Regarding the "TheOtherGCCBasedFortranCompiler" Wiki Page.

Brooks Moses bmoses@stanford.edu
Wed Nov 2 20:37:00 GMT 2005


Tobias Schlüter wrote:
> Brooks Moses wrote:
>>I would probably add, for purposes of that page: "Since that time, the 
>>two compilers have diverged significantly," and leave it at that.
> 
> Yes that makes sense.  The fork mail is rather polite in that saying "Andy
> took away our CVS write access" leaves out several details :-)

I'd gathered, yes.  :)

> As far as the discussion of GPL legalities (which should really be taken to
> misc.gnu.discuss, so I won't go into depth), there are at least two
> well-documented cases where linking software to GPLed code resulted in it
> being GPLed in the end (by out of court settlements) 1. Next's Objective C
> frontend, which is now part of GCC and 2. the FSF doesn't name the specific
> software, but they say that some program that was linked against the readline
> library (which is GPL) was GPL'd after the FSF found out (this is in the 'why
> the LGPL is not preferable' page linked off somewhere on their license list).

Indeed.  I think I was somewhat unclear -- what I was trying to make was 
a distinction between "your code may become free software without your 
consent" and "you may find yourself in a position where you'd prefer to 
choose to make your code free software rather than paying court-imposed 
damages and disentangling it from the GPL-licensed things that it's 
based on."  And, as you point out, a number of companies have found 
themselves in such a position and made out-of-court settlements.

(I'd take this to misc.gnu.discuss, but I think I'd rather not spend 
that much time on it, if that's ok with you.)

>  Of course this is moot, because if the FSF were asked they would certainly
> relicense g95's library under the GPL + exception license that other GNU
> compilers use, it just shows how little Andy cares about licenses.  (BTW is he
> still using that multiple precision library without the required
> acknowledgement?  I see he says he rewrote something related to internal
> arithmetic, so this might not be the case anymore.)

I have no idea at all on the latter point.

- Brooks



More information about the Fortran mailing list