But you do (because I say so ; ), and guess what? It makes no
difference: your review is not a derivative work anymore than it was before.
Exactly. So what's your difficulty in downloading the Tivo code, reading
it and re-using it in your own projects, on your other devices? How is
the missing signing key preventing you from doing any of that?
Someone buying the book may be free to read it anywhere but if they
insist on reading it at your table you may sensibly require they bring a
copy of your review with them (to prove their genuine interest ; ).
Failure to comply only means they have to read the book someplace else.
Can they read the book? Sure. Can they read it at your table? Only if
you choose to allow them.
GPLv2 guarantees that the book remains readable. It does not grant you
(doesn't even try) the right to execute a modified copy on any
particular piece of hardware. Your kernel is perfectly functional on any
platform that supports it - it just so happens that the Tivo device does
not support it.
So is a digital signature. Again, are you arguing the digital signature
is a derivative work?
I guess you'll have to explain again because copyright law and its
definition of derivative works are the things that make the GPL work:
"0.This License applies to any program or other work which contains a
notice placed by the copyright holder saying it may be distributed under
the terms of this General Public License. The "Program", below, 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".
What I can't find though is any reference to "integral parts" or your
taken-for-granted right to run a modified copy of the program on the
same device used for distribution (or any mention of functionality at
all for that matter). Actually: "Activities other than copying,
distribution and modification are not covered by this License; they are
outside its scope".
But you're missing the whole point: the rules are the same, nothing
changes! You are drawing an artificial distinction between the two cases
and focusing on aggregation, which is totally irrelevant: either the
digital signature is a derivative work or it isn't, and in either case
its distribution method makes no difference in the world.
The reason I brought up the separate-signature example is to illustrate
just how ridiculous is to think of the signature keys as source files:
you can implement an equivalent DRM system without ever modifying the
kernel blob. Only difference is the channel used for signature
distribution, and I hope you won't argue that mere aggregation changes
its nature.
---
fm
-
