[ILUG] EU Rejects Patent Software Law - For Now

[Brian Brazil]

On Fri, Feb 18, 2005 at 11:00:30AM +0000, Colm MacCarthaigh wrote:
> On Fri, Feb 18, 2005 at 11:07:29AM +0000, Bryan O'Donoghue wrote:
> > #define ISNOT(x,y) if(x==y) return 1; else return 0;
> 
> If you're going to post an insane rant, at least try and write code that
> would actually work.
> 
> #define ISNOT(x,y)  ((x) != (y))
> 
> I mean, you've just proved that the implementation of an ISNOT operator
> is not (no pun intended) obvious, and non trivial. Since you managed to
> mess it up this one-liner in quite a few ways.

Sorry, also wrong. The patent spec was 'point to the same location in
memory' - ISNOT isn't NOTEQUAL. Although someone would probably argue
that the patent would also cover the above macro as a trivial extension
(software patents are fun for all the family).

#define ISNOT(x,y) (&(x) != &(y))
or if you want to avoid !=
#define ISNOT(x,y) (&(x) - &(y))

Granted this should never be true in C (unless used as ISNOT(x,x)), but
I think its possible in C++ using references.

Also as we now use 3 operators, we aren't (directly) infringing. On the
other hand - does a macro function count as an operator? Maybe
infringing in the original code, but not after cpp gets to it.

Brian

-- 
Website: http://netsoc.tcd.ie/~bbrazil
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 185 bytes
Desc: not available
Url : http://mail.linux.ie/pipermail/ilug/attachments/20050218/48277544/attachment.pgp