[oc] Beating a dead horse

[Larry Doolittle <ldoolitt@recycle.lbl.gov>]

I have to respectfully disagree with an idea expressed repeatedly
here recently, on the subject of licenses.

Victor the Cleaner <jonathan@canuck.com> (among others) said
> With respect to the license terms,
> I strongly recommend adoption of the NetBSD license or similar.  I believe
> that core users must have the option of not releasing their source if
> building a commercial product, and for that reason the GPL is a problem.

It's only a problem if you can't arrange a seperate license with the
copyright holders.  I think we are all grown-up here.  If you can
afford to mix a proprietary, paid-for core with the OR1K, you can
also afford to license the OR1K.  Dual licenses are real, and work.

For the hobbiest, experimenter, and researcher, the implications of
the GPL are real and positive.  I _want_ a thousand flowers to bloom
around the OR1K.  I am itching to ether get started, or piggyback
while someone else gets started, on the TCP accelerator, the IEEE-1394a
Link interface, a self-virtualizing mode, a stand-alone PDA packaging,
hardware crypto, OR1K-FPGA hybrid SOC ASICs, and much more that I
haven't even thought of.  The "All the world's a PC" mentality has
really put the computer hardware people in a rut.  Capable FPGAs and
smart University departments are just what we need to get pulled
out of it.  I don't see that happening (as much) with NetBSD style
licensing.  We'll all just end up with OR1K based hardware, with
proprietary peripherals, that we can't re-FPGA-program because we
don't have the source to those drivers^H^H^H^H^H^H^Hperipherals.
Except for the licensing cost savings to the manufacturer, I don't
see the advantage to that over an ARM.

Sigh.  I wish I didn't feel compelled to say all this.  It's only
distracting Damjan from releasing his code, and that's what we're
really all here for.  To him, I say, go for the GPL!  But also
keep a close eye on who are really the copyright owners, so you
keep the option open of commercially licensing the core also.
I can't imagine your department would turn down an additional
source of funds.

     - Larry Doolittle   <LRDoolittle@lbl.gov>