Re: First draft of AGPL v3
On Wed, 13 Jun 2007 16:13:30 +0200, Gervase Markham [EMAIL PROTECTED] wrote: But surely the entire point in question is whether presenting the UI to someone across the network is conveying or not? GPLv3 says it isn't, AGPL says it is. Perhaps it would be better (in respect of this particular question) if the AGPL extra clause said simply: Notwithstanding any other provision of this License, if a user interacts with the program remotely through a computer network, then that is considered an act of conveying. (i.e. change the definition of conveying in section 0.) That still leaves open the question of what it means to interact with the progam. How many steps removed can it be? If I interact with an ATM that communicates with a server that requrests data from a web service that uses a database that is covered by the AGPL ... am I interacting with it? Is it being coveyed? What if the AGPL is in the web service? It still has no UI itself. Is the ATM customer interacting with that? If not, the AGPL would be trivial to cicumvent by splitting the covered part into a separate component without a UI and interacting with it remotely. /L -- Lasse R. Nielsen - [EMAIL PROTECTED] 'Faith without judgement merely degrades the spirit divine' Reproduction of this message, or parts thereof, is allowed if proper attribution is given. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: First draft of AGPL v3
On Thu, 14 Jun 2007 09:43:23 +0200 Lasse Reichstein Nielsen wrote: On Wed, 13 Jun 2007 16:13:30 +0200, Gervase Markham [EMAIL PROTECTED] wrote: [...] Perhaps it would be better (in respect of this particular question) if the AGPL extra clause said simply: Notwithstanding any other provision of this License, if a user interacts with the program remotely through a computer network, then that is considered an act of conveying. (i.e. change the definition of conveying in section 0.) That still leaves open the question of what it means to interact with the progam. [...] Exactly. -- http://frx.netsons.org/doc/nanodocs/testing_workstation_install.html Need to read a Debian testing installation walk-through? . Francesco Poli . GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgphJ0Iohu1mM.pgp Description: PGP signature
Re: First draft of AGPL v3
On Thu, 14 Jun 2007 01:52:34 +0200 Josselin Mouette wrote: [...] This is why I propose that alternate wording - which surely deserves improvement - which makes clearer that this doesn't have to be for free or on a network server; just that accessing the software must give access to the source. This should cope with any accusations of this clause implying an additional cost. But this alternate wording would make the clause even more restrictive, as it would extend the requirement to cases where the modified version of the Program does *not* support remote interaction through a computer network (whatever that may mean...). Or did I misunderstand your intended meaning? -- http://frx.netsons.org/doc/nanodocs/testing_workstation_install.html Need to read a Debian testing installation walk-through? . Francesco Poli . GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpcI9uy0XVkJ.pgp Description: PGP signature
Re: First draft of AGPL v3
Francesco Poli wrote: The restriction in the GPL is about the act of conveying copies of the work. The restriction in the AGPL is about *using* the modified work: there's a cost associated with *use*... But surely the entire point in question is whether presenting the UI to someone across the network is conveying or not? GPLv3 says it isn't, AGPL says it is. Perhaps it would be better (in respect of this particular question) if the AGPL extra clause said simply: Notwithstanding any other provision of this License, if a user interacts with the program remotely through a computer network, then that is considered an act of conveying. (i.e. change the definition of conveying in section 0.) This is ultimately a use restriction (from the point of view of whoever runs the modified version of the Program) What does it prevent you using the Program for? If the source doesn't fit in the server the modified version runs on (think of small embedded systems, for instance), I have to set up a separate server to provide source to remote users. But that doesn't prevent you *using* the modified version in your small embedded system. It might perhaps be inconvenient, but as Anthony Towns (I believe) said recently, not every obligation a license puts upon you is a cost. In order to *run* the modified version of the Program! Or to convey it, depending on your point of view :-) As I said above, the GPL restricts the act of conveying, the AGPL also restricts the use of modified versions of the Program. Well, those who would use the Affero GPL would see the use you are referring to as a form of conveying. Gerv -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: First draft of AGPL v3
Le mercredi 13 juin 2007 à 01:27 +0200, Francesco Poli a écrit : Bad: use restriction, with a cost associated to it A better wording would then be: ... an opportunity to receive the Corresponding Source under the same conditions as those of using the Program. I'm sorry, but I really cannot understand what you mean. Could you elaborate, please? The current wording is way too specific, and requires providing the source at no charge on a network server. In the event where the user pays for using the service, the source code availability would then have to be included in the price. This would also fix the remote X11 display case that was imagined earlier, as it makes clear that accessing the software must give access to the source. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée
Re: First draft of AGPL v3
Josselin Mouette [EMAIL PROTECTED] writes: The current wording is way too specific, and requires providing the source at no charge on a network server. In the event where the user pays for using the service, the source code availability would then have to be included in the price. That seems to be intentional (on the part of the AGPL authors), no? -- \ My house is on the median strip of a highway. You don't really | `\notice, except I have to leave the driveway doing 60 MPH. -- | _o__)Steven Wright | Ben Finney -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: First draft of AGPL v3
Le jeudi 14 juin 2007 à 09:36 +1000, Ben Finney a écrit : Josselin Mouette [EMAIL PROTECTED] writes: The current wording is way too specific, and requires providing the source at no charge on a network server. In the event where the user pays for using the service, the source code availability would then have to be included in the price. That seems to be intentional (on the part of the AGPL authors), no? By re-reading myself, I seem to have been even more confusing. The current AGPL requires to put the source code for free on a network server, which is too specific. This is why I propose that alternate wording - which surely deserves improvement - which makes clearer that this doesn't have to be for free or on a network server; just that accessing the software must give access to the source. This should cope with any accusations of this clause implying an additional cost. (I won't try to submit this change on their website, because last time I tried it was broken.) -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée
Re: First draft of AGPL v3
Francesco Poli wrote: Bad: no clear definition of remote users The term user is not clearly defined. Is your point that it is impossible to clearly define, or do you have alternative language? Do you know how the corresponding clause in the current Affero license has historically been interpreted? This ambiguity is really problematic, as it implies that there's no clear way to tell whether a modified version supports remote interaction, and hence there's no clear way to tell whether it is subject to the restriction specified by this section. It's not that bad. If I turn some AGPLed code into a local graphics-editing application which has no network capabilities, it's fairly clear that it doesn't apply. But then, what happens if I access that desktop using remote X? Hmm... Let's say the clause instead said that anyone who interacts with the work had to get access to the corresponding source, full stop (no network need be involved). Would that be less ambiguous, I wonder? (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to copy the Corresponding Source from a network server at no charge. Bad: use restriction, with a cost associated to it This restriction compels whoever runs the modified version of the Program to accommodate the source code on the server or, alternatively, to set up and maintain a separate network server to provide source code: this may be a significant cost in some cases. I don't understand this argument. Having to provide CDs of source or fulfil the terms of a written offer is also a significant cost, but no-one thinks that makes the GPL non-free. This is ultimately a use restriction (from the point of view of whoever runs the modified version of the Program) What does it prevent you using the Program for? and effectively forbids private use of the modified version on a publicly accessible server. Well, it forbids public use of the modified version on a publicly accessible server. :-) But of course it does - that's the point. But then the GPL forbids giving someone the use of the modified version via giving them a copy without handing them the source code at the same time. That's not a use restriction. The AGPL clearly passes the Desert Island test (and the Tentacles of Evil test). I'm not sure the current wording of the Dissident test had this situation in mind, but I think a good argument could be made that it passes. (Incidentally, what part of the DFSG is the Dissident test supposed to help test against?) Gerv -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: First draft of AGPL v3
[EMAIL PROTECTED] wrote: (Incidentally, what part of the DFSG is the Dissident test supposed to help test against?) The imaginary clause. -- ciao, Marco -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: First draft of AGPL v3
Le lundi 11 juin 2007 à 23:57 +0200, Francesco Poli a écrit : (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to copy the Corresponding Source from a network server at no charge. Bad: use restriction, with a cost associated to it A better wording would then be: ... an opportunity to receive the Corresponding Source under the same conditions as those of using the Program. -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée
Re: First draft of AGPL v3
On Tue, 12 Jun 2007 22:24:43 +0200 Josselin Mouette wrote: Le lundi 11 juin 2007 à 23:57 +0200, Francesco Poli a écrit : (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to copy the Corresponding Source from a network server at no charge. Bad: use restriction, with a cost associated to it A better wording would then be: ... an opportunity to receive the Corresponding Source under the same conditions as those of using the Program. I'm sorry, but I really cannot understand what you mean. Could you elaborate, please? -- http://frx.netsons.org/doc/nanodocs/testing_workstation_install.html Need to read a Debian testing installation walk-through? . Francesco Poli . GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpVPgQANBkyW.pgp Description: PGP signature
Re: First draft of AGPL v3
On Tue, 12 Jun 2007 11:33:44 +0100 Gervase Markham wrote: Francesco Poli wrote: Bad: no clear definition of remote users The term user is not clearly defined. Is your point that it is impossible to clearly define, or do you have alternative language? I cannot have *alternative* language, since there currently is *no* language to define the term user at all... ;-) Seriously: I think it's really hard to reasonably define what a remote user is or should be. I don't have a proposed definition, because everytime I try to think of one, I find myself trapped in blurred boundaries and don't know where to draw the line (or how to get back home!). Do you know how the corresponding clause in the current Affero license has historically been interpreted? No idea... This ambiguity is really problematic, as it implies that there's no clear way to tell whether a modified version supports remote interaction, and hence there's no clear way to tell whether it is subject to the restriction specified by this section. It's not that bad. If I turn some AGPLed code into a local graphics-editing application which has no network capabilities, it's fairly clear that it doesn't apply. But then, what happens if I access that desktop using remote X? Hmm... Indeed. Moreover, what if I turn the AGPLed code into an OS kernel? Let's say the clause instead said that anyone who interacts with the work had to get access to the corresponding source, full stop (no network need be involved). Would that be less ambiguous, I wonder? Less ambiguous, maybe, but worse for sure, since it would extend the restriction to any use of the modified program. And all the issues with kiosks, pay-toll machines, arcade games, and so forth, would come back (as for the AfferoGPL v1). (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to copy the Corresponding Source from a network server at no charge. Bad: use restriction, with a cost associated to it This restriction compels whoever runs the modified version of the Program to accommodate the source code on the server or, alternatively, to set up and maintain a separate network server to provide source code: this may be a significant cost in some cases. I don't understand this argument. Having to provide CDs of source or fulfil the terms of a written offer is also a significant cost, but no-one thinks that makes the GPL non-free. The restriction in the GPL is about the act of conveying copies of the work. The restriction in the AGPL is about *using* the modified work: there's a cost associated with *use*... This is ultimately a use restriction (from the point of view of whoever runs the modified version of the Program) What does it prevent you using the Program for? If the source doesn't fit in the server the modified version runs on (think of small embedded systems, for instance), I have to set up a separate server to provide source to remote users. In order to *run* the modified version of the Program! If I cannot afford setting up a separate server, I cannot use the modified version on a network! and effectively forbids private use of the modified version on a publicly accessible server. Well, it forbids public use of the modified version on a publicly accessible server. :-) But of course it does - that's the point. But then the GPL forbids giving someone the use of the modified version via giving them a copy without handing them the source code at the same time. That's not a use restriction. As I said above, the GPL restricts the act of conveying, the AGPL also restricts the use of modified versions of the Program. -- http://frx.netsons.org/doc/nanodocs/testing_workstation_install.html Need to read a Debian testing installation walk-through? . Francesco Poli . GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpwdUdNDSuLY.pgp Description: PGP signature