On Mon, Jan 07, 2013 at 04:34:04PM -0500, Timothy Normand Miller wrote:
> On Mon, Jan 7, 2013 at 4:09 PM, Troy Benjegerdes <[email protected]> wrote:
> 
> > For some comparisons, check out:
> >
> > http://www.gaisler.com/doc/grfpu_dasia.pdf
> >
> > http://opencores.org/project,fpu
> >
> > Are we looking to have full IEEE floating point, or a subset of IEEE?
> >
> > Once there is enough code to do add/subtract/multiply I'd like to
> > write a regression test that runs a couple of floating point validation
> > tools.
> >
> >
> Sweet.  BTW, regression tests, like most of this stuff, should be pure GPL.
>  Or you could do a more permissive license.
> 

If there is anything that should have a license that demands free access
to the code, it should be regression tests. I got paid for awhile to write
regression tests, and what I found was a huge pile of code that was critical
to the proper function of the product, a lot of which was some type of
'Copyleft', and **NEVER** made it back outside the testing lab.

Thus was born the COPYRANT:
https://bitbucket.org/dahozer/tfs/src/587e000f6df1adca6699457b0ca8155155cf2ded/tests/fuse/COPYRANT?at=default

In other words, if I write new test code, I want something *STRONGER* than
the AGPLv3, such that it requires anyone using the code as a test for a 
product to distribute to the end user exactly the same regression tests
that validated to the manufacturing line the product was really functional.

The AGPLv3 might actually already cover this case *mostly*, because most 
regression tests (that I've seen) have some form of network interaction.
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to