center are incredibly difficult to get. The competition among law students for these $10/hour jobs is fierce... so it isn't just megacorporations that have access to the best legal talent.
I often hear people express the sentiment that technology changes too fast to be properly regulated, but I've never heard any evidence for this that I've found convincing.
I agree that bad laws can be rendered obsolete by increases in technology. The classic example is export controls that make it illegal to export processors above a certain fixed number of MIPS, or crypto with more than a fixed number of bits. But the problem here is not with the concept of regulation; the problem is that these laws are written with the assumption that the environment is static, and this problem is not just limited to regulations related to technology. These export controls are no more stupid than the US campaign finance laws, in which the maximum donations are not indexed for inflation.
Look at anti-trust law --- it is over 100 years old and, despite the protestations of those with a vested interest, it still seems very well suited to the current situation.
The "technology moves too fast to be regulated" argument seems to me to just be a rather poorly thought out rationalization for a libertarianism. (Which isn't a terrible thing, IMHO, but certainly much better arguments can made for it than this.)
Can anyone produce a non-specious example of some part of technology that is really inherently beyond any regulation, even if that regulation was to be crafted by knowledgeable hackers?
Errr... even though a full-blown anti-trust action against IBM never took place, IBM *did* operate under a consent decree from 1956 through 1996 that limited its ability to behave monopolistically in the hardware market.
It is impossible to predict the exact impact that it had, but the growth of the computer hardware industry in the 1960's might have been significantly stifled if the Feds had given IBM the "freedom to innovate".
To make the leap from "federal anti-trust action" (which has an approx 100-year history, and hasn't managed to utterly crushed the American Way Of Life(tm) in all of that time) to "bureaucrats dictating what code to write, who you can it be sold to (sic)" is just absurd. You might as well argue that a ban on assault rifles will lead to Green Berets kicking down your door to impound your pocket knife...
I can confirm this. I just talked to E*Trade and was told that, due to a big screw-up, everybody in the DSP was sent this notice, but that the DSP shares have *not* been allocated yet.
What a fiasco this has become. Hopefully VA will steer clear of E*Trade when they do their IPO. (But, despite all this, it sure would be nice if they did a community-DSP program too.:-)
Slashdot Mirror
IMHO, saying (or implying) that the existence of many different Linux distributions is going to inevitably lead to fragmentation is FUD. And (again IMHO), this is what SCO's statement did --- or, at least, I think that this is how that statement would be read, and was meant to be read, by a vast majority of the intended audience.
Just because a statement is not pro-linux doesn't mean it's FUD. I wish people would start acknowledging some of the more serious problems with Linux instead of crying "FUD" all the time.
I agree. However, I don't think that this comes anywhere close to being an example of such behavior. The fact that some producers of non-free binary-only apps choose to distribute backwards-compatible versions for the convenience of their users is, at worst, a minor inconvenience for the producers of non-free binary-only apps. It certainly doesn't imply anything about the unity or disunity of the platform.
In the past, SCO and its representatives has made a number of statements about Linux (and free software in general) that many of us saw as FUD. In the most infamous example, these statements included:
- "Linux at this moment can be considered more a play thing for IT students rather than a serious operating system..."
- "The future of Linux is very uncertain... As there are such a large number of developers it is virtually impossible to predict what form Linux will take thus putting the future security of your business at risk."
- "Currently there are over forty distributions of Linux... and as a result there is no single standard. Potentially, this means that software written for one system will not work on another."
Statements like these damaged SCO's credibility among the community that it now appears to be trying to embrace.Do you/SCO still stand by these statements and opinions? If not, what changed your mind? Do you still assert that these statements were true when they were being made by SCO representatives --- or, in retrospect, do you admit that it was not accurate, but was just marketing FUD?
Well, the way I read the GPL, they could theoretically make the source available only in hardcopy mailed upon request and still be GPL compliant.
Those kind of tricks don't work. Part 3a of the GPL specifies that if you distribute a GPLed program in a binary format, you have to "Accompany it with the complete corresponding machine-readable source code, which must be distributed... on a medium customarily used for software interchange". I don't think that anyone could credibly argue that hard copy would meet those requirements.
Section 3 goes to say that the source code has to be in "the preferred for of the work for making modifications to it". This rules out distributing shrouded or otherwise-obfuscated sources.
Also...and I'm being totally serious here, what part of the GPL says it can't be linked with proprietary componets? I can't seem to find that in the Terms of Use....
It is in condition #2 of the GPL, which reads (in part): "These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it."
This is the (in)famous "viral" part of the GPL. Some people think it is a bug; I see it as a feature.
The combination of autoconf and automake is definitely the way to go.
The autoconf and automake docs can be a bit obscure; most people seem to learn how to use them by studying the configure.in and Makefile.am files from other projects.
There is also a brief discussion of autoconf and automake in Havoc Pennington's Gtk+/Gnome Application Development that might help you get started. This fine book is available on-line: here is the relevant section.
Actually, we should probably have a topical subgrouping, sort of like usenet. For instance, I think it'd be nice to know that linux stuff is under *.linux.comp and microsoft stuff under ms.comp.
OK, I'll bite --- where would Slashdot go in a usenet-style hierarchy? comp.linux.slashdot? sci.nanotechnology.slashdot? misc.legal.patents.slashdot? news.nerds.slashdot? soc.people.natalie-portman.slashdot?
This would lead to the bane of Usenet: massive cross-posting (or cross-URLing, in this case). Instead of everyone clamoring to get a .com, they would clamor to get their address into as many groups as possible.
If you want your internet indexed into hierarchical categories, don't worry. Someone has already done that.
We really need to link the namespace to geographical distribution - like the .us domain hierarchy.
I hear this argument from time to time, and I've never really understood it. It seems to me that trying to superimpose geography onto the net is both unnecessary and wrong-headed. The net is inherently a non-physical place: you never know quite where you are, and it doesn't matter.
How fine would the geographical gradiations be: By state? By city? The problems with this seem insurmountable. One example: If my company is in New York, but if my servers are located in New Jersey, are they at www.foo.com.ny.us or www.foo.com.nj.us? In this case, it seems like either answer is the wrong one.
In any case, this still wouldn't solve the inherent problem of domain name disputes, unless you used very fine geographic specifications and also added some sort of business type designation. Does anyone really want to type www.jadegarden.restaurant.chinese.upper-west-side. manhattan.nyc.ny.us into the location line of their browser?
On a minor note could someone create say an obfusicated version of the source code for release and then release the code so that it would take just a little longer to get ahold of?
No. Term #3 of the GPL specifies that "the source code for a work means the preferred form of the work for making modifications to it." So you are free to write your own obfuscated code by hand, but passing it through some obfuscation program is not allowed.
There is a general sense that, besides trying to increase shareholder value, Red Hat and VA are giving back to the community by employing GNOME hackers, kernel hackers, etc.
Red Hat and VA benefit from doing this, of course, but only in the sense that it benefits the free software community as a whole.
What sort of things are Corel doing along these lines? Have you hired any free software celebrities and given them the mandate to hack on anything want? What non-Corel development projects are you funding? Besides bigger and better graphical installers, what benefits will the average non-Corel Debian user derive from your involvement?
Given how much interesting real stuff is going on in science all of the time, you'd think that Wired would have no trouble finding stories to write. Instead, they seem to have consciously decided to focus on "fringe" stuff (cold fusion, anti-gravity, etc.), publishing long, credulous pieces that parrot various crackpot's claims that they are being "oppressed by the scientific establishment".
Now this article is nowhere near as bad as some of the stuff they've put out in the past, but it seems like further evidence of something I've long suspected: they are using science writers who both don't know and don't like science. Clearly it takes some goofy X-files type stuff to get their attention.
The irony is that the demographic reading Wired is probably much more interested in science, and in a much better position to understand and appreciate non-stupid science coverage, than your average readers.
However, given that the best money buys the best lawyer
Actually, legal positions at organizations like the ACLU and the Southern Poverty Law Center are incredibly difficult to get. The competition among law students for these $10/hour jobs is fierce... so it isn't just megacorporations that have access to the best legal talent.
I often hear people express the sentiment that technology changes too fast to be properly regulated, but I've never heard any evidence for this that I've found convincing.
I agree that bad laws can be rendered obsolete by increases in technology. The classic example is export controls that make it illegal to export processors above a certain fixed number of MIPS, or crypto with more than a fixed number of bits. But the problem here is not with the concept of regulation; the problem is that these laws are written with the assumption that the environment is static, and this problem is not just limited to regulations related to technology. These export controls are no more stupid than the US campaign finance laws, in which the maximum donations are not indexed for inflation.
Look at anti-trust law --- it is over 100 years old and, despite the protestations of those with a vested interest, it still seems very well suited to the current situation.
The "technology moves too fast to be regulated" argument seems to me to just be a rather poorly thought out rationalization for a libertarianism. (Which isn't a terrible thing, IMHO, but certainly much better arguments can made for it than this.)
Can anyone produce a non-specious example of some part of technology that is really inherently beyond any regulation, even if that regulation was to be crafted by knowledgeable hackers?
What are your thoughts on both the current status and the future of interoperability between KDE and Gnome in areas like components, CORBA, etc.
Do you see the two projects moving closer together, moving further apart, or staying about the same?
Errr... even though a full-blown anti-trust action against IBM never took place, IBM *did* operate under a consent decree from 1956 through 1996 that limited its ability to behave monopolistically in the hardware market.
It is impossible to predict the exact impact that it had, but the growth of the computer hardware industry in the 1960's might have been significantly stifled if the Feds had given IBM the "freedom to innovate".
To make the leap from "federal anti-trust action" (which has an approx 100-year history, and hasn't managed to utterly crushed the American Way Of Life(tm) in all of that time) to "bureaucrats dictating what code to write, who you can it be sold to (sic)" is just absurd. You might as well argue that a ban on assault rifles will lead to Green Berets kicking down your door to impound your pocket knife...
I can confirm this. I just talked to E*Trade and was told that, due to a big screw-up, everybody in the DSP was sent this notice, but that the DSP shares have *not* been allocated yet.
:-)
What a fiasco this has become. Hopefully VA will steer clear of E*Trade when they do their IPO. (But, despite all this, it sure would be nice if they did a community-DSP program too.