[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: ReRegarding iraf



On Mon, Jan 10, 2005 at 12:25:11PM -0500, Glenn Maynard wrote:
> On Mon, Jan 10, 2005 at 12:08:08PM -0500, Justin Pryzby wrote:
> > It is maybe complicated than I let on; IRAF includes code from NCAR
> > 1.00, but under a nonfree license.  NCAR 4.X is GPL, and includes
> > mostly-minor differences (some bugfixes, I think, and some changes
> > that the IRAF group made).  I contacted NCAR about making a statement
> > that 1.00 was available under the GPL, but thats an impossibility,
> > because they don't want the overhead of making source code available
> > and similar.  So, I figure I'll package libncar-graphics, make IRAF
> > {,build-}depend on it, and include in the library any changes
> > necessary to make IRAF work (possibly as modifications to the code,
> > and possibly as separate routines).
> 
> If IRAF contains non-free (or rather, GPL-incompatible code, or code
> without source), you can't link GPL code against it, even through a
> shared library.  (I'm not sure if this is what you're suggesting;
> if you're just eliminating the non-free code entirely, that's fine,
> of course.)
IRAF contains code under a GPL-incompatible license.  That code,
however, was relicensed with the GPL (with mostly-minor changes).
IRAF group does not intend to update the code (or, presumably, the
license).

I guess there is exactly one of two problems.  IRAF contains nonfree
code.  *Alternately*, IRAF needs to link with GPL code.  You indicated
that the latter was less of a concern (depending on FSF approval,
contingent on IRAF being GPL compatible).

So, assuming that the *rest* of IRAF (excluding the NCAR graphics
library which it includes) is free, it seems okay.

The only other concern that I have is the nonfree yacc parser (my hope
is that it is no longer implemented as a derivation of yacc (C) 1970
Steven C. Johnson).

Justin



Reply to: