:: Re: [Dng] system scriptinng languag…
Αρχική Σελίδα
Delete this message
Reply to this message
Συντάκτης: T.J. Duchene
Ημερομηνία:  
Προς: Jude Nelson
Υ/ο: dng
Αντικείμενο: Re: [Dng] system scriptinng language.

On 12/23/2014 11:44 PM, Jude Nelson wrote:
> Hi T.J., sorry this is late--this thread got lost in my inbox.
>
> Thank you for your feedback regarding GPLv3. My reason for choosing
> it was because it explicitly grants an irrevocable patent license to
> downstream coders, whereas GPLv2 does not. I would think this would
> make it more attractive to them, since there's no way I or my employer
> could sue them later for patent infringement. I realize that this
> license (and as far as I know, no license) can protect me against
> patent suits filed against me.
>

You're certainly welcome. While my opinion should not necessarily
matter, I do appreciate the acceptance.

My point was that your code, licensed as GPL3 doesn't do you as an 
individual any real favors, unless you are personally incorporating your 
patents into the code and personally granting a license to that patent 
to everyone.  Any patents that exist already in the code are not 
necessarily protected, because 9 chances out of 10 the upstream coders 
did not have authority to grant a license.  The GPL3's patent clauses 
are really only useful to patent holders who want to incorporate their 
patents into the code.  Just because something is licensed GPL or 
otherwise, it does not mean that it is unencumbered.    That is why the 
OIN is very handy, even though we wish it was unnecessary.


I can see why you would use it if you are writing code for your employer
to be opened, but then your employer is the one granting the license and
not yourself.


> I don't think you're being too harsh or overcritical, but I would like
> to hear your thoughts on the specific clauses you oppose beyond the
> "additional terms" section. I read that section again, but it
> doesn't look like they affect the Four Freedoms--they instead affect
> related aspects of software distribution, like warranties, copyright
> notices, trademarks, indemnification, etc. I don't see this as a
> problem per se since it's outside the scope of software freedoms, but
> maybe I'm misinterpreting it? Maybe there's something important I
> haven't thought of?
>


Open source is all about getting the code into the hands of others
(distribution) and its peer review (for trust and security) as well as
its improvement (contributions). The GPL3 undermines these things
because the terms itself discourage or forbid use of the code. There
are plenty of examples, such as the release of private keys, which no
one with any sense, is going to do. The additional terms section
creates doubt because you are giving people the ability to add terms to
the GPL3 and still call it the GPL3. Instead of just having a license
you can trust on sight, you have to check for conditions.



> I don't want to get into an argument over whether or not copyleft is a
> net win or net loss, since I don't have enough real-world data to back
> either claim :) However, I don't buy the argument that the GPL is
> undesirable because it can demotivate would-be contributors whose
> hearts were in the right place
>

I will only say that there are far more small companies than large
ones. In my experience, probably 90% of the time or more, I have never
used GPL code on their behalf, except where there is clear case of "use
but not combine." They simply cannot afford the fear of a license
violation, intentional or otherwise. So they use the code, but they
don't actively contribute to it when they otherwise might. In that
respect, the GPL really defeats the purpose in my opinion. I think
that is why there are far more large corporate sponsors to Linux than
anyone else.

However, small companies are quite happy using the LGPL, BSD or MIT,
because they can release code on their own schedules with no uncertainty
about the whole of their code-bases being automatically re-licensed.