There's something about cooperating with a commercial organization that you're in direct competition with.
The nice thing about Debian (from a corporate-executive point of view) is that they don't compete with _anyone_, so there's no risk of having conflicting interests.
Every vendor that builds a distribution solves a number of integration, configuration, and "user experience" problems, cleans up a few bugs, and provides redundant services (two FTP servers are better than one, three are better than two, etc). If those bug fixes and enhancements are propagated back to the component packages then everyone's experience is improved.
There is a use for short-term copyrights even on commodity software. It takes a week or two for warez sites to catch up, so the absolute minimum copyright period for software should probably be a week or two, possibly a month.
Most video games could benefit from a copyright of five years, after which there is little loss in loss of protection. News and current information programs (TV/radio/WWW) become much less valuable a year after their initial release. Operating system software is actually less useful when it is encumbered by copyright at all, but there is an argument that maybe a 1-year or 3-year copyright is appropriate.
20 years, though, is much too long. By that point everyone who can really benefit from having access to copyright software has probably bit the bullet and re-implemented it anyway. How much does Microsoft make from sales of PC-DOS these days?
Perhaps we should shorten the length of copyrights for software by one year per year until we find a workable duration.
Of course music, movies, and books have to be handled differently; software is generally used as a tool which requires maintenance and extension during its life, while music generally is recorded once and is simply reproduced exactly thereafter. Enterprises live or die on their software; their phone system's hold-music is generally much less critical.
Red Hat relies heavily on trademark protection. They don't use patents or copyrights, but they do use trademarks quite aggressively.
I don't know how trademark protection got into this discussion. There is a requirement that the identities of proprietors of anything (from IP to ice cube makers) be easily and unambiguously identifiable; otherwise, it would be impossible to have informed consumers because nobody would know who makes what.
Trademark law (and corporate name law which is virtually identical) simply says that you can't misrepresent company A as company B, especially if A and B are competitors. You're allowed to use trademarks in any other situation that you would be allowed to use a person's name. People get into silly lawsuits over trademarks, but then they get into lawsuits over libel and slander too.
One of the _nice_ things about copyrights is that they keep the people who produce copyright works on physical media from hiring people who really know what they're doing to find ways to stop such media from being copied without permission. DIVX makes money without legal copyright protection too. I find DVD's "copying restrictions" (breakage by design) much too annoying to enjoy any movies in that format--I can't imagine how intolerable DIVX must be.
If copyright law is abolished, so are laws against reverse engineering. The FSF could then set itself up as a central repository and information clearing house for discoveries from backward engineers.
Trade secrets were only a legal problem if such secrets were disclosed to you by their owners, and there was some framework in place wherein you would be obligated not to reveal that information (e.g. you signed an NDA or agreed to one by opening a package). Otherwise, any trade secrets you discover are fair game.
"Hi, I'm Joe Blow. I wrote the little spinning globe thingy on Internet Explorer and several other critical library functions. I just quit Microsoft...and I'm not granting them a license to use my code any more. Thpppt."
I'd rather buy only the tracks I like instead of being forced to purchase hundreds of megabytes of music data that I don't want. The problem is that 90% crappy CD's cost the same as 10% crappy CD's, not one of exceess copyright protection or lack thereof.
If nobody else can support the product, it means nobody can use it as well. That mistake will correct itself given a little time.
I think it's a little simplistic to say that the problem of open-source vs closed-source is congruent to the problem of copyright vs non-copyright. There is significant overlap but I think there is a valuable place for each of the proprietary, freely-available, open-source and closed-source combinations.
Patenting a compression algorithm means you cannot communicate with someone who only understands that algorithm, or who is broadcasting data encoded in that algorithm.
"editors weed out (much of) the crap that we wouldn't want to waste our time on, publishers provide a distribution channel."
The Internet can provide this function at very low cost (certainly low enough that revenue streams such as advertising can pay for it).
Star Wars costs $100 million today. Ten years from now it might be something that could be produced in someone's basement with a normal desktop computer. We are already seeing some fairly high-quality productions (certainly not major-studio-quality, but not bad either) being produced by groups of people with Pentium-based video editing software.
We would still have Star Wars, it would just happen later. Or perhaps earlier, given that more people would be working on the necessary video editing and CGI animation software.
Possibly the allegation is that the person on the other end of the Internet chat was Arkan, while the criminal nature of Arkan himself is not in dispute.
On the Internet, nobody knows you're a brutal, murderous warlord.
Too true. Most of the significant contributions (aside from simple bug fixes) I've seen in "real" free software (e.g. Linux, gcc, etc) have been made by people who needed to build a tool to get some other job done.
For an ISP who has to pay (and therefore feed) a solid technical person, all the costs for contributing to free software are already paid for. At that point the incremental costs of using closed-source software (with all its quality and service problems) are just too high.
A number of research groups (both government-sponsored and from private industry) do significant work in the form of software projects, which become open-source.
gcc has become a de facto standard compiler among chip vendors, particularly microcontroller (yeah, it has _multiple_ 64 bit registers and they call it a microcontroller...right...backinmyday, a CPU with more than 64 bits in all registers _combined_ powered a high-end desktop machine...) and risc CPU's, because a chip vendor has no interest in making it hard for anyone to use their CPU's, least of all themselves.n
Commercial software is capitalistic _outside of producers_ (i.e. when you go from producer to producer or producer to consumer, you have competition and economic exchange). Commercial software is very much not capitalistic _within its producers_ (i.e. within a single corporation, there is ultimately a handful of people who make resource allocation and production planning decisions). This is why commercial software is so inefficient compared to OSS--only a few people within a corporate software vendor have a chance to make important decisions about the direction of software development.
Commercial software development is to natural selection in software what farming is to natural selection in biology. The resulting product in both cases is often something that would not survive for very long outside of its artificial habitat--and it's really the distributors, not the working farmers, that make all the big profits.
Consider where Windows would be today if Microsoft and its large corporate customers weren't pouring _billions_ of dollars in cash on top of it every year, like a farmer in the desert pouring water and chemical fertilizer on ground that really shouldn't be growing anything but cacti and salt crystals. If corporations were not islands of planned economy within a capitalistic world, then corporations would not standardize on operating systems--they would tell their employees to select for themselves whatever tools create the best personal productivity and therefore earn the most money. Arguably it is this _lack_ of capitalism that is all that holds many corporations together--look at a venture capital startup software company and see if the desires of the executives match the desires of the customers or even other employees.
By contrast, OSS is more like fishing or mining. (Hmmm...I hear the sound of an analogy being stretched to breaking). There is a lot of software out there, and a lot more people who would like to write some. OSS software vendors can be thought of as just a filtering mechanism that scoops up gigabytes of crappy software, extracts the 0.1% of it that can be packaged into a product, throws the rest away, and charges a nominal fee for the service. They may cultivate or manufacture some software with their own money (analogous to fish farming or chemical mineral manufacture) if they so choose, meaning that the two models for software development can be mixed to an extent.
Look at the problem from the other direction: How many people who have any self-respect would deliver pizza?
My impression of the typical job duties of a pizza delivery person leads me to believe that pizza delivery is something you do only if there are no other alternatives, and you do it when you need the money for survival (yours or your loved ones), or to pay for something that will get you out of the pizza delivery job.
Are there any people who actually _enjoy_ pizza delivery?
And that marketing company may in fact rely on the contributions of its users, and thereby avoid doing any R&D spending at all. They continue to make money promoting, supporting (or organizing support for), and distributing free software. If necessary, others can compete with them in market niches, but ultimately if the market does not support more than one multi-billion-dollar vendor then there will be only one multi-billion-dollar vendor.
I don't see a problem yet.
It seems perfrectly fair to me--whoever has the best marketing wins in the business world. Since the software itself would be free, those of us with real work to do can either 1) use an off-the-shelf system, or 2) customize an off-the-shelf system, or 3) build the whole thing from scratch, as we have to do today. The difference is that those of us using option #2 don't have to spend more than the cost of option #3 for non-technical reasons. The rest of you can go be the little lemmings you are, if you want to.
The real problem I have with the business world is that it keeps intruding into software development. Once that problem is solved (and I don't particularly care how it ends, as long as it ends) then we can get on with our coding in peace.
The entire _concept_ of non-open-source software comes to us exclusively from the business world--real software developers would never dream of withholding source code because it is simply counterproductive to making good software (except as an educational exercise in the difference between interface and implementation, of course).
Wow, if you can't earn $50 from each of 2500 copies of "a really useful and killer piece of software" for $50 each, _regardless of license_, then you are doing something wrong. Go get some business clues from any shareware vendor or contract consultant.
Really useful and killer pieces of software are exceedingly hard to come by; only a handful are made each year, and most of those are niche products.
Most money spent on software per year is spent on maintaining or improving existing applications, or on supporting them. This is the service that big corporations will fork over the big dollars for. If you want to reliably earn money in software over the long term, this is where you do it.
Of course under the GPL it is possible that e.g. Red Hat will make more money than you do. But that doesn't stop you from earning money--in fact, if your business strategy is right, it simply increases your own marketability while eliminating your distribution and advertising costs.
It doesn't take a whole lot of thought to make money out of GPL software--just organize your business so you aren't competing with your own distributors (i.e. your users).
1. the quite severe impact that a software bug can have on a business that relies on Internet-connected computer systems, and
2. the history of almost total disregard for quality in the software industry,
I would say that an OS vendor should be under legal obligation to prove that their product is _not_ defective (e.g. by releasing source code), because the common case is that it _is_, and the aggregate cost of failures is large.
In fact, I would go further and say that a user should be under no obligation to a vendor until after the user has approved the source code and put it into production--in other words, a money-back guarantee if it doesn't work. Anything less is highway robbery and software vendors know it.
Like...ummm...active signatures? We've been doing just that with PGP for _years_ now.
Kernel-of-the-week Club
on
Linux 2.2.4
·
· Score: 1
Some of us are _paid_ to test out these kernels.
We have customers who don't care about kernel releases as long as they work, so we keep track of which ones do work (with specific test cases for each customer).
Kernel-of-the-week Club
on
Linux 2.2.4
·
· Score: 1
Huh? I have it automated. If my test system ever goes down, I get whatever kernel was most recently released when it reboots. Since it generally only crashes due to kernel bugs, this is a pretty fun way to keep up.
One system crashed on 2.1.128, it rebooted in 2.1.131, and is still running 2.1.131...
Linux's NFS client implementation (allegedly) works now. One thing that does work is that the client actually uses an NFS lock server, whereas 2.0 didn't. So if you aren't running a lockd on your NFS server, you need the nolock mount option.
That's nothing. 2.2.2 had some real karma
on
Linux 2.2.4
·
· Score: 1
2.2.2 came out on 22/2/1999. Sounds like another Pentium floating-point bug.;-)
Yes, but I'd rather listen to MP3 lossiness instead of AM or FM lossiness any day. Plus AM and FM both have really awful noise filtering, and there's no way to pre-queue audio for when you drive under a bridge or into underground parking.
Slashdot Mirror
There's something about cooperating with a commercial organization that you're in direct competition with.
The nice thing about Debian (from a corporate-executive point of view) is that they don't compete with _anyone_, so there's no risk of having conflicting interests.
Every vendor that builds a distribution solves a number of integration, configuration, and "user experience" problems, cleans up a few bugs, and provides redundant services (two FTP servers are better than one, three are better than two, etc). If those bug fixes and enhancements are propagated back to the component packages then everyone's experience is improved.
There is a use for short-term copyrights even on commodity software. It takes a week or two for warez sites to catch up, so the absolute minimum copyright period for software should probably be a week or two, possibly a month.
Most video games could benefit from a copyright of five years, after which there is little loss in loss of protection. News and current information programs (TV/radio/WWW) become much less valuable a year after their initial release. Operating system software is actually less useful when it is encumbered by copyright at all, but there is an argument that maybe a 1-year or 3-year copyright is appropriate.
20 years, though, is much too long. By that point everyone who can really benefit from having access to copyright software has probably bit the bullet and re-implemented it anyway. How much does Microsoft make from sales of PC-DOS these days?
Perhaps we should shorten the length of copyrights for software by one year per year until we find a workable duration.
Of course music, movies, and books have to be handled differently; software is generally used as a tool which requires maintenance and extension during its life, while music generally is recorded once and is simply reproduced exactly thereafter. Enterprises live or die on their software; their phone system's hold-music is generally much less critical.
Red Hat relies heavily on trademark protection. They don't use patents or copyrights, but they do use trademarks quite aggressively.
I don't know how trademark protection got into this discussion. There is a requirement that the identities of proprietors of anything (from IP to ice cube makers) be easily and unambiguously identifiable; otherwise, it would be impossible to have informed consumers because nobody would know who makes what.
Trademark law (and corporate name law which is virtually identical) simply says that you can't misrepresent company A as company B, especially if A and B are competitors. You're allowed to use trademarks in any other situation that you would be allowed to use a person's name. People get into silly lawsuits over trademarks, but then they get into lawsuits over libel and slander too.
One of the _nice_ things about copyrights is that they keep the people who produce copyright works on physical media from hiring people who really know what they're doing to find ways to stop such media from being copied without permission. DIVX makes money without legal copyright protection too. I find DVD's "copying restrictions" (breakage by design) much too annoying to enjoy any movies in that format--I can't imagine how intolerable DIVX must be.
If copyright law is abolished, so are laws against reverse engineering. The FSF could then set itself up as a central repository and information clearing house for discoveries from backward engineers.
Trade secrets were only a legal problem if such secrets were disclosed to you by their owners, and there was some framework in place wherein you would be obligated not to reveal that information (e.g. you signed an NDA or agreed to one by opening a package). Otherwise, any trade secrets you discover are fair game.
There's a nasty hole here...
"Hi, I'm Joe Blow. I wrote the little spinning globe thingy on Internet Explorer and several other critical library functions. I just quit Microsoft...and I'm not granting them a license to use my code any more. Thpppt."
I'd rather buy only the tracks I like instead of being forced to purchase hundreds of megabytes of music data that I don't want. The problem is that 90% crappy CD's cost the same as 10% crappy CD's, not one of exceess copyright protection or lack thereof.
If nobody else can support the product, it means nobody can use it as well. That mistake will correct itself given a little time.
I think it's a little simplistic to say that the problem of open-source vs closed-source is congruent to the problem of copyright vs non-copyright. There is significant overlap but I think there is a valuable place for each of the proprietary, freely-available, open-source and closed-source combinations.
Patenting a compression algorithm means you cannot communicate with someone who only understands that algorithm, or who is broadcasting data encoded in that algorithm.
"editors weed out (much of) the crap that we wouldn't want to waste our time on, publishers provide a distribution channel."
The Internet can provide this function at very low cost (certainly low enough that revenue streams such as advertising can pay for it).
Star Wars costs $100 million today. Ten years from now it might be something that could be produced in someone's basement with a normal desktop computer. We are already seeing some fairly high-quality productions (certainly not major-studio-quality, but not bad either) being produced by groups of people with Pentium-based video editing software.
We would still have Star Wars, it would just happen later. Or perhaps earlier, given that more people would be working on the necessary video editing and CGI animation software.
"The military will not allow this for a variety of obvious reasons,"
Huh? Which reasons are those? Surely by the time the bomb-cam coverage hits the Internet, it's far too late for defenders to do anything about it?
"Oh, look Sarge...you can see our tank from the nose cone of the cruise missi--"
Possibly the allegation is that the person on the other end of the Internet chat was Arkan, while the criminal nature of Arkan himself is not in dispute.
On the Internet, nobody knows you're a brutal, murderous warlord.
Too true. Most of the significant contributions (aside from simple bug fixes) I've seen in "real" free software (e.g. Linux, gcc, etc) have been made by people who needed to build a tool to get some other job done.
For an ISP who has to pay (and therefore feed) a solid technical person, all the costs for contributing to free software are already paid for. At that point the incremental costs of using closed-source software (with all its quality and service problems) are just too high.
A number of research groups (both government-sponsored and from private industry) do significant work in the form of software projects, which become open-source.
gcc has become a de facto standard compiler among chip vendors, particularly microcontroller (yeah, it has _multiple_ 64 bit registers and they call it a microcontroller...right...backinmyday, a CPU with more than 64 bits in all registers _combined_ powered a high-end desktop machine...) and risc CPU's, because a chip vendor has no interest in making it hard for anyone to use their CPU's, least of all themselves.n
Ummm...no.
Commercial software is capitalistic _outside of producers_ (i.e. when you go from producer to producer or producer to consumer, you have competition and economic exchange). Commercial software is very much not capitalistic _within its producers_ (i.e. within a single corporation, there is ultimately a handful of people who make resource allocation and production planning decisions). This is why commercial software is so inefficient compared to OSS--only a few people within a corporate software vendor have a chance to make important decisions about the direction of software development.
Commercial software development is to natural selection in software what farming is to natural selection in biology. The resulting product in both cases is often something that would not survive for very long outside of its artificial habitat--and it's really the distributors, not the working farmers, that make all the big profits.
Consider where Windows would be today if Microsoft and its large corporate customers weren't pouring _billions_ of dollars in cash on top of it every year, like a farmer in the desert pouring water and chemical fertilizer on ground that really shouldn't be growing anything but cacti and salt crystals. If corporations were not islands of planned economy within a capitalistic world, then corporations would not standardize on operating systems--they would tell their employees to select for themselves whatever tools create the best personal productivity and therefore earn the most money. Arguably it is this _lack_ of capitalism that is all that holds many corporations together--look at a venture capital startup software company and see if the desires of the executives match the desires of the customers or even other employees.
By contrast, OSS is more like fishing or mining. (Hmmm...I hear the sound of an analogy being stretched to breaking). There is a lot of software out there, and a lot more people who would like to write some. OSS software vendors can be thought of as just a filtering mechanism that scoops up gigabytes of crappy software, extracts the 0.1% of it that can be packaged into a product, throws the rest away, and charges a nominal fee for the service. They may cultivate or manufacture some software with their own money (analogous to fish farming or chemical mineral manufacture) if they so choose, meaning that the two models for software development can be mixed to an extent.
Look at the problem from the other direction: How many people who have any self-respect would deliver pizza?
My impression of the typical job duties of a pizza delivery person leads me to believe that pizza delivery is something you do only if there are no other alternatives, and you do it when you need the money for survival (yours or your loved ones), or to pay for something that will get you out of the pizza delivery job.
Are there any people who actually _enjoy_ pizza delivery?
And that marketing company may in fact rely on the contributions of its users, and thereby avoid doing any R&D spending at all. They continue to make money promoting, supporting (or organizing support for), and distributing free software. If necessary, others can compete with them in market niches, but ultimately if the market does not support more than one multi-billion-dollar vendor then there will be only one multi-billion-dollar vendor.
I don't see a problem yet.
It seems perfrectly fair to me--whoever has the best marketing wins in the business world. Since the software itself would be free, those of us with real work to do can either 1) use an off-the-shelf system, or 2) customize an off-the-shelf system, or 3) build the whole thing from scratch, as we have to do today. The difference is that those of us using option #2 don't have to spend more than the cost of option #3 for non-technical reasons. The rest of you can go be the little lemmings you are, if you want to.
The real problem I have with the business world is that it keeps intruding into software development. Once that problem is solved (and I don't particularly care how it ends, as long as it ends) then we can get on with our coding in peace.
The entire _concept_ of non-open-source software comes to us exclusively from the business world--real software developers would never dream of withholding source code because it is simply counterproductive to making good software (except as an educational exercise in the difference between interface and implementation, of course).
Wow, if you can't earn $50 from each of 2500 copies of "a really useful and killer piece of software" for $50 each, _regardless of license_, then you are doing something wrong. Go get some business clues from any shareware vendor or contract consultant.
Really useful and killer pieces of software are exceedingly hard to come by; only a handful are made each year, and most of those are niche products.
Most money spent on software per year is spent on maintaining or improving existing applications, or on supporting them. This is the service that big corporations will fork over the big dollars for. If you want to reliably earn money in software over the long term, this is where you do it.
Of course under the GPL it is possible that e.g. Red Hat will make more money than you do. But that doesn't stop you from earning money--in fact, if your business strategy is right, it simply increases your own marketability while eliminating your distribution and advertising costs.
It doesn't take a whole lot of thought to make money out of GPL software--just organize your business so you aren't competing with your own distributors (i.e. your users).
Given:
1. the quite severe impact that a software bug can have on a business that relies on Internet-connected computer systems, and
2. the history of almost total disregard for quality in the software industry,
I would say that an OS vendor should be under legal obligation to prove that their product is _not_ defective (e.g. by releasing source code), because the common case is that it _is_, and the aggregate cost of failures is large.
In fact, I would go further and say that a user should be under no obligation to a vendor until after the user has approved the source code and put it into production--in other words, a money-back guarantee if it doesn't work. Anything less is highway robbery and software vendors know it.
Like...ummm...active signatures? We've been doing just that with PGP for _years_ now.
Some of us are _paid_ to test out these kernels.
We have customers who don't care about kernel releases as long as they work, so we keep track of which ones do work (with specific test cases for each customer).
Huh? I have it automated. If my test system ever goes down, I get whatever kernel was most recently released when it reboots. Since it generally only crashes due to kernel bugs, this is a pretty fun way to keep up.
One system crashed on 2.1.128, it rebooted in 2.1.131, and is still running 2.1.131...
Uhhh...read clue-point-two...
Linux's NFS client implementation (allegedly) works now. One thing that does work is that the client actually uses an NFS lock server, whereas 2.0 didn't. So if you aren't running a lockd on your NFS server, you need the nolock mount option.
2.2.2 came out on 22/2/1999. Sounds like another Pentium floating-point bug. ;-)
Wow, talk about not understanding Microsoft.
If Microsoft decides to modify the Linux kernel, they will modify the Linux kernel and that will be the end of it.
They won't move their Office suite to any new kernel revisions except their own. Period.
Yes, but I'd rather listen to MP3 lossiness instead of AM or FM lossiness any day. Plus AM and FM both have really awful noise filtering, and there's no way to pre-queue audio for when you drive under a bridge or into underground parking.