[Koha] Proposal To Switch Koha's License to GPLv3 and AGPLv3 or AGPLv3

[Chris Nighswonger]

Given that this discussion has been somewhat carried on through the
wiki (http://wiki.koha-community.org/wiki/General_Meeting,_June_2_2010),
I thought it would be good to keep it on list as well.

MJ posts the following argument to the above referenced agenda:

>  * AGPL does not prevent unfriendly vendor lock-in (still achievable through access control, particularly to the databases) while
> introducing onerous burdens on friendly hosters. http://lists.katipo.co.nz/pipermail/koha/2010-May/023816.html

This would probably be a good time to point out that 3.4 will have the
capability of doing a total db dump from within the staff client. This
should greatly mitigate the potential of an hostile vendor holding
data hostage. A vendor would have to deliberately remove or disable
this functionality to prevent a client from retrieving a full copy of
their database. (Of course any client who contracts with a vendor
*without* writing data protection assurances into their contract is
asking for trouble to start with.)

The matter of "onerous burdens on friendly hosters" has been addressed
in previous communications and so those arguments are not repeated
here.

Kind Regards,
Chris



On Mon, May 10, 2010 at 11:14 PM, Chris Nighswonger
<cnighswonger at foundations.edu> wrote:
> On Mon, May 10, 2010 at 9:54 PM, MJ Ray <mjr at phonecoop.coop> wrote:
>>
>> Christopher Nighswonger wrote:
>> > On Mon, May 10, 2010 at 9:47 AM, MJ Ray <mjr at phonecoop.coop> wrote:
>> > > [...] It might mean that everyone who sees a page is entitled to
>> > > the 227Mb source tarball.  Who wants to pay for those downloads?
>> >
>> > Github and the like provide very simple solutions to this problem both with ease
>> > of administration and ease of cost. AGPLv3 does not specify (see my previous
>> > response to Lars' post.) Besides, there are hosting plans available that provide
>> > unlimited bandwidth for very, very few $$$ per month if the FTP route was a
>> > necessity.
>>
>> Firstly, it seems unethical to impose even "very, very few $$$ per
>> month" of extra cost on charitable libraries.
>>
>
> In that case, github has a free offering... no $$$ per any time increment.
>
>>
>> Secondly, this leads to another of what I think is still one of the
>> Great Unknowns of AGPLv3: if you don't host the source alongside, must
>> the app go offline if it thinks the source has gone offline?
>>
>
> Where in the license does it say anything about hosting source
> alongside the application?
>
>>
>> There are so many of these lawyerbombs around AGPLv3 that I feel the
>> whole thing is best avoided by sticking with GPLv2, at least until
>> others have trod on some of the big ones, in the absence of any
>> pressing need to switch.
>>
>
> GPLvX was virtually untested in court until Progress Software Corp. v.
> MySQL began back in 2002. Apparently that line of reasoning did not
> stop the originators of Koha from selecting GPLv2+ when they released
> it in 2000.
>
>>
>> > <snip>
>> > > Fundamentally, AGPLv3 is based on an absurd idea that one can "ensure
>> > > cooperation with the community" (source: AGPLv3 preamble).  However,
>> > > cooperation by definition must be voluntary (source:
>> > > ICA.coop/coop/principles.html ) so legal compulsion is not cooperation.
>> [...]
>> > It appears that any form of licensing is an attempt to ensure
>> > cooperation of some sort among some people.  *All* licensing is, in
>> > fact, some form of coercion, period. The unfortunate fact of life is
>> > that there is somebody, somewhere who will do wrong even if you will
>> > not. It would be wonderful if it were otherwise. The we would not need
>> > licenses... or laws for that matter. [...]
>>
>> I don't see how that's true, unless "cooperation of some sort" means
>> something other than cooperation, such as mere trading.
>> Free software licensing is usually just setting out the terms of trade,
>> but AGPLv3's clod-handed attempts to force public sharing go beyond it.
>>
>
> View it how you will. In the case you propose, we are simply
> establishing an additional clarification to the terms of trade. The
> intent of GPL licenses is to preserve the open nature of code released
> as free and open source. Not only is it the right of the receiver to
> have access to the code, etc. but it is also the right of the author
> to have their intent that the code be free and open respected. The
> terms of the trade are such that in return for the right to use, etc.
> this code, you agree to release any changes you make to said parties.
>
> As I pointed out previously: All licensing agreements "force" some
> points. Just violate one and see how quickly it is en-forced. (Pun
> intended.)
>
> It actually seems to me that the matter of not forcing cooperation is
> in contradiction to the expressed desire to force cooperation in the
> matter of trademarks, etc. by de-listing, etc. Cooperation is, after
> all, cooperation.
>
>> Laws and licences do not prevent anyone doing wrong.  I'm sure all of
>> us are quite capable of breaching a copyright licence without much work.
>
> Quite right. If laws and licenses did that, we'd only need a few to
> fix all problems. What they do do is give recourse for redressing
> wrongs.
>
>> If anyone is hoping that adopting AGPLv3 will prevent bad people
>> refusing to share progress with the community, you are doomed to fail.
>
> I don't think anyone is under that illusion. As I said, "The
> unfortunate fact of life is that there is somebody, somewhere who will
> do wrong even if you will not."
>
>>
>> So what is the burning desire for AGPLv3?
>>
>
> To put in place a mechanism for redressing violations of terms of
> license. We can chatter away about how a company reneged on its verbal
> obligations to contribute code back to the community,but we have
> nothing to fall back upon with any (even hopeful) force of law. It is
> not a panacea for all ills, only another hopeful obstacle in the path
> of misbehavior.
>
>> > > So may we postpone the rest of this discussion to post-3.2.0?
>> >
>> > As I stated in my original proposal: We are already very active atm, and now is
>> > the time to at least begin discussing this change.
>>
>> I am disappointed by this desire to press ahead with holding a
>> discussion of such a complex topic at such a busy time.  It will limit
>> participation and likely leave the discussion incomplete.
>
> Participation will only be limited if we want it to. No one involved
> in this project is not busy. If we believe that we have a good thing
> going in Koha, we must make the time to do the not so nice parts of
> the project too.
>
> Kind Regards,
> Chris
>