Why not call the OS by the name of the distribution? That means Debian is 'Debian GNU/Linux' and Redhat is 'Redhat Linux' and something else might be 'Joe-Bob Monkey/Linux'.
If you want to use an OS called 'GNU/Linux', use Debian.
(If you want an OS with real package management, use Debian, too.)
Hate to nitpick, but that's wronger. Put a space before and after your em-dashes (or faked em-dashes -- like this one). This visually and logically separates them from the words before and after them.
Visual separation increases readability; logical separation tells the browser that it can break a line.
Incidentally, --- isn't uncommon as a replacement for an em-dash, because that's the TeX code for it.
This is also what separates spam from junk mail. Junk mail has to be delivered because the Post Office is a public entity (and, technically, your mailbox is federal property).
Totally negatory. Bulk mail is delivered because the bulk mailers pay for the privilege. They're consuming Postal Service resources, and they pay for those resources. In fact, the bulk-mail industry effectively subsidizes the rest of the Postal Service's operations, because they actually pay more than the cost of the resources their mail consumes.
Spammers do not pay for the privilege. Spam takes up disk space on the mail server I administer, and my employer is not compensated for this unauthorized use of our services. This is why spamming is an act of theft-of-services.
The ACLU tends to err on the side of free speech. This means that they often end up defending (legally, not morally) unpopular causes such as the right of Nazis and Klansmen to demonstrate.
The argument that needs to be made, then, is that spamming is not protected speech; it is in fact theft of services (which is already a crime). Because it consumes resources belonging to people who have not agreed to be involved in the spamming -- and who would, if asked, refuse to be involved in spamming -- it amounts to unlawful conversion of those resources.
What are the attributes of socialism? Socialism involves the forced common ownership of the means of production, generally by the state and in the name of the people, with the intention of creating economic egalitarianism.
Free Software/Open Source Software has NONE of these attributes.
1. Force. The establishment of socialism invariably involves some degree of force; this may be violent revolution or may merely be expropriation or forced nationalization ("eminent domain") of resources. Free Software doesn't involve any sort of force: nobody is attempting a violent overthrow of Microsoft, nor "expropriating" (pirating) Windows source code to use in FS projects.
2. Common ownership of the means of production. In socialism, the means of production (factories, farms, etc.) are owned by the government. In Free Software, the "means of production" would be computers, compilers, and the Internet infrastructure itself. None of these are owned in common --- you own your own computer; the compiler is licensed to you by its owners under the GPL; the Internet infrastructure is owned by the various ISPs, telcos, and other such entities.
3. Involvement of the state. In socialism, the state administers the use of resources and means of production. The state is not involved at all in Free Software, except insofar as a state agency contributes to FS (e.g. Beowulf, from NASA), or uses FS. At no point is the state involved in the administration of FS as a whole: FS is administered severally by such people as package maintainers, kernel maintainers (thank you Alan and Linus!), and such.
4. Operation in the name of the people. A socialist regime operates (or claims to operate) in the name of the people. Very few Free Software authors write their software with "the public good" as their primary aim; they generally write it because they need to solve a problem, or because they believe it to be interesting.
5. Intention of economic egalitarianism. It is the aim of socialism (though it has never been accomplished) to achieve economic equality among the people --- to abolish the so-called "class system" of "capitalism". Economics is rarely a concern in the world of Free Software; the "freedom" involved is that of "free speech" (i.e. liberty) and not "free beer" (i.e. unlimited economic opportunity). Those who theorize about the economics of FS tend to compare it neither to a "command economy" (socialism) nor a "market economy" but rather a "gift economy".
I am really rather tired of hearing FS/OSS referred to as socialist. If I had to ascribe any political ideology to FS in general, it would be something like "communitarian libertarianism". This would take the ideal of voluntary cooperation (communitarianism) and couple it with the ideal of freedom from coercion (libertarianism).
This would distinguish this ideology from "communitarian socialism" (that variety of socialism which would use communitarian structures such as intentional communities and local governments to establish a larger socialist regime) and from "individualist libertarianism" (that variety of libertarianism which proclaims competition to be the sole law of the jungle, ignoring the fact that cooperation is just as natural).
gcc was first compiled with a non-gcc compiler, which could not have known how to infect gcc, because gcc hadn't existed when the previous compiler was written.
Since then, gcc has always been compiled using gcc, and its source has been open for public inspection. egcs has also been compiled from gcc and egcs versions which are also publicly inspected.
The proof that publicly-inspected versions of gcc and egcs are not infected follows via recursive proof. G(0) is clean, because it was compiled with a non-gcc compiler; G(n+1) is clean if G(n) is clean, because its source is publicly inspected. Hence G(n) is clean for all n.
The problem enters when a version of gcc or egcs is **secretly** forked from the main (publicly inspectable) source tree, and *its* code is contaminated, and it is then distributed in binary form, as if it were legitimate. Once binaries enter the arena from an unverifiable source, you have a problem, since if anyone compiles a (clean-source) version of gcc with an infected compiler, they only get infected compilers as a result.
Solutions:
* Maintain a "reference compiler binary" for each platform, located at FSF or CYGNUS. This would be a copy of gcc or egcs which had been derived from an old (pre-gcc) compiler, and which was thus known (by the above recursive proof) to be clean.
* Trace through all compilers with a debugger. While it's possible that a debugger could be hacked to skip over the trojan routines, it would be MUCH harder to hide than the largely black-box process of compilation.
An analogy:
Every newly-compiled compiler has two "parents": its own source code, and the compiler which was used to compile it. Infection is a dominant genetic trait: if *either* parent is infected, then the "baby" compiler will be infected too.
If you detect an infection in a compiler (by stepping through it with a debugger, for instance) then you know that one or both of its parents were themselves infected. This could theoretically be used to trace an infection back to the source patch that introduced it.
One reason that the "source is language" meme is a good one to spread is that it would tend to bring programs under the protection of existing protections for speech and press. For instance, it would undercut encryption-export laws, whose Constitutionality in the U.S. has hinged in court on source *not* being speech.
Your comparison of free software to free clip-art demonstrates this point even better: graphic arts are already considered speech, at least under U.S. law.
Source code is an expression of ideas, just as much as a mathematical treatise is. The difference is that source code, besides *expressing* the ideas, *implements* them at the same time.
This is the kind of pressure we need. This may well spur U.S. software companies (including, perhaps, our big-stick-carrying friends in Redmond) to pressure the Government to dispose of the crypto export restrictions.
Why this bugs the heck out of me
on
Cyber Vigilantes
·
· Score: 1
This bugs the heck out of me for two very simple reasons:
1. Floods don't just flood the target system; they increase load (sometimes dramatically) on all the routers and links between the flooder and the target. While many DoS attacks are not floods, a lot of the simpler ones (e.g. ICMP directed-broadcast amplified ping flooding, aka "smurf" attacks) are.
2. I administer Linux and Unix systems for a small college. If some freshman IRChead here decides to do stupid things to some remote site, I would much prefer that the remote sysadmin send me logs and ask nicely for the problem to be solved, rather than trying to attack my systems.
When I find a system here being portscanned, I don't start plotting revenge against the evil hAx0rZ. I do a reverse DNS on the originating site, get the admins' addresses from whois, send them the appropriate log clippings, with a nice note saying "I think you're harboring a cracker; please do something about it."
This gets results.
I'm *certain* it gets better results than smurfing the offending site back.
Quick version: Bill Gates dropped out. Linus Torvalds didn't. Which would you rather be like?
Long version: There's a lot more to an education than learning professional skills. At least part of it is learning how to think; how to distinguish the excellent from the merely adequate; how to understand and appreciate the work that has been done before you, as well as the project you're currently working on; and so forth.
If you, O reader, are a "geek" looking at college in the future, I strongly recommend that you get a good liberal-arts education as well as any CS/EE/etc. you're planning on taking. "Well-rounded" isn't just some idea that your high school guidance counselors invented to mock you. Very few of the great geek heroes are nothing but computer geeks.
Take Larry Wall, for instance. When he began his studies, he and his wife went into theology and linguistics, seeking to become translator-missionaries. And while I can't recommend theology as a responsible discipline, one can't help but notice the positive influence of Wall's linguistic work on Perl.
And don't bother trying to do real programming without at least some theoretical CS. Where else are you going to learn that you really shouldn't sort directory listings with bubble-sort? The early programmers of MS-DOS never learned that...
Sure, you can get a job without a degree. Is it a good idea for your life? No. Consider again Gates and Torvalds. Which do you think is better in the sack? Where do you think you're going to get more practice at that -- in a full-time tech job at age 19, or in college?
Slashdot Mirror
Why not call the OS by the name of the distribution? That means Debian is 'Debian GNU/Linux' and Redhat is 'Redhat Linux' and something else might be 'Joe-Bob Monkey/Linux'.
If you want to use an OS called 'GNU/Linux', use Debian.
(If you want an OS with real package management, use Debian, too.)
Hate to nitpick, but that's wronger. Put a space before and after your em-dashes (or faked em-dashes -- like this one). This visually and logically separates them from the words before and after them.
Visual separation increases readability; logical separation tells the browser that it can break a line.
Incidentally, --- isn't uncommon as a replacement for an em-dash, because that's the TeX code for it.
This is also what separates spam from junk mail. Junk mail has to be delivered because the Post Office is a public entity (and, technically, your mailbox is federal property).
Totally negatory. Bulk mail is delivered because the bulk mailers pay for the privilege. They're consuming Postal Service resources, and they pay for those resources. In fact, the bulk-mail industry effectively subsidizes the rest of the Postal Service's operations, because they actually pay more than the cost of the resources their mail consumes.
Spammers do not pay for the privilege. Spam takes up disk space on the mail server I administer, and my employer is not compensated for this unauthorized use of our services. This is why spamming is an act of theft-of-services.
The ACLU tends to err on the side of free speech. This means that they often end up defending (legally, not morally) unpopular causes such as the right of Nazis and Klansmen to demonstrate.
The argument that needs to be made, then, is that spamming is not protected speech; it is in fact theft of services (which is already a crime). Because it consumes resources belonging to people who have not agreed to be involved in the spamming -- and who would, if asked, refuse to be involved in spamming -- it amounts to unlawful conversion of those resources.
What are the attributes of socialism? Socialism involves the forced common ownership of the means of production, generally by the state and in the name of the people, with the intention of creating economic egalitarianism.
Free Software/Open Source Software has NONE of these attributes.
1. Force. The establishment of socialism invariably involves some degree of force; this may be violent revolution or may merely be expropriation or forced nationalization ("eminent domain") of resources. Free Software doesn't involve any sort of force: nobody is attempting a violent overthrow of Microsoft, nor "expropriating" (pirating) Windows source code to use in FS projects.
2. Common ownership of the means of production. In socialism, the means of production (factories, farms, etc.) are owned by the government. In Free Software, the "means of production" would be computers, compilers, and the Internet infrastructure itself. None of these are owned in common --- you own your own computer; the compiler is licensed to you by its owners under the GPL; the Internet infrastructure is owned by the various ISPs, telcos, and other such entities.
3. Involvement of the state. In socialism, the state administers the use of resources and means of production. The state is not involved at all in Free Software, except insofar as a state agency contributes to FS (e.g. Beowulf, from NASA), or uses FS. At no point is the state involved in the administration of FS as a whole: FS is administered severally by such people as package maintainers, kernel maintainers (thank you Alan and Linus!), and such.
4. Operation in the name of the people. A socialist regime operates (or claims to operate) in the name of the people. Very few Free Software authors write their software with "the public good" as their primary aim; they generally write it because they need to solve a problem, or because they believe it to be interesting.
5. Intention of economic egalitarianism. It is the aim of socialism (though it has never been accomplished) to achieve economic equality among the people --- to abolish the so-called "class system" of "capitalism". Economics is rarely a concern in the world of Free Software; the "freedom" involved is that of "free speech" (i.e. liberty) and not "free beer" (i.e. unlimited economic opportunity). Those who theorize about the economics of FS tend to compare it neither to a "command economy" (socialism) nor a "market economy" but rather a "gift economy".
I am really rather tired of hearing FS/OSS referred to as socialist. If I had to ascribe any political ideology to FS in general, it would be something like "communitarian libertarianism". This would take the ideal of voluntary cooperation (communitarianism) and couple it with the ideal of freedom from coercion (libertarianism).
This would distinguish this ideology from "communitarian socialism" (that variety of socialism which would use communitarian structures such as intentional communities and local governments to establish a larger socialist regime) and from "individualist libertarianism" (that variety of libertarianism which proclaims competition to be the sole law of the jungle, ignoring the fact that cooperation is just as natural).
What's a "Lintel" box? Nobody who actually *uses* Linux names their system after a part of a door.
gcc was first compiled with a non-gcc compiler, which could not have known how to infect gcc, because gcc hadn't existed when the previous compiler was written.
Since then, gcc has always been compiled using gcc, and its source has been open for public inspection. egcs has also been compiled from gcc and egcs versions which are also publicly inspected.
The proof that publicly-inspected versions of gcc and egcs are not infected follows via recursive proof. G(0) is clean, because it was compiled with a non-gcc compiler; G(n+1) is clean if G(n) is clean, because its source is publicly inspected. Hence G(n) is clean for all n.
The problem enters when a version of gcc or egcs is **secretly** forked from the main (publicly inspectable) source tree, and *its* code is contaminated, and it is then distributed in binary form, as if it were legitimate. Once binaries enter the arena from an unverifiable source, you have a problem, since if anyone compiles a (clean-source) version of gcc with an infected compiler, they only get infected compilers as a result.
Solutions:
* Maintain a "reference compiler binary" for each platform, located at FSF or CYGNUS. This would be a copy of gcc or egcs which had been derived from an old (pre-gcc) compiler, and which was thus known (by the above recursive proof) to be clean.
* Trace through all compilers with a debugger. While it's possible that a debugger could be hacked to skip over the trojan routines, it would be MUCH harder to hide than the largely black-box process of compilation.
An analogy:
Every newly-compiled compiler has two "parents": its own source code, and the compiler which was used to compile it. Infection is a dominant genetic trait: if *either* parent is infected, then the "baby" compiler will be infected too.
If you detect an infection in a compiler (by stepping through it with a debugger, for instance) then you know that one or both of its parents were themselves infected. This could theoretically be used to trace an infection back to the source patch that introduced it.
One reason that the "source is language" meme is a good one to spread is that it would tend to bring programs under the protection of existing protections for speech and press. For instance, it would undercut encryption-export laws, whose Constitutionality in the U.S. has hinged in court on source *not* being speech.
Your comparison of free software to free clip-art demonstrates this point even better: graphic arts are already considered speech, at least under U.S. law.
Source code is an expression of ideas, just as much as a mathematical treatise is. The difference is that source code, besides *expressing* the ideas, *implements* them at the same time.
This is the kind of pressure we need. This may well spur U.S. software companies (including, perhaps, our big-stick-carrying friends in Redmond) to pressure the Government to dispose of the crypto export restrictions.
This bugs the heck out of me for two very simple reasons:
1. Floods don't just flood the target system; they increase load (sometimes dramatically) on all the routers and links between the flooder and the target. While many DoS attacks are not floods, a lot of the simpler ones (e.g. ICMP directed-broadcast amplified ping flooding, aka "smurf" attacks) are.
2. I administer Linux and Unix systems for a small college. If some freshman IRChead here decides to do stupid things to some remote site, I would much prefer that the remote sysadmin send me logs and ask nicely for the problem to be solved, rather than trying to attack my systems.
When I find a system here being portscanned, I don't start plotting revenge against the evil hAx0rZ. I do a reverse DNS on the originating site, get the admins' addresses from whois, send them the appropriate log clippings, with a nice note saying "I think you're harboring a cracker; please do something about it."
This gets results.
I'm *certain* it gets better results than smurfing the offending site back.
Long version: There's a lot more to an education than learning professional skills. At least part of it is learning how to think; how to distinguish the excellent from the merely adequate; how to understand and appreciate the work that has been done before you, as well as the project you're currently working on; and so forth.
If you, O reader, are a "geek" looking at college in the future, I strongly recommend that you get a good liberal-arts education as well as any CS/EE/etc. you're planning on taking. "Well-rounded" isn't just some idea that your high school guidance counselors invented to mock you. Very few of the great geek heroes are nothing but computer geeks.
Take Larry Wall, for instance. When he began his studies, he and his wife went into theology and linguistics, seeking to become translator-missionaries. And while I can't recommend theology as a responsible discipline, one can't help but notice the positive influence of Wall's linguistic work on Perl.
And don't bother trying to do real programming without at least some theoretical CS. Where else are you going to learn that you really shouldn't sort directory listings with bubble-sort? The early programmers of MS-DOS never learned that...
Sure, you can get a job without a degree. Is it a good idea for your life? No. Consider again Gates and Torvalds. Which do you think is better in the sack? Where do you think you're going to get more practice at that -- in a full-time tech job at age 19, or in college?