[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: DFSG-freeness of any license that fixes the ASP loophole



On Sun, Nov 11, 2007 at 05:55:53AM +0530, Shriramana Sharma wrote:
> Would this corrected clause then be DFSG-compliant? Added text marked with 
> carets.
>
> When you make it possible for this work or derivative works to be directly 
> or indirectly used over a network, you must prominently provide information 
> as to how to obtain the complete source code for such work ^^at no more 
> charge than the cost of transfer^^:
> * on the same interface that is provided for the usage of such work
> * or if the above is not possible, on another publicly accessible 
> ^^network^^ location closely related to your activities

Better, but I'm still not sure how that would work out in practice: of
someone decides they "need" to set up a separate server with high
bandwidth in order to service the (probably fairly small) number of
source downloads, they could probably present the cost of this as
being non-trivial. I'm sure this is why the AGPL puts the cost burden
back onto the operator by insisting the source is made available
without charge.

Another problem with the proposed text is the use of works "directly
or indirectly". This gets us straight into the issue of "how far down
the stack" the licence reaches: you could have hundreds of different
programs that are "indirectly" used by anyone accessing the site, from
the kernel upwards.

Also, this licence places conditions merely on *use* of the unmodified
software ("When you make it possible for *this work*..."), which is a
major departure from the principles of the GPL. Even the AGPL places
the condition on those who *modify* the software, following the GPL
practice of only placing obligations on modifiers and distributors,
rather than users.

I begin to understand the thought processes that have led to the
current draft of the AGPL. I suspect the AGPL is heading towards being
the best way currently available of doing this. What I'm sceptical
about is whether it needs to be, or should be, done in the first
place.

John



Reply to: