Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Mon, Jun 07, 2004 at 11:32:22PM +0200, Francesco Poli wrote: > That would bring me to the conclusion that I must accept the GPL in > order to make a copy of a GPL'd work. > > See for example GPL#4: > > [ 4. You may not copy, modify, sublicense, or distribute the Program > [ except as expressly provided under this License. > > > On the other hand GPL#5 itself says I'm required to accept the GPL only > in order to distribute or modify... > > Could you explain how's that possible? http://www4.law.cornell.edu/uscode/17/117.html Interpret as you will. I interpret this to mean that I can make copies of and/or modifications to a program for personal use, and that that act is not defined as 'copying' for purposes of copyright. Copyright only becomes involved when I decide to distribute copies or modifications to a third party, at which time I will have to accept the GPL and abide by its terms. --Adam -- Adam McKenna <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Sun, 6 Jun 2004 10:45:59 -0700 Adam McKenna wrote: > On Sun, Jun 06, 2004 at 11:08:50AM +0200, Francesco Poli wrote: > > On Fri, 4 Jun 2004 23:25:18 -0700 Adam McKenna wrote: > > > > > the reason you can copy a file > > > which has been released under the GPL without accepting the GPL is > > > because you are explicitly granted that right by the GPL. > > > > I don't think so: you are not granted any right by a license, unless > > you accept the license itself. > > Hence, I don't see how could the GPL grant any right to make copies > > to someone who is not willing to accept the GPL itself. > > Please read the GPL, specifically sections 5 and 6. You only have to > accept the license if you wish to modify or distribute the work. Ooops. I apologize for not having recalled this. > > I'm not going to participate in any more semantic arguments or legal > 101 on what the definitions of the words 'distribute' or 'copy' are. Could you please provide a good reference to some clear definition? I hope I know the difference, but I'd better check anyway... :p > If you are confused about copyrights in general, please go to > http://www.loc.gov/copyright and read the text entitled "copyright > basics". Well, *now* I'm a bit confused. At http://www.copyright.gov/circs/circ1.html I read the following: é[ Section 106 of the 1976 Copyright Act generally gives the owner of [ copyright the exclusive right to do and to authorize others to do the [ following: [ [* To reproduce the work in copies or phonorecords; ... AFAIK the reasoning behind GPL#5 is that only the license gives me some rights, hence I must have implicitly accepted it, if I exercise those rights. U.S. Copyright law (and many other copyright laws, I suppose) lists the right to make copies as one of the copyright owner's exclusive rights. That would bring me to the conclusion that I must accept the GPL in order to make a copy of a GPL'd work. See for example GPL#4: [ 4. You may not copy, modify, sublicense, or distribute the Program [ except as expressly provided under this License. On the other hand GPL#5 itself says I'm required to accept the GPL only in order to distribute or modify... Could you explain how's that possible? -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgpIU94hxpjZ4.pgp Description: PGP signature
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Sun, Jun 06, 2004 at 11:08:50AM +0200, Francesco Poli wrote: > On Fri, 4 Jun 2004 23:25:18 -0700 Adam McKenna wrote: > > > the reason you can copy a file > > which has been released under the GPL without accepting the GPL is > > because you are explicitly granted that right by the GPL. > > I don't think so: you are not granted any right by a license, unless you > accept the license itself. > Hence, I don't see how could the GPL grant any right to make copies to > someone who is not willing to accept the GPL itself. Please read the GPL, specifically sections 5 and 6. You only have to accept the license if you wish to modify or distribute the work. I'm not going to participate in any more semantic arguments or legal 101 on what the definitions of the words 'distribute' or 'copy' are. If you are confused about copyrights in general, please go to http://www.loc.gov/copyright and read the text entitled "copyright basics". --Adam -- Adam McKenna <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Fri, 4 Jun 2004 11:46:51 +0200 Bernhard R. Link wrote: > * Henning Makholm <[EMAIL PROTECTED]> [040602 16:42]: > > If you want to *download* the sofware, then you'd better do it by > > the GPL's terms.[...] > > If you log on some computer and make a copy there and transmit it to > you (like ssh'ing into a solaris box and copying /bin/true), this may > be true. > > But normally someone set up a computer do make a copy and sent it to > me, if I request it. As when someone makes copies of a CD and sends > them to me, when I send him a postcard. I agree. Henning, consider how, say, the HTTP protocol works: an HTTP server serves some content upon my *request*. I send a GET /path/lbreakout2_2.2.2-1woody1_i386.deb to the web server, it processes my request (this choice of word is not a coincidence!), determines if I have permission to get that resource and serves a response to me. The response may contain a 403 forbidden error code, or a copy of $DOCUMENTROOT/path/lbreakout2_2.2.2-1woody1_i386.deb Is that any different (from a legal point of view) from visiting one friend of mine, seeing that she has a great game installed on her Debian Woody machine, *asking* her "May I have a copy of the deb package?" and going back home with a floppy disk containing lbreakout2_2.2.2-1woody1_i386.deb? My friend is the one who is making the copy. In the HTTP example, the copy is being made by server, that is (from a legal standpoint) by the person who put online the file. In the HTTP example, you can also consider the case in which the response contains dynamically-generated content: that content may be a derived work of something (a template, perhaps...). Do you claim that *I* am the person who generated that content? I don't even necessarily have access to the template! I think that derivation is made by the server: after all, they are called *server-side* scripting languages... -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgprcWqdBekmw.pgp Description: PGP signature
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Fri, 4 Jun 2004 23:25:18 -0700 Adam McKenna wrote: > the reason you can copy a file > which has been released under the GPL without accepting the GPL is > because you are explicitly granted that right by the GPL. I don't think so: you are not granted any right by a license, unless you accept the license itself. Hence, I don't see how could the GPL grant any right to make copies to someone who is not willing to accept the GPL itself. -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgp3AM4y2lpZZ.pgp Description: PGP signature
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Fri, Jun 04, 2004 at 01:09:00PM +0100, Henning Makholm wrote: > I'll have to retract my assertation that one has to accept the GPL > before downloading a work covered by it. In most jurisdictions that I > know of, people by default have the right to create copies of most > copyrighted works for their own personal use, even without permission > from the copyright holder. This is what gets the downloader and the > button-pusher off the hook, rather than a fiction that they are not > really creating copies. Your semantic arguments are missing the point; the reason you can copy a file which has been released under the GPL without accepting the GPL is because you are explicitly granted that right by the GPL. --Adam
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Jun 3, 2004, at 20:27, Henning Makholm wrote: But that is actually irrelevant. The relevant part is that no matter where you consider the copy to be "made", *I* am the one who is causing the computers (my own and the server) to make a copy at that particular time and place. So then the server operator is guilty of at worst contributory infringement? Someone please go fill in RIAA...
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Scripsit Glenn Maynard <[EMAIL PROTECTED]> > If I make photocopies of a book and put them on a shelf with a "Free!" > sign, and you then take a copy, I'm the one who made the copy available, > and the one needing permission from the copyright holder. The thing that needs permission is not making the copy *available*; it's making the copy at all in the first place. > It don't see how it's any different if I set up a printer with a > button saying "Push for free book!". "I didn't actually make the > copy; he's the one who pushed the button!" isn't something I'd try > in court. No, not if you were accused of contributory infringement by making it too easy for third parties to make copies of a specific work that they are not allowed to make copies of. The fact that *you* are in trouble, however, does not in itself stop the person pushing the button from being in trouble *too* if he knew that said button would cause the machine to manufacture a copy that he did not have any right to manufacture. I'll have to retract my assertation that one has to accept the GPL before downloading a work covered by it. In most jurisdictions that I know of, people by default have the right to create copies of most copyrighted works for their own personal use, even without permission from the copyright holder. This is what gets the downloader and the button-pusher off the hook, rather than a fiction that they are not really creating copies. -- Henning Makholm "Hør, hvad er det egentlig der ikke kan blive ved med at gå?"
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
* Henning Makholm <[EMAIL PROTECTED]> [040602 16:42]: > If you want to *download* the sofware, then you'd better do it by the > GPL's terms. "Downloading" implies that you are instructing some > computer to make create a copy of the Work on your hard drive. Because > computers, legally speaking, do not *do* anything by themselves, *you* > are the one who are creating the copy on your hard drive. And creating > a copy is smack in the middle of the copyright holder's legal > monopoly. If you log on some computer and make a copy there and transmit it to you (like ssh'ing into a solaris box and copying /bin/true), this may be true. But normally someone set up a computer do make a copy and sent it to me, if I request it. As when someone makes copies of a CD and sends them to me, when I send him a postcard. Now when I send a postcard somewhere, there those are scanned and when they are a printout of the request-formular, the CD is burned, copied the address field on an envelope and sends it out. Would I still be the one copying in your understanding of the situation? what if the form says: "might be processed without human intervention"? Hochachtungsvoll, Bernhard R. Link -- Sendmail is like emacs: A nice operating system, but missing an editor and a MTA.
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
> Scripsit Raul Miller <[EMAIL PROTECTED]> > > Except, the copy is being made on the server. On Fri, Jun 04, 2004 at 01:27:00AM +0100, Henning Makholm wrote: > When I download something, the copy is being made on a hard disk that > sits in a box below my desk. Current is being modulated and passed > through a coil, which causes an area of the disk surface to be made > into a copy of the work. Fundamentally, the way computers work, is by making copies of everything they're working on. There's a copy on disk, it gets copied into the hard disk controller, then into memory, then into L2 cache and L1 cache through the cpu, resulting in more copies being made in the ethernet controller, then out onto the network. Many other copies are made (at least one at each router) before it finally winds up on the user's machine where a similar process results in a copy being made on disk. This is all really fundamental and basic. But the thing is, almost all those copies are transient -- they're destroyed shortly after they're made. From a non-technical point of view, the concept of "transmission" is more relevant than the concept of "copying". It's only at server where there's an original which lasts from before the copying started till after it's concluded. The server makes another copy by transmitting it to the client machine. > But that is actually irrelevant. The relevant part is that no matter > where you consider the copy to be "made", *I* am the one who is > causing the computers (my own and the server) to make a copy at that > particular time and place. You're neglecting the person running the server -- if they want to stop you from making a copy, they're free to do so. In essence, putting something up on a publically reachable server is an act of distribution. If this were not the case, everybody on the internet would be violating copyright every time they downloaded a page which doesn't have an appropriate grant of copyright on it. If I accepted your logic, I'd think that people listening to the radio or watching TV are liable for copyright violation, and that the people transmitting that content are uninvolved in making these copies. > When I knowingly cause machines to make a copy for me, I am doing the > exact thing that copyright regulates. Are you suggesting that copyright law is preventing you from making a copy of nothing? You don't have a copy until after you've received it. -- Raul
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Fri, Jun 04, 2004 at 01:27:00AM +0100, Henning Makholm wrote: > But that is actually irrelevant. The relevant part is that no matter > where you consider the copy to be "made", *I* am the one who is > causing the computers (my own and the server) to make a copy at that > particular time and place. If I make photocopies of a book and put them on a shelf with a "Free!" sign, and you then take a copy, I'm the one who made the copy available, and the one needing permission from the copyright holder. It don't see how it's any different if I set up a printer with a button saying "Push for free book!". "I didn't actually make the copy; he's the one who pushed the button!" isn't something I'd try in court. I've only seen free licenses written in terms of the sender: "here are the terms under which you can give copies to others", and never "here are the terms under which you can receive copies from others". Also, you often can't view even free licenses until *after* receiving the work. /usr/share/doc/foo/copyright is part of Debian packages; I can't read it until I already have it (unless I bend over backwards, such as by finding the license elsewhere, apart from the package). -- Glenn Maynard
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Henning Makholm wrote: When I download something, the copy is being made on a hard disk that sits in a box below my desk. Current is being modulated and passed through a coil, which causes an area of the disk surface to be made into a copy of the work. But that is actually irrelevant. The relevant part is that no matter where you consider the copy to be "made", *I* am the one who is causing the computers (my own and the server) to make a copy at that particular time and place. When I knowingly cause machines to make a copy for me, I am doing the exact thing that copyright regulates. An interesting question, however, is if you are knowingly causing the computer(s) to copy a specific copyrighted work, considering that you don't know what you're downloading until you've already done it. As the RIAA lawsuits have shown, many entities consider the act of downloading to be distribution of the work by the server, rather than the client. Under this interpretation, downloading a GPLed work is no different than being given it on a CD; the owner of the server has accepted the GPL and is distributing the software under its terms, while the owner of the client computer need not accept the GPL unless he wants to. If he doesn't, he may still use the software, just not copy or modify it. -- Lewis Jardine IANAL IANADD
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Scripsit Raul Miller <[EMAIL PROTECTED]> > On Wed, Jun 02, 2004 at 12:52:37PM +0100, Henning Makholm wrote: > > If you want to *download* the sofware, then you'd better do it by the > > GPL's terms. "Downloading" implies that you are instructing some > > computer to make create a copy of the Work on your hard drive. > Except, the copy is being made on the server. When I download something, the copy is being made on a hard disk that sits in a box below my desk. Current is being modulated and passed through a coil, which causes an area of the disk surface to be made into a copy of the work. But that is actually irrelevant. The relevant part is that no matter where you consider the copy to be "made", *I* am the one who is causing the computers (my own and the server) to make a copy at that particular time and place. When I knowingly cause machines to make a copy for me, I am doing the exact thing that copyright regulates. -- Henning Makholm "Den nyttige hjemmedatamat er og forbliver en myte. Generelt kan der ikke peges på databehandlingsopgaver af en sådan størrelsesorden og af en karaktér, som berettiger forestillingerne om den nye hjemme- og husholdningsteknologi."
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Wed, 2 Jun 2004 16:27:28 +0100 Edmund GRIMLEY EVANS wrote: > It seems to me that the person who puts something on line is usually > regarded as the person doing the copying. That is indeed what I have thought till a few days ago... And it's still the most reasonable interpretation I can think of... -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgp7fLBvLRORP.pgp Description: PGP signature
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Wed, Jun 02, 2004 at 12:52:37PM +0100, Henning Makholm wrote: > Scripsit Francesco Poli <[EMAIL PROTECTED]> > > > That is: I'm not required to accept the GPL if I simply want to download > > (and install and use) a GPL'd piece of software. > > If you want to *download* the sofware, then you'd better do it by the > GPL's terms. This is absolutely false, and I believe the GPL itself contains language that refutes this. --Adam
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On 02 Jun 2004 12:52:37 +0100 Henning Makholm wrote: > If you want to *download* the sofware, then you'd better do it by the > GPL's terms. "Downloading" implies that you are instructing some > computer to make create a copy of the Work on your hard drive. Thus a downloaded package (e.g. from Debian mirror network) is legally different from a package obtained on a CD (e.g. provided by a Debian CD vendor). Is that right? If this is the case, I see it as a bit surprising... Or does the copy from CD to HD (that I must perform in order to install the package) trigger copyright laws too? In that case, I cannot install a package without accepting its license: that sounds strange. The user can use the software without accepting the license, but the sysadmin must have accepted it, otherwise he could not install it... -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgpVc9JBA9rma.pgp Description: PGP signature
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Henning Makholm <[EMAIL PROTECTED]>: > If you want to *download* the sofware, then you'd better do it by the > GPL's terms. "Downloading" implies that you are instructing some > computer to make create a copy of the Work on your hard drive. Because > computers, legally speaking, do not *do* anything by themselves, *you* > are the one who are creating the copy on your hard drive. And creating > a copy is smack in the middle of the copyright holder's legal > monopoly. It seems to me that the person who puts something on line is usually regarded as the person doing the copying. I don't know of any legal basis for this point of view, but it makes more practical sense if you think of search engines, Usenet, other situations where copying happens to all intents and purposes automatically once something has been put onto a server, and in general that the person who puts something on line has had a chance to examine and decide whether they are allowed to do so, whereas the person who starts to download a file has no way of knowing what they getting.
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Scripsit Francesco Poli <[EMAIL PROTECTED]> > > That is: I'm not required to accept the GPL if I simply want to download > > (and install and use) a GPL'd piece of software. On Wed, Jun 02, 2004 at 12:52:37PM +0100, Henning Makholm wrote: > If you want to *download* the sofware, then you'd better do it by the > GPL's terms. "Downloading" implies that you are instructing some > computer to make create a copy of the Work on your hard drive. Because > computers, legally speaking, do not *do* anything by themselves, *you* > are the one who are creating the copy on your hard drive. And creating > a copy is smack in the middle of the copyright holder's legal > monopoly. Except, the copy is being made on the server. It's true that the copy is being stored on the client, but it seems meaningless to make a distinction between that copy and the copy on the wire. -- Raul
Re: You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
Scripsit Francesco Poli <[EMAIL PROTECTED]> > That is: I'm not required to accept the GPL if I simply want to download > (and install and use) a GPL'd piece of software. If you want to *download* the sofware, then you'd better do it by the GPL's terms. "Downloading" implies that you are instructing some computer to make create a copy of the Work on your hard drive. Because computers, legally speaking, do not *do* anything by themselves, *you* are the one who are creating the copy on your hard drive. And creating a copy is smack in the middle of the copyright holder's legal monopoly. -- Henning Makholm "Det er jo svært at vide noget når man ikke ved det, ikke?"
You can't get a copy unless you accept the GPL [was: Re: libkrb53 - odd license term]
On Tue, 1 Jun 2004 20:35:09 +1000 Matthew Palmer wrote: > I guess, though, in a way > it's another wording of the GPL's "you can't legally get a copy except > by the permissions we've granted here, so we'll take it as read you > accept this licence" clause. Wait, wait! I'm not sure I understand what you're saying here (maybe it's my english-language-parser's fault... :p ) Do you mean I have to accept the GPL in order to *get* a copy of a GPL'd work? I suppose you're referring to the following GPL-2 clause: ] 5. You are not required to accept this License, since you have not ] signed it. However, nothing else grants you permission to modify or ] distribute the Program or its derivative works. These actions are ] prohibited by law if you do not accept this License. Therefore, by ] modifying or distributing the Program (or any work based on the ] Program), you indicate your acceptance of this License to do so, and ] all its terms and conditions for copying, distributing or modifying ] the Program or works based on it. I have always interpreted it as covering actions such as distributing, modifying... but not getting copies! Am I wrong? That is: I'm not required to accept the GPL if I simply want to download (and install and use) a GPL'd piece of software. The person who distributes the work to me must have accepted the GPL, but I'm not required to. License acceptance is instead necessary when *I* want to modify or distribute the work (or a modified version of the work). Is that right? -- | GnuPG Key ID = DD6DFCF4 | You're compiling a program Francesco |Key fingerprint = | and, all of a sudden, boom! Poli| C979 F34B 27CE 5CD8 DC12 | -- from APT HOWTO, | 31B5 78F4 279B DD6D FCF4 | version 1.8.0 pgpxSf5UbHKg7.pgp Description: PGP signature