... Thus, the justification for patent and copyright laws....
Bzzzt. Wrong!
The justification for Patent & Copyright laws was to encourage printers and innovators to make the substantial investments that were required to bring books & inventions to a point where they might recoup the investment.
In the case of printed works, typesetting used to be a very significant cost, since a typo early in a book could require some or all of the following pages to be re-typeset, with the associated possibility of introducing new errors.
Thus, without copyright to protect that particular printing of a book, a second unscrupulous publisher could undercut the original publisher (since they didn't have to pay the typesetters & proofreaders), rendering the original typesetting of books unprofitable.
If that situation had been allowed to persist many fewer books would have been published, so copyright was introduced to provide an incentive to publishers.
This clearly does not apply to most software production, since the setup costs for many software companies can be as little as the cost of a cheap PC.
Obviously, people that sell shrink-wrap software are able to make more money by enforcing an artificial scarcity, and copyright allows them to do this, but equally clearly the vast bulk of the world's software is written to solve a problem, and that would happen regardless of the charging structure that you put in place to pay for it.
I've been making my living from Free Software programming and support since 1993, and I assure you, I'm not starving:-)
The very idea of the lay person using Linux is ridiculous...
As a data point that refutes that assertion, I cite a friend of mine who counts as a lay person IMO.
He's a commercial estate agent, and has been cheerfully using Linux (Debian GNU/Linux & Star Office) for all his computing needs for over a year.
His level of technical competence with M$ stuff was such that he knows how to CD around a DOS file system, and perhaps edit an autoexec.bat file with a 75% certainty of not screwing it up, so perhaps you think he counts as a Windows expert;-)
His main comment about Linux is that he no longer needs to wonder if the machine will refrain from crashing before he managed to print the document he's editing.
I get fewer calls for help from him than I used to when he used DOS/Windows, and almost all of those are due to Netscape screwing up in some way, and all of them were remotely fixable (ssh over the internet) in under 5 minutes, whereas Windows problems would often result in me wasting a decent chunk of a weekend, and having to re-install the machine from scratch.
So, he doesn't do his own sysadmin work, but then he didn't for Windows either, and it takes about one twentieth as much of my time to support him, and the fact that it is possible to diagnose problems on Linux means that supporting him doesn't make me incoherent with rage, which is nice.:-)
That may be true of your example because, by the sounds of it, the programmer neither assigned copyright, nor granted the right to use/sell/redistribute (in the form of a license).
This is not relevant in the case of code released under the GPL, because they have already granted permision to fork when they released the code under the GPL, so they have no right to revoke that license.
As a software vendor, why should it be of any interest to you if you derive your per copy income from a license fee, or a copying fee?
The FSF has been selling their software tapes for a significant per copy fee, in reasonable quantities pretty much from the start, so your argument that it is not possible to make money from selling copies of GPLed software is demonstrably false.
The GPL demands that the price charged for a copy... be reasonable
Like all license restrictions, that doesn't aply to the copyright holder.
Are you suggesting that you're upset that you cannot make vast amounts of money out of code you didn't even write? Well, Duh!
Yes, apt-get can resume partial files, does download everything to a holding area before installing and with apt-zip you can do the sneakernet thing by copying files to Zip (or other removable media) at work, and taking them home to upgrade you machine (for example).
This article assumes that the total time spent developing Free
Software is a limited resource, and that it is being squandered by
using some of it on redundant applications.
I would prefer to think that the people he's complaining about were
going to want total control over their code base no matter what
license they chose, and the fact that they chose Free Software
licenses means that if they come up with a particularly good idea it
can be easily adopted by other Free Software authors.
If you make this second assumption then the main effect these
authors will have is positive.
OK, so it's a pain trying to find out which text editor one should
use, because the choice is so wide and there is a fair amount of alpha
software around, but I take the fact that we can make such a complaint
as proof that Free Software is in an extremely healthy.
you're missing the fact that (judging from some other comments here) the main site is suffering from the/. effect, whereas the Debian mirrors have enough bandwidth available between them that there is no real chance of them being/.ed
By mentioning the fact that it's in woody he's being helpful to Debian users, and being helpful to everyone else (because he's helping to spread the load).
The package in question is "mail-transport-agent". There is a section of the Debian Policy Manual which defines the features expected in an MTA.
If the feature you seek is within that policy definition of an MTA, then you can depend upon "mail-transport-agent" with confidence. If not, then you need to determine which other package(s) (real or virtual) satisfy your needs.
If this hypothetical need is something that more than one package requires, then defining a new virtual package, defined in terms of that need, provides pretty much what you request.
It would be entirely possible to break the features covered by mail-transport-agent into a whole list of feature-based virtual packages, but from expeience (in Debian) is seems that that is not in fact required.
There are one or two very specific virtual packages in Debian, for example:
c-shell - Anything providing a suitable/bin/csh
wish - Anything that provides/usr/bin/wish
which are effectively what you are calling features. Note, these are more than file dependancies, because they carry with them the requirement that the functionality of the file be "standard"
Are you really trying to tell us that your income is based on the license sales of some software you wrote, or are you just repeating the myth that it's imposible to make a living out of Free Software?
Hint: I make a living out of Free Software - many others do too.
For US based distributions, there's still the issue of US crypto export laws, which while greatly relaxed still gets in the way.
Debian has included OpenSSH pretty much since it was available. It's available from our Non-US software site, and is on CD#1 of Debian 2.2 (unless you get the cut-down US exportable version).
Patents are supposed to allow a time limited monopoly to exploit an invention, to give chance to recoupe development costs, in return for the inventor publishing details of the technique that makes their invention novel.
The details of RSA was published before it was patented, and it was not funded by RSA in the first place.
Also, it seems to me that it's a discovery and not an invention.
Sounds more like a classic abuse of the whole concept of patents to me.
To add insult to injury, they didn't even write a decent implementation of it.
Swap 2 weeks of unenforcible patent rights for some free publicity and the opportunity to witter on about how your patent helped to promote development.
Personally, I never let their patent affect my life in any way (which is why there's no ssh-rsaref package in Debian), but if I had done, it would not have been helpful.
The claim that some benefit accrued to anyone but themselves is drivel.
If you think software patents are a silly idea, sign this petition to help keep Europe software patent free
However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.
Also, libqt is different in that it is distributed with it's source, and until recently was distributed under a license which conflicted with the terms of the GPL in a way that you're probably not equipped to comprehend, given that you failed to spot the above.
That the original coder would still be in violation, and so would now not be allowed to use, modify or distribute that program.
Also, anyone that ever distributed the old version of the program was in violation of the sensitive code's license, and so would still be in violation, as would anyone that modified the program after the sensitive lines were added.
Who does that leave? Well, appart from people that never saw the program, not many.
RMS was not attempting to claim that KDE were in violation of their copyright, since a copyright holder can do what they like without reference to copyright permissions.
So, what was he on about? Clearly, he must be referring to people that violated the GPL by distributing GPLed code against its terms, which would mean people that distributed GPL+Qt binaries.
Who falls into that category?
Everyone except KDE who distributed KDE (and other GPL+Qt) binaries.
KDE (and other GPL+Qt authors) in the few cases where they have imported pure GPL code into their code base
What needs to be done to fix this?
Well in the first case, KDE's "That's Absurd" statement seems to constitute a statement of prior permission --- Problem Solved
That only leaves us with a few lines of pure GPL code which were included in Qt-linked programs without asking the author for their permission. This can be fixed by asking those authors for permission/forgiveness, which seems entirely reasonable to me, given that the code was used in violation of the license. I'm sure that all the authors in question would be perfectly willing to grant such a request.
In conclusion. RMS is simply asking for the few authors who's copyright was violated, to be asked if they hold a grudge. Is that so wrong?
One aspect of this that amuses me is that Debian doesn't need anyone's permission to distribute any of this, since we were in compliance (or made best efforts to be so) all along.
or often even better, simply set up a local squid cache, with a line like this in the/etc/squid.conf:
refresh_pattern debian.org/.*\.deb$ 129600 100% 129600
and point your hundreds of machines at the cache to get all the benefits of a local mirror, none of the admin overhead, and only downloading packages you actually use.
You want the latest greatest Gnome? You want to install software from third party sources? You want the latest greatest everything else?
Just put these two lines in/etc/apt/sources.list:
deb http://www.uk.debian.org/debian woody main non-US/main
deb http://spidermonkey.helixcode.com/distributions/de bian unstable main
and you get the latest (albeit, so called unstable) "woody" distribution, with the addition of Gnome from HelixCode.
Admittedly, it's about as unstable as most other distro's are when they're released, but if you can put up with that, it's what you were looking for.
The point is that the author of a work should be free to choose her own license, and not have someone else impose their idea of the ?right? license upon it.
Absolutely correct.
And if an author chooses a license that states the equivalent of "You are forbiden from distributing binaries built from this code" you can trust Debian to comply with that license.
That is the case with the GPLed portions of KDE. The fact that others chose to ignore the license is not our concern.
I think you'll find that the KDE executable contains code derived from Qt header files, and function call references derived from the Qt dynamic libraries.
Given that Troll Tech have asserted that they own the API to the extent that they beleive that if Harmony had succeeded, that KDE/Harmony binaries would be a violation of Troll's copyright, it seems fair to assume that they also think that the binary is a derived work.
If you look at last month's thread on debian-legal, you'll see that your assertion that this is all sorted out is somewhat inacurate.
I don't see Debian changing its stance until either there is a license change on KDE or Qt, or RMS reinterprets the issue (which I'm not expecting to happen, ever).
Don't expect Debian to ignore the terms of the GPL in order to win some sort of popularity contest. We're not interested.
Don't expect Debian to act in a way that undermines the authority of the GPL, by distributing code that combines GPLed source and GPL-conflicting source in one binary.
If you don't like it, use another distribution, or perhaps start your own, by adding Debian to the already available KDE.debs
At the start of this article, he states that good people can do bad things and vice versa.
Why then does he try to draw some conclusions from ESR's gun obsession, when that is nothing to do with free software? Why does he try to draw a connection between anecdotes about RMS's table manners and Free Software?
He later tries to imply that Free Software is all built with stolen or donated time.
My business survives by selling solutions to my clients. My clients are fully aware that I use and write Free Software in the course of providing them with the solutions they require, and cheerfully pay me all the same.
Is the resulting software "Donated", or am I immorally diverting the funds from my clients?
He then goes on to accuse the Free Software movement of a series of moral lapses ("stealing" other people's ideas) which I agree should be addressed with proper acknowledgements of inspiration used. If it wasn't for the fact that the commercial software world seems to have no compunction about failing to give credit where credit is due, he might be saying something of interest with respect to free software, but as it is this is an industry wide failing for which the Free Software cannot be held answerable.
Slashdot Mirror
Try reading the article again.
The programs identified prior to getting shell access were not FSF owned.
glibc1, which is mentioned later, is FSF owned.
Bzzzt. Wrong!
The justification for Patent & Copyright laws was to encourage printers and innovators to make the substantial investments that were required to bring books & inventions to a point where they might recoup the investment.
In the case of printed works, typesetting used to be a very significant cost, since a typo early in a book could require some or all of the following pages to be re-typeset, with the associated possibility of introducing new errors.
Thus, without copyright to protect that particular printing of a book, a second unscrupulous publisher could undercut the original publisher (since they didn't have to pay the typesetters & proofreaders), rendering the original typesetting of books unprofitable.
If that situation had been allowed to persist many fewer books would have been published, so copyright was introduced to provide an incentive to publishers.
This clearly does not apply to most software production, since the setup costs for many software companies can be as little as the cost of a cheap PC.
Obviously, people that sell shrink-wrap software are able to make more money by enforcing an artificial scarcity, and copyright allows them to do this, but equally clearly the vast bulk of the world's software is written to solve a problem, and that would happen regardless of the charging structure that you put in place to pay for it.
I've been making my living from Free Software programming and support since 1993, and I assure you, I'm not starving
The very idea of the lay person using Linux is ridiculous ...
;-)
:-)
As a data point that refutes that assertion, I cite a friend of mine who counts as a lay person IMO.
He's a commercial estate agent, and has been cheerfully using Linux (Debian GNU/Linux & Star Office) for all his computing needs for over a year.
His level of technical competence with M$ stuff was such that he knows how to CD around a DOS file system, and perhaps edit an autoexec.bat file with a 75% certainty of not screwing it up, so perhaps you think he counts as a Windows expert
His main comment about Linux is that he no longer needs to wonder if the machine will refrain from crashing before he managed to print the document he's editing.
I get fewer calls for help from him than I used to when he used DOS/Windows, and almost all of those are due to Netscape screwing up in some way, and all of them were remotely fixable (ssh over the internet) in under 5 minutes, whereas Windows problems would often result in me wasting a decent chunk of a weekend, and having to re-install the machine from scratch.
So, he doesn't do his own sysadmin work, but then he didn't for Windows either, and it takes about one twentieth as much of my time to support him, and the fact that it is possible to diagnose problems on Linux means that supporting him doesn't make me incoherent with rage, which is nice.
That may be true of your example because, by the sounds of it, the programmer neither assigned copyright, nor granted the right to use/sell/redistribute (in the form of a license).
This is not relevant in the case of code released under the GPL, because they have already granted permision to fork when they released the code under the GPL, so they have no right to revoke that license.
IANAL but then, clearly, neither are you.
What does this mean?
If you are trying to say that the SPL is an Open Source (or Free Software) license, then you're wrong, it is not.
It fails to meet clause 6 of the The Debian Free Software Guidelines (DFSG), or if you prefer, Clause 6 of the Open Source Definition.
Don't expect to see any SPL covered code in Debian.
As a software vendor, why should it be of any interest to you if you derive your per copy income from a license fee, or a copying fee?
... be reasonable
The FSF has been selling their software tapes for a significant per copy fee, in reasonable quantities pretty much from the start, so your argument that it is not possible to make money from selling copies of GPLed software is demonstrably false.
The GPL demands that the price charged for a copy
Like all license restrictions, that doesn't aply to the copyright holder.
Are you suggesting that you're upset that you cannot make vast amounts of money out of code you didn't even write? Well, Duh!
Cheers, Phil
Yes, apt-get can resume partial files, does download everything to a holding area before installing and with apt-zip you can do the sneakernet thing by copying files to Zip (or other removable media) at work, and taking them home to upgrade you machine (for example).
This article assumes that the total time spent developing Free Software is a limited resource, and that it is being squandered by using some of it on redundant applications.
I would prefer to think that the people he's complaining about were going to want total control over their code base no matter what license they chose, and the fact that they chose Free Software licenses means that if they come up with a particularly good idea it can be easily adopted by other Free Software authors.
If you make this second assumption then the main effect these authors will have is positive.
OK, so it's a pain trying to find out which text editor one should use, because the choice is so wide and there is a fair amount of alpha software around, but I take the fact that we can make such a complaint as proof that Free Software is in an extremely healthy.
you're missing the fact that (judging from some other comments here) the main site is suffering from the /. effect, whereas the Debian mirrors have enough bandwidth available between them that there is no real chance of them being /.ed
By mentioning the fact that it's in woody he's being helpful to Debian users, and being helpful to everyone else (because he's helping to spread the load).
The package in question is "mail-transport-agent". There is a section of the Debian Policy Manual which defines the features expected in an MTA.
/bin/csh
/usr/bin/wish
If the feature you seek is within that policy definition of an MTA, then you can depend upon "mail-transport-agent" with confidence. If not, then you need to determine which other package(s) (real or virtual) satisfy your needs.
If this hypothetical need is something that more than one package requires, then defining a new virtual package, defined in terms of that need, provides pretty much what you request.
It would be entirely possible to break the features covered by mail-transport-agent into a whole list of feature-based virtual packages, but from expeience (in Debian) is seems that that is not in fact required.
There are one or two very specific virtual packages in Debian, for example:
c-shell - Anything providing a suitable
wish - Anything that provides
which are effectively what you are calling features. Note, these are more than file dependancies, because they carry with them the requirement that the functionality of the file be "standard"
Are you really trying to tell us that your income is based on the license sales of some software you wrote, or are you just repeating the myth that it's imposible to make a living out of Free Software?
Hint: I make a living out of Free Software - many others do too.
For US based distributions, there's still the issue of US crypto export laws, which while greatly relaxed still gets in the way.
Debian has included OpenSSH pretty much since it was available. It's available from our Non-US software site, and is on CD#1 of Debian 2.2 (unless you get the cut-down US exportable version).
Patents are supposed to allow a time limited monopoly to exploit an invention, to give chance to recoupe development costs, in return for the inventor publishing details of the technique that makes their invention novel.
The details of RSA was published before it was patented, and it was not funded by RSA in the first place.
Also, it seems to me that it's a discovery and not an invention.
Sounds more like a classic abuse of the whole concept of patents to me.
To add insult to injury, they didn't even write a decent implementation of it.
Swap 2 weeks of unenforcible patent rights for some free publicity and the opportunity to witter on about how your patent helped to promote development.
Personally, I never let their patent affect my life in any way (which is why there's no ssh-rsaref package in Debian), but if I had done, it would not have been helpful.
The claim that some benefit accrued to anyone but themselves is drivel.
If you think software patents are a silly idea, sign this petition to help keep Europe software patent free
Looks like you didn't check very thoroughly:
However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.
Also, libqt is different in that it is distributed with it's source, and until recently was distributed under a license which conflicted with the terms of the GPL in a way that you're probably not equipped to comprehend, given that you failed to spot the above.
That the original coder would still be in violation, and so would now not be allowed to use, modify or distribute that program.
Also, anyone that ever distributed the old version of the program was in violation of the sensitive code's license, and so would still be in violation, as would anyone that modified the program after the sensitive lines were added.
Who does that leave? Well, appart from people that never saw the program, not many.
So, what was he on about? Clearly, he must be referring to people that violated the GPL by distributing GPLed code against its terms, which would mean people that distributed GPL+Qt binaries.
Who falls into that category?
- Everyone except KDE who distributed KDE (and other GPL+Qt) binaries.
- KDE (and other GPL+Qt authors) in the few cases where they have imported pure GPL code into their code base
What needs to be done to fix this?Well in the first case, KDE's "That's Absurd" statement seems to constitute a statement of prior permission --- Problem Solved
That only leaves us with a few lines of pure GPL code which were included in Qt-linked programs without asking the author for their permission. This can be fixed by asking those authors for permission/forgiveness, which seems entirely reasonable to me, given that the code was used in violation of the license. I'm sure that all the authors in question would be perfectly willing to grant such a request.
In conclusion. RMS is simply asking for the few authors who's copyright was violated, to be asked if they hold a grudge. Is that so wrong?
One aspect of this that amuses me is that Debian doesn't need anyone's permission to distribute any of this, since we were in compliance (or made best efforts to be so) all along.
or often even better, simply set up a local squid cache, with a line like this in the /etc/squid.conf:
refresh_pattern debian.org/.*\.deb$ 129600 100% 129600
and point your hundreds of machines at the cache to get all the benefits of a local mirror, none of the admin overhead, and only downloading packages you actually use.
You want the latest greatest Gnome? You want to install software from third party sources? You want the latest greatest everything else?
/etc/apt/sources.list:
e bian unstable main
Just put these two lines in
deb http://www.uk.debian.org/debian woody main non-US/main
deb http://spidermonkey.helixcode.com/distributions/d
and you get the latest (albeit, so called unstable) "woody" distribution, with the addition of Gnome from HelixCode.
Admittedly, it's about as unstable as most other distro's are when they're released, but if you can put up with that, it's what you were looking for.
why? because nobody would be interested in getting a copy? ;-)
That is REALLY STUNNINGLY UNHELPFUL
The box you link to is a P166, which is currently running at a load average of around 10 just dealing with the rsyncs from the mirrors.
Please moderate this moron to the nether reaches of hell --- Thank you.
The point is that the author of a work should be free to choose her own license, and not have someone else impose their idea of the ?right? license upon it.
Absolutely correct.
And if an author chooses a license that states the equivalent of "You are forbiden from distributing binaries built from this code" you can trust Debian to comply with that license.
That is the case with the GPLed portions of KDE. The fact that others chose to ignore the license is not our concern.
I think you'll find that the KDE executable contains code derived from Qt header files, and function call references derived from the Qt dynamic libraries.
.debs
Given that Troll Tech have asserted that they own the API to the extent that they beleive that if Harmony had succeeded, that KDE/Harmony binaries would be a violation of Troll's copyright, it seems fair to assume that they also think that the binary is a derived work.
If you look at last month's thread on debian-legal, you'll see that your assertion that this is all sorted out is somewhat inacurate.
I don't see Debian changing its stance until either there is a license change on KDE or Qt, or RMS reinterprets the issue (which I'm not expecting to happen, ever).
Don't expect Debian to ignore the terms of the GPL in order to win some sort of popularity contest. We're not interested.
Don't expect Debian to act in a way that undermines the authority of the GPL, by distributing code that combines GPLed source and GPL-conflicting source in one binary.
If you don't like it, use another distribution, or perhaps start your own, by adding Debian to the already available KDE
Cheers, Phil.
At the start of this article, he states that good people can do bad things and vice versa.
Why then does he try to draw some conclusions from ESR's gun obsession, when that is nothing to do with free software?
Why does he try to draw a connection between anecdotes about RMS's table manners and Free Software?
He later tries to imply that Free Software is all built with stolen or donated time.
My business survives by selling solutions to my clients. My clients are fully aware that I use and write Free Software in the course of providing them with the solutions they require, and cheerfully pay me all the same.
Is the resulting software "Donated", or am I immorally diverting the funds from my clients?
He then goes on to accuse the Free Software movement of a series of moral lapses ("stealing" other people's ideas) which I agree should be addressed with proper acknowledgements of inspiration used. If it wasn't for the fact that the commercial software world seems to have no compunction about failing to give credit where credit is due, he might be saying something of interest with respect to free software, but as it is this is an industry wide failing for which the Free Software cannot be held answerable.
I think you are correct.
As I see it, they have a choice. Either publish the full source of the driver under the GPL, or pay the copyright holder for violating the copyright.
If they opt for the GPL route, great.
If not, they need to be sued. Presumably, proving the case will not be too dificult, since they've already admitted the violation.