Nicolas,

I figured a simple NACK would not suffice, given the range of topics
this thread has covered.  This is not all meant for you, as I expect you
have heard these arguments once or twice in the past. :)  I just wanted
to post to this thread once (and for all).

On Mon, 2009-06-15 at 00:28 -0400, Nicolas Pitre wrote:
[snip]
> Now... who can make that call?  Is there someone with code in OpenOCD 
> who is against such a relicensing?

Yes.  OpenOCD *is* GPL.  I would _not_ have contributed as much as I
have to the code, if I thought anyone could later distribute versions
with proprietary extensions.  I will vehemently oppose and challenge all
such use cases, though this statement deserves some clarification.

The GPL is about freedom -- to the extent that it forces its "freedoms"
down your throat when you distribute GPL-licensed binaries.  In that
respect, the code is _definitely_ not as "free" as the public domain,
but any "lost" freedom assures it remains _more_open_ moving forward.
The GPL is better to keep source code open than "open source licenses,"
at the cost of the freedom to distribute binaries without source code.

Since "no exceptions" should now be clear, I would consider distribution
of OpenOCD 0.2.0 binaries with support for the FTD2XX libraries as a
flagrant violation of the GPL.  This express sentiment serves as
unequivocal and sufficient warning for anyone who would consider
otherwise.  I am very sorry if this is a problem; however, those are the
consequences of the licensing terms for the respective packages.  To me,
this was always clear; debate seems moot.

With dynamic module loading capabilities, it _might_ be possible to
independently distribute a ftd2xx-plugin package, as such would be in a
potentially "gray area" that avoids consequences (as you pointed out).
Maybe.  Personally, I think that would be a GPL violation too, but
perhaps I am simply an ideologue backing a Stallmanesque agenda.  ;)
The community should be discouraged from trying the limits to find out.

Others have pointed out that this licensing violation occurs only when
distributing the binaries, so I do want to provide assurance that I have
no problem supporting these types of features -- for developers to use
in local builds _only_.  I think such support is great, in fact.

Specifically, these kinds of restrictions should not impact OpenOCD's
functionality for developers.  If libftdi is insufficient, the community
should have someone work on improving that code until it performs
competitively.  The same goes for the underlying libusb support or for
any other component lacking a comparable GPL-compatible alternative.  

For developers, in-tree proprietary solutions give implementations that
can be compared with open solutions, to ensure these stay complete and
competitive in our code base.  Further, forbidding the distribution of
these proprietary features helps the community by providing a strong
impetus to develop or improve equivalent open source solutions, which
leads to pressure on the proprietary vendors to open their libraries.

Any feature support is good, but open is better; however, only features
"compatible with the GPL" can be distributed in OpenOCD binaries, and
proprietary libraries do not fall into this last category.  I do not
care if this is an unpopular declaration; my obstinance runs deep,
though my level of compromise should demonstrate that I stand apart from
those that mindlessly follow the GNU Hurd. ;)

Sincerely,

Zachary T Welch
Corvallis, OR

P.S.  I believe it is possible to deliver a technical solution that
would allow a binary FTD2XX solution to be distributed for OpenOCD,
sidestepping the present legal hurdles.  Such would require a lot of
work (much of which would be added to and benefit OpenOCD as GPL code),
but the point I want to make is simple.  We do not need to change the
license when technical loopholes exist that will achieve the same ends.

P.P.S.  Full disclosure: I am a professional software developer who is
looking for "GPL-compatible" work, so I have both vested and immediate
interests in promoting more open source software development activities.

_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to