Mirror of the gdb mailing list
 help / color / mirror / Atom feed
From: "Martin Schröder" <gschroeder@onlinehome.de>
To: "Robert Dewar" <dewar@adacore.com>
Cc: <gdb@sourceware.org>
Subject: Re: generic query regarding GPL and licensing terms associated with gdb
Date: Fri, 13 Aug 2010 18:15:00 -0000	[thread overview]
Message-ID: <B62E4B18CADE41B399FB2B5D083CCFE3@igor> (raw)
In-Reply-To: <4C657043.3020206@adacore.com>

Since this posting has nothing to do with GDB anymore, I think it's best if 
we moved this from the mailing list to regular E-Mail.

Basically, my closing statement is:
I concur with Tom Tromey: If in doubt, ask the FSF. It's just that I don't 
see how I could doubt the legitimacy of talking about code that is correctly 
and willingly licensed to you under GPL terms.


For completeness sake, below is my complete argument, which can be skipped 
by everyone who is not interested in it. ;)


Robert Dewar wrote:
> Martin, what you are saying is seriously wrong!
>
> The GPL never automatically forces disclosure of anything.
>
> If you take GPL code, modify it by adding trade secrets,
> and then distribute it without giving a GPL license, it
> is not the case that somehow you have lost the trade
> secrets, or that anyone is free to disclose them. That
> is not at all the case.
>
> Yes, it is true that if you grant a GPL license to someone
> for the distrtibuted object, then of course there are no ytrade
> secrets.
>
> But NO ONE EVER forces you to issue a GPL license.
>
> You most certainly can redistribute without granting
> such a license (and if there are trade secrets present,
> then indeed you cannot grant a valid GPL license anyway).
>
> Now if you *DO* redistribute in this manner, you have likely
> committed a copyright violation, actionable in the usual way.
>
> In response to such a claim of copyright violation, you can
> AT YOUR DISCRETION, cure the infringement in the future by
> granting a GPL license, but no one forces you to do so,
>
> The GPL NEVER forces you to disclose anything, it simply
> says that if you meet certain disclosure and distribution
> requirements then you have a license to redistribute, that's
> all!


I think our problem stems from the fact, that we look at the problem from 
diametrically opposed sides.

As far as I understand you, you quite correctly point out that the GPL does 
not carry the force of law. It is merely a license which is not enforced by 
any specific law, but instead supported by the laws concerning copyright and 
licensing issues. The GPL in and of itself does not grant you any liberties 
or restricts your action in the direct sense that criminal and public laws 
do.

I can't argue with that fact, indeed it would be foolish to do so.


But that's not at all where I'm coming from. The original question contained 
the following sentence:
"And the organisation has used gdb and modified gdb undder GPL licensing 
terms."

I am fully aware that you can understand this in two ways:

1.) The organisation has taken GDB and modified it, either for internal use 
without redistribution (1.a), or with redistribution to someone else (1.b), 
and witholds the code.

2.) The organisation has taken GDB, modified it and redistributes or plans 
to redistribute it under the GPL.


If it's case 1.a) *without* redistribution, I fully support your point. The 
changes have to be regarded as trade secrets and the GPL fully covers this 
use. You can't touch them, but they quite surely can sue you.

If it's case 1.b) *with* redistribution, we enter muddy territory. Yes, they 
are not forced to publish the code under the GPL, but you *can* sue them for 
copyright / license violation and given the FSF's/EFF's track record, you 
would most likely win, whereupon the GPL does apply to the redistributed 
code, which means that the code is freely visible, which means that no judge 
or jury would be able to convict you for just discussing what the code 
*does*. Afterall, you just rephrase what everyone can see for themselves 
anyway. If the code's patented, you won't be able to use it, but you may 
talk about it.

Of course, the company can still sue you anyway, especially during the time 
they themselves are sued, so it's still dangerous in a way to talk about the 
code. So yes, I do agree with your point there, too.


But what I argued for was case 2. What happens when the company *does* 
publish the changes and licenses it under GPL terms. Could they sue you for 
saying what the code does? I already explained why I don't think they could 
have any trade secrets that they could enforce that would be touched just by 
viewing and explaining the source.

That leaves patents and selling the software. If it is patented, you *can* 
freely talk about the code, even if you can't use it without paying 
licensing fees. Selling the source code (patent or not) seems to be covered 
by the GPL, but the code you acquire *must* be licensed to you under the GPL 
terms in that case, or you again have the second version of case 1.). And if 
you have it under GPL terms, you can talk about it, because you yourself 
must be able to distribute the changes (after paying the appropriate 
licensing fees if applicable).

Anyway, I don't see how you could be sued for talking about a piece of code 
that *is* correctly licensed to you under GPL terms. Or at least, you could 
be sued, but they wouldn't have a chance in hell to win.


So, you see, my approach is not what they are forced to do, but what you can 
do without getting (successfully) sued. And I simply don't see a way to be 
sued for a correctly GPL-licensed piece of code.



But other than that, I concur with Tom Tromey: If in doubt, ask the FSF. 
It's just that I don't see how I could doubt the legitimacy of case number 
2.). :)


So long,
    Martin. 


  parent reply	other threads:[~2010-08-13 18:15 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-13  4:08 paawan oza
2010-08-13  5:04 ` Robert Dewar
2010-08-13  6:16   ` paawan oza
2010-08-13  7:56     ` Martin Schröder
2010-08-13 14:26       ` Robert Dewar
2010-08-13 15:35         ` Martin Schröder
2010-08-13 16:18           ` Robert Dewar
2010-08-13 17:50             ` paawan oza
2010-08-13 18:15             ` Martin Schröder [this message]
2010-08-24 21:13               ` Steffen Dettmer
2010-08-24 21:33                 ` Robert Dewar
2010-08-27 11:38                   ` Steffen DETTMER
2010-08-15 16:40           ` Florian Weimer
2010-08-13 14:24     ` Robert Dewar
2010-08-13 15:50 ` Tom Tromey

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=B62E4B18CADE41B399FB2B5D083CCFE3@igor \
    --to=gschroeder@onlinehome.de \
    --cc=dewar@adacore.com \
    --cc=gdb@sourceware.org \
    --cc=lionhead@onlinehome.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox