Slashdot Mirror
First Draft of GPL Version 3 Released
njan writes "The first draft of version three of the GNU General Public License was released to the public this afternoon. Major improvements touted in version three include changes designed to mitigate the damage posed by new threats to free software such as software patents. One individual stated about the release: 'It is changes in law, not computer technology, that pose the principal challenges to the free software community. Chief among these changes has been the unwise and ill-considered application of patent law to software. Software patents threaten every free software project, just as they threaten proprietary software and custom software. Any program can be destroyed or crippled by a software patent belonging to someone who has no other connection to the program.'"
I think I will continue to stick with the MIT license. It has plain, easy to comprehend terms. It's concise.
I appreciate the effort the FSF is making, but things may be getting out of hand. I know of many developers who feel the same as I do. They just want to create software, without having to get bogged down with legalities. Thankfully, licenses like the BSD license and the MIT license work wonderfully well for us.
Cyric Zndovzny at your service.
GNU GENERAL PUBLIC LICENSE
Discussion Draft 1 of Version 3, 16 Jan 2006
THIS IS A DRAFT, NOT A PUBLISHED VERSION OF THE GNU GENERAL PUBLIC LICENSE.
Copyright (C) 2006 Free Software Foundation, Inc.
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software--to make sure the software is free for all its users. We,
the Free Software Foundation, use the GNU General Public License for
most of our software; it applies also to any other program whose
authors commit to using it. (Some Free Software Foundation software
is covered by the GNU Lesser General Public License instead.) You
can apply it to your programs, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
To protect your rights, we need to make requirements that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if you
distribute copies of the software, or if you modify it.
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have. You must make sure that they, too, receive or can get the
source code. And you must show them these terms so they know their
rights.
Developers that use the GNU GPL protect your rights with two steps: (1)
assert copyright on the software, and (2) offer you this License which
gives you legal permission to copy, distribute and/or modify the software.
For the developers' and author's protection, the GPL clearly explains
that there is no warranty for this free software. If the software is
modified by someone else and passed on, the GPL ensures that recipients
are told that what they have is not the original, so that any problems
introduced by others will not reflect on the original authors'
reputations.
Some countries have adopted laws prohibiting software that enables users
to escape from Digital Restrictions Management. DRM is fundamentally
incompatible with the purpose of the GPL, which is to protect users'
freedom; therefore, the GPL ensures that the software it covers will
neither be subject to, nor subject other works to, digital restrictions
from which escape is forbidden.
Finally, every program is threatened constantly by software patents. We
wish to avoid the special danger that redistributors of a free program will
individually obtain patent licenses, in effect making the program
proprietary. To prevent this, the GPL makes it clear that any patent must
be licensed for everyone's free use or not licensed at all.
The precise terms and conditions for copying, distribution and
modification follow.
GNU GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0. Definitions.
A "licensed program" means any program or other work distributed under
this License. The "Program" refers to any such program or work, and a
"work based on the Program" means either the Program or any derivative work
under copyright law: that is to say, a work containing the Program or a
portion of it, either modified or unmodified. Throughout this License, the
term "modification" includes, without limitation, translation and
extension
why is it so difficult to read?
...will hold up legally, and how much of it is just hot air and rants?
"DRM is fundamentally incompatible with the purpose of the GPL, which is to protect users' freedom; therefore, the GPL ensures that the software it covers will neither be subject to, nor subject other works to, digital restrictions from which escape is forbidden." Sounds good and noble, but will it work?
I saw it on Slashdot, it must be true!
So "freedom for users" has now been redefined to "freedom for users, except for one group of users that we don't like". I'm curious to see who the second group is going to be...
What I'm listening to now on Pandora...
Gah... why couldn't there be a web page that didn't have BR tags at the 80-character mark every time. This is like reading e-mail in the 90s! (Actually this looks like a plone-based site so it's probably serving up auto-generated *ml from a text file... which is no excuse, really. If vim can fix up stuff like that then plone could too.)
What keeps me from using the GPL is the "any later version" option. How do I know that GPL version 17 wont give every user of my software a right to come by my house for a free lunch? Or a car. You know, unless most significant software, everything in this world isn't free as in lunch. That holds for example for most lunches.
Swedish plasma phys. PhD student; MSc EE; knows maths, programming, electronics; finance interest; seeks opportunities
I'm reading this as (bold area): if I compile my code with GCC and link with a GNU library, my code will not fall under the GNU license unlese I sat it does.
The same reason that code is hard to read by non-programmers, or medical papers are hard to read by people without medical training. The law, like any field, needs precise language to communicate. Many words have special legal meanings that are subtly different from common speech (or not so subtle if language has diverged over time). This is necisarry for the same reason that you can't use plain english to write code - plain english leaves to much open for interpretation. When you write legal documents, you want the judge interpreting your document, should it ever go to court, to read it the way you intended it to be read. The best way to do this is to use the accepted legal terminology.
Well, it'll stop those fsckers at TiVo from using Linux in hardware devices that are locked down so you can't read the data or modify the software without serious hardware hacking.
GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
The new GPL have the following:
So patent law mixed with how I use the software, and privately at that. Can I use GPLv3 software in a company (it's not private, usually)? Can I modify it, but not distribute it outside the company? If I don't do this privately, but as a "corporate" person, then it's not private, so I can do what I want (of course not). This is just in the beginning of the new license, and it goes on and on and on and on etc.
Really, why not make a license that I don't need to be a lawyer to understand?
"Freedom for users" was already redefined to mean "freedom for users, except for ones we don't like". You were just ok with who they didn't like before (people using non-GPL licenses). They are simply extending the scope of who they don't like.
Is some smart-ass making a joke about the GIMP image editor?
Only skimmed it so far, but I wonder if their eagerness to stop copyprotection and DRM scheemes means that you can't use the GPL3 for email applications supporting GPG encrpytion, or not at least without giving away your private GPG keys :)
It's a) to long and b) a lot more complicated than previous revisions. At least I kind of understood those. Now it's more a political manifesto, forcing everyone who touches the software to become vegeterian etc..
Fleur de Sel
*NOT* terms and conditions. This is the style of GNU licenses - preamble describing intent, but not legally binding, then the legally-binding terms .
The Raven
I was surprised there would be the ability to add clauses to the license. It seems this will make it more difficult for many people or (especially) organizations to accept, if they have to evaluate the legality for their situation of many similar but different licenses.
I have only read the license through once and have lots of questions to think about. My first impression though is it seems a lot more readable than v2 for which I read "This is legalese what does this mean in English?"
It means that you don't have to release Visual Studio, MFC, and the .Net runtime under the GPL, if you distribute a GPL application that requires those things to run. Unless your code is a compiler or object code intepretor, I don't see why you would think it is excluded.
I like this one, specially the 'unencumbered by patents' part, but I'm not sure if this was already present in GPL v2.
DRM:
The only thing i didn't like was the "illegally invade users' privacy". It's the 'illegally' that concerns me, what if DRM is made legal? I can see a loophole in there. The 'nor for modes of distribution'... does it include 'illegally', too?
Perhaps they should be more explicit, I'm not sure...
Curious that a discussion of the GPL should bring out so many annoymous cowards.
I wonder why?
Must do some stats some day on proportion of AC comments verses subject matter.
Anyone else noticed a correlation?
Open Source Drum Kit, LPLC deve board - mjhdesigns.com
I can't wait to try it out! But are there any binaries to download? I can only seem to find the source code, and I don't have a compiler for Lawyer++.
I didn't see any wording in the draft that addresses this issue either way; every time I thought I did, I found the same or similar wording in version 2. So, is it in there? Will it affect how we publish web applications?
Dewey, what part of this looks like authorities should be involved?
>Shouldn't there be a possibility to make changes to the GPL and release your software under your own, derivative license? (Scary, I >know, but someone might want to do it.) They seem to forbid it as it is now.
I was under the impression that you could make derivatives, you just couldn't then go and call it the GPL, you had to call it something else. I could be entirely wrong though.
What could be better than a jet powered motorcycle? http://www.youtube.com/watch?v=u8l6GTHLSWE
here
My turnips listen for the soft cry of your love
Valid legalese is money. My lawyer can beat up your lawyer.
All those words do exactly the opposite of what you think they do, each one is another "attack vector," if you will, for a well paid lawyer.
This license may very well "attempt to uphold the traditions of the FSF" but will ultimatly bring them down over an arguement along the lines of "depends on what the definition of 'is' is."
The BSD and MIT licenses, while accomplishing something very different, are short and non-ambiguous and therefore (more) defensible.
Did you read the rest of the license, or just stop halfway through? There's an exception for libraries that come as standard with the OS or development environment. What you say is not true at all.
Bogtha Bogtha Bogtha
" As a special exception, the Complete Corresponding Source Code need not include a particular subunit if (a) the identical subunit is normally included as an adjunct in the distribution of either a major essential component (kernel, window system, and so on) of the operating system on which the executable runs or a compiler used to produce the executable or an object code interpreter used to run it, and (b) the subunit (aside from possible incidental extensions) serves only to enable use of the work with that system component or compiler or interpreter, or to implement a widely used or standard interface, the implementation of which requires no patent license not already generally available for software under this License."
THis is the clause that avoids that problem, specificly subclause a. As major essential components of the OS, they do not have to include code (unless the work you are distributing is an OS, such as the Linux kernel).
I still have more fans than freaks. WTF is wrong with you people?
Did you see any earlier versions of this thing? btw political is good.
My turnips listen for the soft cry of your love
One question for OSS projects currently using GPL, will be, should they relicense?
For example, should Linux become GPL3'd?
Discuss...
This reminds me of something similar which has been explored in depth by Karl Popper, namely the problem of how does a free society protect itself from people who would use that freedom to change the society so that its no longer free. For example how does a democracy prevent a party from getting elected which would end the democracy? I cant give you a 25c summary, because it is a complicated subject, and I cant remember all of it anyway (its been years). However, if you're interested, I highly recommend reading Popper. His stuff is easy to read, and he deserves to be more widely known.
My life is an open book ... up to a point.
Some vagueness here, based on "applicable copyright law":
To "propagate" a work means doing anything with it that requires
permission under applicable copyright law, other than executing it on
a computer or making private modifications. This includes copying,
distribution (with or without modification), sublicensing, and in some
countries other activities as well.
And there's other disclaimers about "private" this and that:
This License gives unlimited permission to privately modify and run the
Program, provided you do not bring suit for patent infringement against
anyone for making, using or distributing their own works based on the Program.
But, finally, does this give specific permission for web services?
Propagation of covered works is permitted without limitation provided it
does not enable parties other than you to make or receive copies.
Propagation which does enable them to do so is permitted, as
"distribution", under the conditions of sections 4-6 below.
"I assumed blithely that there were no elves out there in the darkness"
Except, ya know, that the license then goes on to say
As a special exception, the Complete Corresponding Source Code need
not include a particular subunit if (a) the identical subunit is
normally included as an adjunct in the distribution of either a major
essential component (kernel, window system, and so on) of the
operating system on which the executable runs or a compiler used to
produce the executable or an object code interpreter used to run it,
and (b) the subunit (aside from possible incidental extensions) serves
only to enable use of the work with that system component or compiler
or interpreter, or to implement a widely used or standard interface,
the implementation of which requires no patent license not already
generally available for software under this License.
How we know is more important than what we know.
I'd like to point out that in a more perfect world, we wouldn't need the GPL v3. What has happened is that the world has become tremendously more complicated since v2. The v3 helps to address some of those issues.
Let's look at some of the issues that make licensing your software more difficult to do properly.
PATENTS. Software patents have gotten out of control. At the time of v2, there was some concern about patents, but now we have had a direct attack on the GPL from the angle of patents trump copyright. What v3 does is spell out what patents mean and how companies can use patents and GPL in peace. It says that you can't distribute software under the GPL if you own a patent that doesn't allow others to distribute the same software. You have to license that patent to allow distribution by others. Isn't this what we want? This is one technical and hairy issue that most people never think about, but needs to be thought about.
DRM. The DRM issue is another viper's nest. What happens if Sony releases software under the GPL to play movies, and then try to protect the movies under DRM legislation? Think especially if Sony takes a GPL software out in the wild, and then adds their own things to it, and then to satisfy the GPL they try to distribute it under the GPL. Are they doing the right thing, even if it is legal? Of course not. If you build or work on a data interpretation program and then license it under the GPL, you can't ask people not to try to understand how your program works or write another program to deal with the same software. The v3 license is explicit about this.
BSD license folks like to pick on the GPL because it is getting so complicated. They forget that the BSD license has its own problems. The GPL is trying to solve those problems. If you don't understand what the problems with the BSD license are, you can't understand what the GPL is intended to do, nor can you be convinced to use it.
Hint about BSD's problems: MS Windows has a good portion of BSD code in the Windows kernel. They spend a great deal of time modifying and adapting the code and then turning around and selling it, without allowing people the same rights they had. I have a co-worker whose job was to work with this code. The folks who wrote the original code get NOTHING in return, except a by-line in the Windows documentation.
Imagine if I took the Linux Kernel, made it better, then turned around and started selling it for profit, without allowing people to see the changes, nor giving them an opportunity to do the same. The GPL forbids this, and it has been a perfect solution up 'til now. The FSF is trying to predict future "attack vectors" and closing them off before they can become a problem, perhaps through future legislation or actions.
The radical sect of Islam would either see you dead or "reverted" to Islam.
yes, you are entirely wrong. You can certainly derive from the /spirit/, but the text itself is copyrighted (presumably to avoid requiring users re-read the full text every time they download something new, scanning for changes)
-- 'The' Lord and Master Bitman On High, Master Of All
I would say "of other software capable of accessing the same data" needs expansion and or definition in a footer or otherwise. What is this "other software" you're talking about? External, third-party, binary software perhaps?
The sentence itself, while parted with a colon, also feels grammatically unweildy. Should the "of" in "of other software [...]" not be "for"?
This is so totally fallacious that I'm not sure even how to refute it.
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
The longer the lines of text are, the more space you need inbetween each line in order to maintain the same level of readability. Long lines make it difficult to find the next line when you finish a line unless there is a good deal of space inbetween. This is the same reason newspapers break up text into columns. Imagine a newspaper with 18" long lines.
It apparently means proprietary software distributors, those who seek to lock up software and make Free Software unavailable to others by DRM and an unshared key, and the like. In short, the people who the Free Software movement hasn't liked for the past 20 years.
The alternative of absolute freedom to do anything one wants all too quickly becomes a power to distribute proprietary derivatives, and thus becomes unsavory for preserving software freedom; just as it has been for the past 20 years too.
One cannot have all possible freedoms. As the FSF has reminded us many times, my freedom to walk down the street in safety conflicts with your freedom to drive anywhere you want. There are times when one must choose which freedoms to secure and which freedoms to trade away.
You hint at unspecified malevolence suggesting that you're actually casting FUD instead of providing analysis of the new GPL or what the Free Software movement has been talking about for two decades now.
Digital Citizen
You weren't very specific as far as what the difference is between opinion and ideology.
So far what I can tell, is that you're saying an opinion is something you think that you will do, whereas an ideology is something you think others should do. Is this correct? Define your terms, as you made little sense.
GENERATION 26: The first time you see this, copy it into your sig on any forum and add 1 to the generation.
I've made a unified diff of the differences between the GPLv2 and GPLv3-draft, enjoy!
12. Liberty or Death for the Program.
...
If conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot distribute
the Program, or other covered work, so as to satisfy simultaneously your
obligations under this License and any other pertinent obligations, then as
a consequence you may not distribute it at all. For example, if a patent
license would not permit royalty-free redistribution by all those who
receive copies directly or indirectly through you, then the only way you
could satisfy both it and this License would be to refrain entirely from
distribution.
I guess all this means is that if lawmakers want to void or weaken part of the GPL, they'll have to void this part as well.
The FSF have provided a document explaining the rationale behind the changes in each section of the license.
Before blindly criticizing the wording of a certain section, I suggest reading the rationale behind the changes.
- Does the text in the License do what they intended it to do?
- Do you agree or disagree with what they intended?
- Are the possibly-bad side effects of the text which aren't mentioned in the rationale?
I don't think I like it, it has a few clauses that are ambiguous and could be used in a way I don't agree with.
Good thing one can still use the v2 only.
GPLv3 is calling DRM Digital Restrictions Management. A good description, but isn't the term actually Digital Rights Management? Ought to state both terms. Would hate to lose a court case on a technicality based on imprecise terminology. I thought renaming it Restrictions was a clever joke, and we don't want the GPL to look like a joke. Gives opponents a wedge they can use to try to discredit the whole thing.
Intellectual Property is a monopolistic, selfish, and defective concept. It is "tyranny over the mind of man"
Hmm.. so it's now legal to make GPL Java software (v2 only had a restricted version of first part of the clause, and java is *not* shipped OS libraries so didn't count.. not that that stopped anyone).
I'm not really asking if they can be merged with pure pure v3 code. I'm asking whether the baseline version ought to be dragged up to v3. Ought GPL2 to be abandoned?
Note that merging v2+ with v3+ code makes the resuling whole become v3+, the baseline has been moved up. So in that regard the questions overlap. In fact, Linus for example is likely to face that question when considering merging patches that were licensed v3+ or derived from v3+ code. It will eventually get rather uncomfortable trying to hang onto v2+ purity.
"e) They may impose software patent retaliation, which means permission
for use of your added parts terminates or may be terminated, wholly or
partially,"
The draft of Version 3 of the GPL is comptable with CDDL and similar GPL2 incompatable free software licences, this is very good news for developing software that uses components from for example, say FSF and Sun.
My little Linux and tech blog
Suppose I develop a closed-source OS that ships the Gnu ls source and
/bin/ls that was developed independently of /bin or /usr/local/bin is first in the script's path.
/bin/ls that does not recognize -S. The script /usr/local/bin is first in the script's path.
a binary (/usr/local/bin/ls). Within my OS, I have a large non-GPL shell
script that generates reports about system usage. It has the line:
ls -lS | head -10 | mail -s "largest files" root
(Note that -S is the "sort by file size" option, which Gnu ls has, but
many other ls programs don't have.) Consider two scenarios:
(1) My OS has a non-GPL
Gnu ls but does recognize -S. Therefore the script works regardless
of whether
(2) My OS has a non-GPL
works only if
My reading is that scenario (2) is a GPL violation, according to the
"But when you distribute the same sections for use in combination with
covered works" clause. Scenario (1) is not a GPL violation.
I suspect the GPLv3's intent here is to prohibit a closed-source
program that accomplishes the bulk of its work by executing a GPL
program and reformatting its outputs (e.g., displaying them within an
extensive non-GPL GUI). The actual stipulation is that a non-GPL
program cannot rely on the existence of a GPL program for even a very
minor task. With GPLv3, GPL software becomes much less useful on a
closed-source OS, unless the user would be manually interacting with
the GPL program without any vendor-supplied scripting.
Only the other hand, my point is, if you strand your code in v2 like you're suggesting, and accept patches under those terms, then in the entirely predictiable future in which something you meed to include is v3 you'll be forced to chase everyone down and get their permission to relicense. For example, if you need to link a GPL (not LGPL) shared library, and it's 2010, and no modern distro uses the creaky antique version of that library which is GPLv2 (except Debian stable, heh). So you're either forced to recode, relicense, link a copy of the old v2 library statically, or allow your program to fade into historical irrelevance. Like I said, uncomfortable.
The term "widely used interface" seems to take a big bite off the Viralness. Think about things like Netscape plugins or ODBC drivers.
Whenever I hear the word 'Innovation', I reach for my pistol.
AC said something worth reading
Why is it that when you believe something it's an opinion, but when I believe something it's a manifesto?
These sections seem off, and even dangerous:
Under what theory of copyright law is a copyright owner allowed to restrict modification of the work? AFAIK, a copyright holder can only restrict copying of the work or derived works, but cannot restrict the creation of such derived works. I don't need FSF's permission to modify its source any more than I need a book publisher's permission to write notes on the margins.
This is a dangerous precedent; if such a clause were to be upheld, it would make any modifications of a copyrighted work subject to similar prohibitions without requiring redistribution. Freedom to tinker would be history.
Am I totally misreading this?
...meight be more useful: have a look.
I have discovered a truly remarkable sig which this 120 chars is too small to contain.
That definitely rules out StuffIt archives (unless you use StuffIt 1.5.1, which is over 15 years old). I'm not sure that Apple's disk image format (.dmg) is documented either.
In general, it is safe and legal to kill your children. -- POSIX Programmer's Guide
it's just that lots of software is licensed as gpl2 or any later version 'at your option'. it's not a fixed part of gpl2.
for instance the linux kernel is licensed only under gpl2.
my password really is 'stinkypants'
You would be wise to consider the problems with this fair seeming license. What good is simplicity if a license can be subverted? Those who don't learn history are doomed to repeat it.
an ill wind that blows no good
...about downloading and trying any GPL library because ultimately I've found too many unclear explanations on what I can and cannot do with it related to code I make money on. In some cases, I'd like to pay for the right to use it commercially but can't even figure out who to write the check to and for how much.
I wish authors releasing good libraries under GPL would also assume that is OK to make money with it too. Just spell out the terms and conditions for us commercial types too. Not all of what I do is commercial of course, some is free (as in beer) in which case GPL tends to be fine.
I hate being the kind of person who uses other people's contributions but can't contribute back all the time. I hope the new license clears some of this up for me.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
from : http://www.gnu.org/licenses/gpl-faq.html#ModifyGPL
---
Can I modify the GPL and make a modified license?
You can use the GPL terms (possibly modified) in another license provided that you call your license by another name and do not include the GPL preamble, and provided you modify the instructions-for-use at the end enough to make it clearly different in wording and not mention GNU (though the actual procedure you describe may be similar).
If you want to use our preamble in a modified license, please write to for permission. For this purpose we would want to check the actual license requirements to see if we approve of them.
Although we will not raise legal objections to your making a modified license in this way, we hope you will think twice and not do it. Such a modified license is almost certainly incompatible with the GNU GPL, and that incompatibility blocks useful combinations of modules. The mere proliferation of different free software licenses is a burden in and of itself.
---
I guess that implies that I was indeed correct.
What could be better than a jet powered motorcycle? http://www.youtube.com/watch?v=u8l6GTHLSWE
It seems Ty Coon has been dismissed and replaced by Richrt Anus. Why this potty humour in such an important document?
Oh well, what the hell...
We'd all be better off if Microsoft took even more code from open source projects. Why is that? Because it could very well increase the quality of their products.
How does that directly affect you, you may ask? Well, first of all it may reduce the number of Windows-based PCs that can be compromised, and used to send the spam that clogs up your mail servers or your inbox.
Second of all, it may lead to the adoptation of open source developed software and open standards. This allows for better interoperability between Windows and non-Windows systems. For those of us running massive heterogenous networks, anything that eases interaction between different systems is a blessing.
Cyric Zndovzny at your service.
How exactly is Microsoft, for instance, going to suddenly make FreeBSD "unfree"? Sure, they may take code from FreeBSD, modify it, and not release the changes. But that doesn't change the fact that the original FreeBSD code would still be freely available.
Cyric Zndovzny at your service.
But the changes are irrelevant. The original BSD licensed software is still freely available.
If your theory about those with resources throwing money at the projects until they disappear was correct, then we wouldn't have directly competing projects such as FreeBSD, NetBSD, and OpenBSD around, because Microsoft would have bought them out. But that didn't happen, and most likely wouldn't, thus your theory is incorrect. And nevertheless, even if your theory were correct, that yet again wouldn't prevent the original code from being freely available.
Cyric Zndovzny at your service.
As a figter against software patents I have now some experience reading legal documents and here is what I think of this one.
r st-gplv3-draft.html
http://aigarius.blogspot.com/2006/01/ok-i-read-fi
Interesting that in the rationale, they decided against taking action to close the web services/remote execution loophole.
It's doubly interesting that, while they made some changes to combat DRM, they said nothing at all about Trusted/Treacherous Computing, which is the foundational layer enabling most modern DRM.
I think that's good, because I believe you can use the latter to combat the former, just by leveraging existing terms in GPLv2, combined with the public's natural interest in retaining privacy.
It appears to me that communication between a wide subset of software would be outlawed by the "intimate data communication or control flow" clause.
This is a much broader definition of "derived work" than was traditionally supported by the filtration/abstraction/comparison tests used by copyright legislation.
Basically, CORBA and other RPC schemes cannot be used between GPL V3 and other software?
Where is the definition of "intimate data communication" and "flow control".
For example, are Win32 API messages included? How about RPC across the internet? Or HTTP messaging between a webserver and a browser?
It seems to be just as vague and open to interpretation as the previous definitions of program linkage.
It's too bad that you have to understand any kind of legaleese to be a programmer, but that's life.
I think programmers aren't the ones to complain about obscure notation and languages, given that we have invented many more of them than lawyers ever have in the history of the world.
Does someone mind explaining to me the deal with GPL and linking? AFAIK, you can only link GPL'd programs to GPL'd programs.
But what is the inherent difference between linking and communicating with a program in another manner?
If my code communicates with a GPL program via tcp/ip, or via function calls the only logical difference i can see is speed?
The FSF never was about "free as in beer" software, nor was it about "free as in you-can-do-whatever-you-like-with-it" software. I'm sorry you misunderstood this point, but the FSF has been completely clear about it.
The FSF is trying to guarantee a specific set of freedoms, and they are using whatever methods the legal system gives them in order to make those guarantees.
Of course, your kind of confusion extends into other areas. For example, there are many people who think that a free society is a society in which everybody can do whatever they like, but that's false. A free society is a society in which people can do whatever they like, up to the point where it starts interfering with other people's freedoms.
I know some of you new Eclipse/Visual Studio DOT NET guys love 30000
character lines, and don't get me started on perl, but for the projects
I work on having long lines is a drawback.
I wish that programming languages didn't allow freedom to place whitespace (as C, perl, and so forth do). Python is partway there.
The reason why is the same reason that HTML (at least as originally envisioned) was set up to dissociate the storage format from the display formatting. I can stick newlines wherever I want when writing HTML, but it will be displayed however the user wants. If he wants emphasized text to appear in orange, or be set off by *stars*, or to appear in an italic face, he can do so. Dissociating the storage and viewing format provides this benefit.
The problem is that people take advantage of the flexibility of the storage format of programming languages like C to insert display formatting -- a few spaces here and there to line things up -- and there is no clean way to separate the storage format and the display format. A better approach would be to have a fixed storage format (preventing display formatting from being encoded with the file). Each user could make the file display however they want while working with it, and their additions would be free of any of their own display preferences.
With the existing schemes, you can always make a file display however you want -- ram it through GNU indent -- but after you save the file, you've lost the original embedded display formatting information.
If all files used single spaces as separators between all tokens or something along those lines, then an editor could display the file however it wanted.
Of course, it would make editing the language with a regular text editor less good, which is probably why this hasn't really caught on (I think that languages that aren't easily editable in standard text editors are slowly approaching -- I listened to a lecture on aspect-oriented programming, and I think that the largest drawback is that working with the thing in a standard text editor kind of sucks).
Any program relying on (nontrivial) preemptive multithreading will be buggy.
When it comes to patents, the draft is actually not very aggressive about them.
.... lawsuits that lack
the justification of retaliating against other software patent lawsuits
that lack such justification.
Under the draft, you are required to grant a patent license to whoever may use the software. That may not be as agressive as some licences, but it's definitely something to think about.
The wording needs to be cleaned up. This sentence is a doozy:
The conditions must limit retaliation to
GPL v2 was legalese, but the wording was much clearer than this. Hopefully it will be cleaned up before the final version.
Qxe4
I really wish that software development courses (and CS courses, because to be honest, more people with a CS major wind up developing software than writing papers) included a legal component. One course should be enough to at least cover the crucial basics.
Any program relying on (nontrivial) preemptive multithreading will be buggy.
"No, we don't make all of our source code for everything (the UI) available.
Big deal; that doesn't make us Darth Vader."
Err, I never said you were darth vader. I simply said you appreaciate very much the free coding the BSD programmers do for you. It allows you to profit nicely from their labors.
One supposes they don't mind. After all if they didn't want to volunteer code to apple then they would have coded using the GPL. Then again if they did that then you would have to open up your UI. Thank for the BSD programmers, you get to use their code while keeping your UI proprietary.
evil is as evil does
Yeah, but what legal document doesn't?
No. The sentence is saying "development of other software [...], distribution of other software [...], and use of other software [...], as 'development,' 'distribution,' and 'use' are defined under the terms of this license." Factor out the common "of other software [...]" from all three portions and put it at the end.
That particular sentence is fine. There are others, though, that don't have good parallel structure. Consider this passage from section 6:
It's saying that the distribution must be in a format that is publicly documents, that it must be in a format that is unencumbered by patents, and that it must be in a format that is must require no special password. Is must? The alternative reading of that sentence is that the distribution must be in a format that is publicly documented, the distribution unencumbered by patents, and the distribution must require no special password. Distribution unencumbered? There are either too many verbs or not enough.
I can only hope that the authors of the license simply haven't given it to a copy editor yet -- this is just a draft, after all.
Rob
That definitely rules out StuffIt archives [and possibly] Apple's disk image format (.dmg)
Which platforms still use StuffIt? As for .dmg, isn't source code more commonly distributed in .tar.gz or .tar.bz2 files?
Fact is, I'm more worried about traditional (hardware) patents on the physical distribution medium. CD-R, CD-RW, and all DVD formats are still subject to at least physical layer patents and thus are likely to fail the "unencumbered by patents" test.
It seems Ty Coon has been dismissed and replaced
Because coon is a synonym for nigger.
Are you saying those who invented/use the term "DRM" don't say it stands for anything?
You're just plain wrong. The earliest recorded uses of this term do indicate the middle "R" stands for "Rights".
Yeah, it's a doublespeak-type acronym, but it's equally Orwellian to try to change the perception of DRM simply by changing its name.
http://lkml.org/lkml/2005/8/20/95
will hold up legally, and how much of it is just hot air and rants?
I'm not really thrilled with this part:
Regardless of any
other provision of this license, no permission is given to distribute
covered works that illegally invade users' privacy, nor for modes of
distribution that deny users that run covered works the full exercise of
the legal rights granted by this License.
I don't like spyware, but I think that the GPL is a really, really bad place to try to "fix" spyware. And illegally under what laws? Can I now not distribute my software in Podunk, North Dakota because it's illegal, under Podunk city law, to do Tivo-style recommendations based on phone-home viewing data? It seems like it's pretty broad, and doesn't seem within the scope of promoting Free Software.
I wouldn't really like to see restrictions on spammers using GPL software either -- not because I like spammers much, but because I don't think that the GPL should be saddled with anything non-crucial.
Any program relying on (nontrivial) preemptive multithreading will be buggy.
RTFL: Section 2 "This License gives unlimited permission to privately modify and run the Program, provided you do not bring suit for patent infringement against anyone for making, using or distributing their own works based on the Program."
When I write my program and my program is dynamically linked, I have no control over which libraries it will be linked against. If it links against a GPL'd library, do I have to make my program GPL? What if I developed it with a GPL'd library, but it can link against a corresponding proprietary library? How about if I linked against a proprietary library, the only copy of this library being my own, and everyone else links against a GPL'd library; does my application need to be GPL'd? Is it up to the user to check that none of his proprietary programs link against GPL'd libraries?
The Affero license has been "testing" the Web Services clause for a while.
The problem for Free Software purists is that it's a EULA: It restricts use, not just copying. That seems to contradict the "not a contract" part of the GPL, and probably can't be enforced in many jurisidctions (although it is a way of fighting back against UCITA or similar laws / court decisions that make EULAs binding).
Ah, but there's the rub - many people wouldn't see a publicly hosted server as being something that is "privately run". In fact, a lot of people think that there should be a clause in the GPL that allows access to the source of GPLed server code - eg for websites that use GPLed bulletin board software.
As to the other part of the question, of course you could still charge for access to the game, regardless of the GPL status of the code. What you couldn't do would be prevent someone from setting up their own competing service.
If I were to set up a MMORPG or similar and GPL some of it, I'd probably GPL the code but keep the copyright on the content (ie sounds, graphics, etc). That's what id do when they GPL their old games - just because you can now get the source of Quake 3 doesn't mean you can legally distribute the content that it requires.
It's official. Most of you are morons.
You can _always_ make money with GPL programs. Or at least, you can legally try. You could take e.g. gimp, and sell it, provided you comply with the gpl terms: that anyone can take the source code you're distributing and modify it, and sell it themselves if they wish. The thing is, if you do this precise kind of stuff, you might have negative publicity. It is nevertheless your right under the GPL.
More clarifications at the FSF's GPL FAQ's page.
"I think it would be a good idea!"
Gandhi, about Internet Security
The patents chunk is still very significant. What it comes down to is that any developer can inadvertently effectively nullify a company's patents (at least WRT GPLv3 implementations, which can be shoehorned via a separate process model into allowing even non-GPLv3 software to do patented things).
The company I work for (very large) is dicey about even using software components under the MIT license, because they're unsure about the risks of this whole open source thing -- they require that a lawyer sign off an OK. Given their patent portfolio, I can almost guarantee that they would never allow a programmer to release anything under GPLv3.
As a matter of fact, I would be surprised if IBM does not have significant feedback on this.
If GPLv3 does go through, I could even see IBM spawning a child company just to produce open source software, to keep their patent portfolio secure.
Another point I'm interested in -- by my reading, the requirement to provide a free license is on a per-patent basis, *not* on a per-claim basis.
Many, many patents contain intentionally over-broad claims. Adding a claim only costs about $100 or so, plus lawyer's fees. If you have 30 claims, ranging from the absurdly broad to the unnecessarily narrow, if someone challenges your patent, they might only get 17 of the claims rejected -- so you still have almost as much IP as you theoretically could have had if you claimed the very maximum amount of IP you could get away with. If you had only two claims (one very broad, one very narrow), and the broad claim is rejected, you are left with only the narrow claim.
Under this GPLv3 draft, if my reading is correct, someone releasing software affected by a patent that is over-broad must provide a license providing free use of *all* claims of that patent. This is very significant if GPLv3 catches on. This means that all the patents out there with extremely broad claims will be effectively nullified by any of these companies that want to release GPLv3 software.
I like most of the GPLv3, but I think that the patents bit may make it unacceptable -- the FSF has a *lot* of weight to throw around in the form of software under their license, but I think that they're being too ambitious in trying to fight software patents with the GPL. Unless, of course, my reading is incorrect.
Any program relying on (nontrivial) preemptive multithreading will be buggy.
Lots of lawyers are wrong in their legal views and some of them even lose cases. Lots of basketball players who can perform slam dunks sometimes miss the basket and sometimes lose games. Being an expert doesn't mean you're above being questioned. But you apparently don't even like for people to question whether you're really a lawyer, so you post anonymously, giving us no information to determine if you're actually a lawyer as you claim to be.
Eben Moglen, counsel for the FSF, is also a lawyer who has said repeatedly that "Licenses are not contracts". Perhaps you would take the matter up with him instead. After reading his essays and listening to his talks, I find him to be a far more patient and informative speaker than you appear to be. Then again, he might object to some anonymous nobody claiming to be a lawyer arguing a point with him that he's rehashed so many times.
Pamela Jones, a journalist with a paralegal background who runs Groklaw, has gone on record saying that "The GPL is a license, not a contract" in which she cites Moglen's essay and expands on it a bit. Perhaps you'd rather discuss the matter with her, since she too might be more on your level of expertise.
But I think it's telling that instead of patiently explaining the difference between the terms 'license' and 'contract' you instead chose to take a needlessly confrontational and remarkably uninformative route to point out that most non-lawyers don't understand the terms. It's unfortunate that the /. moderators don't seem to penalize such posts in an attempt to raise the level of discourse here.
Digital Citizen
>Do you think it could be just your own expectations projected onto what you actually see?
:o)
Isn't that exactly what statistics are about?
I dont it's the case, but I'll try an analysis on past stories, if I can.
Open Source Drum Kit, LPLC deve board - mjhdesigns.com
For some, but even now over a decade after it came out in its final form many people still don't understand it (including people who use or license programs under it). You can often tell how deep people's understanding is by listening intently to their questions and counterarguments—people aren't stupid, but they are sometimes ignorant. Fortunately, there is a well-known cure for ignorance.
Give the new GPL some time to be discussed and eventually we'll get to the same point we are now with GPLv2—a lot of people in the Free Software community understand what it says in large part and they get on with using it (it's the most popular Free Software license). Lawyers will study the new GPL (or so I'm told from the recent New York Times article about it) and a few cases will be tried that involve the new GPL and more people will feel like doing business with it is acceptable.
Finding out that non-lawyers find legalese to be "incomprehensible" is nothing new, just as non-programmers don't understand what programmers and computer technicians are talking about when they throw around terms in common programming parlance. I'm sure most trades have their in-speak that is hard to (shall we say, grok?) for outsiders.
Digital Citizen
I'm slightly concerned about the wording on startup messages (section 5.2c).
I've two issues with this. One is prolixity. When a system starts up, dozens and perhaps hundreds of individual components are started. If each component is required to print:
then the startup log is going to become unmanageably big, and hard to search for important information.
The other issue is granularity. The stuff I build is webapps, comprising sometimes as many as forty or fifty individual servlets. I don't control which servlet is started first, because servlets are started when they are first required to serve content a user (or other system) has requested. If I'm required to 'show' this at startup, am I required to show it as each servlet starts, or only when the first servlet in a given webapp starts? If the latter, do I have to keep track of which servlet is started first in order that it can know to print the message?
I'm old enough to remember when discussions on Slashdot were well informed.
There are two parts where I've found clearly unclear formulations:
Section 6.[3] b): "accompanied by a written offer, valid
for at least three years and valid for as long as you offer spare parts
or customer support for that product model"
Is that an and or an or? That is, does it mean if a company stops offering spare parts/support before the end of the three years, it doesn't need to give source code any more, despite the three years are not over, or does it mean that if after those three years, they still offer spare parts or support, they also need to provide the source, despite the three years have passed? I suspect the latter, in which case the appropriate formulation would IMHO be "valid for as long as you offer spare parts or customer support for that product model, but at least for three years".
7. e): "The conditions must
limit retaliation to a subset of these two cases: 1. Lawsuits that lack
the justification of retaliating against other software patent lawsuits
that lack such justification. 2. Lawsuits that target part of this
work, or other code that was elsewhere released together with the parts
you added, the whole being under the terms used here for those parts."
Same question: and or or? That is, has it to be a subset of the union of the two sets defined by the conditions given, or of the intersection?
BTW, I think the definition of the term "aggregate" in section 5.[2] would logically belong in the section 0. Definitions.
The Tao of math: The numbers you can count are not the real numbers.
To me this sounds at least in places as though the FSF are using the GPL as an opportunity to further dictate to their already-converted more than anything else.
For one thing, it insists that software patents should either be made freely usable by everyone, or else not obtained at all. Whether you argue that this is a desirable sentiment or not, the bottom line is that this directive is beyond Stallman's legitimate power to make. The draft tries to decree a similar prohibition against DRM, which again, is beyond its power and will be found unenforcable.
I've said for a while that Stallman's ego and delusions about his own level of authority were getting out of control...to me, this draft fairly conclusively proves that assertion.
If you are looking for libraries that you can add to your own commercial projects with impunity, I suggest looking towards LGPL code. With the LGPL you can distribute the library along with your binary and are obligated to release the source code to only the library. In fact, you can change the library around for your benefit and use it and still only have to publish the library.
:-)
Much more commercial friendly that the GPL.
Help! I'm a slashdot refugee.
Thank you, that's helpful. I've looked at LGPL to some extent,and it makes sense. GPL is too heavy handed. Its like you give me a hammer and tell me that if I build a house with it that I must let everyone live in it.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
Some tools are valuable as 'free' for use. Some represent a small niche market for very specific kinds of work. Making the choice between either not doing the work at all or doing it without hope of recovering revenue from it is the place the GPL leaves you.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
> ..about downloading and trying any GPL library because ultimately I've found
> too many unclear explanations on what I can and cannot do with it related to code
> I make money on.
I find them quite clear. My experience is that people who find it unclear are generally trying to do something against the spirit of the license, that might or might not be against the letter of the license. In that case, I'd suggest you find another way.
> In some cases, I'd like to pay for the right to use it commercially but can't even
> figure out who to write the check to and for how much.
It is exactly the same as for all other software.
If there are multiple contributors to the software and no indication of a "transfer of copyright" procedure for patches, don't bother. It is too complicated to get permission from all the copyright holds.
If the copyright holder is the FSF, don't bother. They are not for sale.
Otherwise, send an email to the copyright holder (i.e. either the sole author, or the address the "copyright transfer proecedure" refers to= and ask.
How difficult can it be?
I swear I read something completely different on that same page a year ago :/
-- 'The' Lord and Master Bitman On High, Master Of All
I like the new clauses, and the clarifications of older clauses. I expect to be releasing my new (and old, since I rarely used the, "...or...any later version" clause) under this license.
One of the interesting clarifications regards GPL derivatives. Formerly, a developer could not use the GPL as a basis for a new license. That meant, depending on interpretation, that all that software licensed with terms that said stuff like, "this software is released under the terms of the GPL, but with the following changes" may have been violating the FSF copyright on the GPL license itself. Version 3 specifically allows for this behavior.
Oh great, so now we don't get to have a Trusted Computing Base that's will be as secure as Vista or its successor? The Trusted Computing Platform has good uses and bad. A good use is protecting the operating system *you* installed from modification by malware - a chain of trust extending up from the hardware - as long as you are in ultimate control of the TCP, of course.
I'm not too clear about the anti-DCMA provisions in GPL3, but frankly it worries me that this global license is being worded to get around bits of unbalanced U.S. legislation.
GPL needs to protect the freedom of the source code and derived code to be seen and modified by others, when software based on it is distributed, for a global audience. No more and no less than that - certainly not to tie too closely with any one country's current legal system, or to impose ideological restrictions on the nature of the code that can or can't be written using it.
If you are looking for libraries that you can add to your own commercial projects with impunity, I suggest looking towards LGPL code
.o's for static linking)?
The LGPL is very unclear in many areas, and it hasn't been seriously vetted by the FSF's lawyers. What happens if you staticly link instead of dynamicly link (the FSF states that you may then provide the non-LGPL object files used for linking as well, but that's not clear from the license wording)? What if there's code in the header files (then it's no longer just the interface -- it's object code as well, and compilers will link that directly into your object files; now it's not sufficient to allow for dynamic linking or provide the
My company (which is, uh, rather large) has a formal policy -- only BSD/MIT licensed code. It's the only licenses that the lawyers feel comfortable with. The terminology is simple, and there's no question of whether or not it can be used without code release.
The GPLv3 was supposed to clarify things on the whole linking-to-libraries bit, but based on my reading of this first draft it has not. If anything, it's less clear for a layman.
In the simple case (dynamic linking, no functions defined in header files), then you're right -- all you have to do is publish changes to the library that you make. But one of the other caveats is that you must be able to use a later version of the LGPL library -- if that library substantially changes then you could run into problems again. Bleh.
Now all of this is only for the case where you provide executable or object code to outside parties. If all the code runs interally to your company (application service provider) then you can use GPL or LGPL code without issue since you never distribute the code externally. This is something that RMS wanted to address in v3, but, again, I don't see anything that does so. The only danger here is if at some point in the mysterious future you change business models and want to distribute a client-side application, you have to be damned sure that there's no code from a GPL or LGPL project that's snuck in (copy/paste, link, whatever) or else you'll violate the licensing.
"For one thing, it insists that software patents should either be made freely usable by everyone, or else not obtained at all. Whether you argue that this is a desirable sentiment or not, the bottom line is that this directive is beyond Stallman's legitimate power to make. The draft tries to decree a similar prohibition against DRM, which again, is beyond its power and will be found unenforcable."
It's enforcable (IMO -- IANAL) in that it doesn't prohibit you from doing either activity. What it *does* prohibit you from being able to use GPLv3'd software if you do the things they don't want you to do.
Note, I'm not taking a stance on whether or not this is a desirable position..
The GPL is (very clearly this iteration) a license, and the only stick that the GPL has is that all the rights it grants you to use/modify software go away if you do not follow its license proscriptions. Once breached, you can be sued for copyright infringement if you do not comply.
-- Have you ever imagined a world with no hypothetical situations?
Using libraries and such you don't always know which you'll want to use until you've tried several. For me, its also an iterative process as I'm learning the technology at the same time I'm looking into the available code. That means trying out lots of things. Since the commercial use arrangements are difficult to track down, I usually try to avoid the whole mess except in those rare occasions when I already KNOW the tool I need.
Recent example - I'm doing some custom work which uses VOIP as a backbone transport. IAX2 is, IMO, the better protocol. The open sourced project which produces the iaxclient.dll is built to make that easier. Figuring out just how to take advantage of that, however, is quite hard. Since the client is Windoze based, it was WAY easier to use the Microsoft RTC Client libraries and use SIP for the session management. I hate SIP compared to IAX2, and I'm working toward my own implimentation of the IAX2 protocol so I can use that -- but its a great case where some excellent technology is just unworkable (for me, and those like me) in its present form resulting from its open source roots. Its great tech, its functional, its a better design, the code is probably much better --- but it isn't really "finished" at all.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
...only to accept reasonable payment to allow it to be used as part of a closed solution in addition to remaining open for others to use under GPL.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
I think that for most users of the license, more important than the legal terms it self is an explanation of what the license is trying to accomplish. Such clarifications may also prevent the abuse of legal terms in court as the clarifications may be valuable in court as well as the clarifications indicate the intent of the wording used.
Why not add a Wiki style link system to explain what the different sections try to accomplish?
After all it is the users and the developers who needs to understand their rights and responsibilities, not lawyers.
Maybe I don't have a good feel for the community then. The success of open source, the availability of tools, and the usefulness of those tools are not diminished by the existance of proprietary software. Keep in mind, I'm talking about things like libraries which are only useful if the software is going to - by its nature - take part in an open standard.
In the example I gave, if I wanted to build something that "speaks" IAX2 instead of SIP, and to do so I want to have the legal right (which I would presumably have to pay for) to incorporate the existing work someone has done to build the iaxclient.dll itself, isn't that in general GOOD not BAD for the overall success of IAX2, Open Source in general, and the expansion of the platforms that use the open standard connection protocol itself?
I guess I just don't see how that's bad. Is it bad to make a career out of developing software? I'm all for support and living within the standards. 100% for it. I'd MUCH rather support a standard than not. Not everything is standardized, and not everything should be. Does that mean if I build something which is inherently proprietary that communicates openly with something that isn't, that I'm hurting the industry?
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
Gpl'd DRM is going to have to be real real good, because it won't have the protection of the DMCA as a crutch to hide behind. If any DRM program using this license doesn't have good code, it will be almost useless. This is going to result in all popular DRM programs (gpl'd) to be far superior on average than those that aren't, because those others have used the DMCA as a crutch for bad code.
I agree completely with what you're saying. If you improve the library or add to it -- you should be contributing that back and of course not making money on the work that's already been done for you. Stack libraries and protocols are a bit different though, no? The code you do has literally nothing to do with the protocol other than as a user of it. Is it the same?
Not trying to be an ass, or weasel out of something -- the gpl is what is and I agree with it. I just want to bring up that it often fails to meet this kind of need. By virtue of the poor way its addressed (individually) it prevents me even trying out many GPL'd tools because I don't want to invest the time to learn them only to later find out I can't use them.
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
I think another issue with previous versions of the GPL is addressed by requiring not just distribution of source code, but of anything necessary to make the code actually work, including encryption keys. Quoting from the draft:
"Complete Corresponding Source Code also includes any encryption or
authorization codes necessary to install and/or execute the source code of
the work, perhaps modified by you, in the recommended or principal context
of use, such that its functioning in all circumstances is identical to that
of the work, except as altered by your modifications. It also includes any
decryption codes necessary to access or unseal the work's output."
This is very important if treacherous computing schemes become widespread (that is, "trusted computing" where the owner of the computer cannot get code of his choosing to execute).
The reason is that if I write code and release it under the old GPL, company A can port it to a treacherous computing platform, compile it and sign the binary with a digital signature enabling it on the platform. Company A then distributes it with the source code. But when another user takes the code, modifies it, and compiles he is unable to sign it (unless he is also a big company with a TCP-enabling key) and is therefore unable to run it on the platform. Company A has therefore effectively stolen a user's right to further modify derivations of a GPL work.
The proposed changes stops Company A distributing a derivative work for a treacherous computing platform, unless they also publish their TC private key (which they can't). It doesn't stop Company A from using GPL code on a trusted computing platform, where anyone can use their own key to enable code to run on their own computer, so Company A does not need to give us their key.
"Can I just release the client (complete with full source code and under the GPL) while never releasing any bit of the server?"
Yes, this is one of the points they discussed before. The source sharing currently only applies to distribution of an application. Some people would like to see the source distributed for the server-hosted application, but that isn't likely to end up in GPL3, and it would only apply to your hypothetical MMORPG server software if you had used some other GPL3'd code to construct it.
If you wrote the software without using other code, it's yours to do whatever you wanted with, including modifying whatever license you release it with. If you believe modified hosted software should have its source shared, you can always release your own code under "the GPL3 plus an additional requirement that you share the source if you modify and host the application without distributing said application."
I think something like Mambo/Joomla! might be a good example of the type of code where such a license may have a strong effect. Personally I'm on the fence over whether people should be able to host modified GPL apps without sharing it back to the community. It would be nice if everyone saw the long-term benefits of sharing (net acceleration in code improvement, reduced redundancy of labor) but for the most part, I think that since the original source would still be available, someone could just clone the new feature pretty easily.
-jpowers
I love the header "Liberty or Death for the Program." Essentially, what it says is that the so-called "liberty" of the program (which is an absurd concept; a program isn't a thinking being and cannot have liberty) is more important than PEOPLE's liberties.