Re: Source licensing

From: Arthur Keller <arthur_at_kellers_dot_org>
Date: Mon May 10 2004 - 05:01:56 CDT

At 9:13 PM -0400 5/9/04, David Mertz wrote:
>On May 9, 2004, at 7:47 PM, Arthur Keller wrote:
>>How about, closed proprietary source, published proprietary source,
>>and free software?
>
>Those look like good phrases to me.
>
>>You mention a variety of "free" software. Which "free" software
>>model should be used for the licensing/ownership of UC-developed
>>software?
>
>I actually don't care nearly as much as a lot of people. If I were
>making the decision all by myself, I'd probably say Public Domain
>(or BSDish... the 'B' in there being UC-Berkeley, after all).

The original BSD license was a mess. It locked up too much control
to the UC licensing people. I want to avoid that problem. That's
why there had to be the independent development of Free BSD.

>But when we went over the discussion back in August, a lot of
>developers felt strongly about the GPL's guarantee that proprietary
>companies not be able to incorporate our code into derived,
>closed/proprietary products. I'm certainly not against GPL either
>(or EVMPL, which isn't much different, and hopefully compatible).
>
>Something like Karl suggests--registering service marks or
>trademarks on "OVC compliant" or some logo--seems like a great idea
>to me. I've never been through ISO 9000 myself, but most people
>seem to shudder at those words :-). But he might be right about the
>political benefit of that.
>
>Even if a proprietary vendor used our code, they couldn't win an
>"OVC compliant" mark without meeting OVC standards. Obviously,
>those standards would need to be non-discriminatory; they might
>involve certification fees though. The real "value proposition" IMO
>should not be using some lines of code developed in
>EVM2003/OVC-reference-design, but in being able to claim OVC
>standards compliance legally.
>
>>Second, I wanted to make the distinction between VoteHere's
>>approach to published software, and the OVC's approach (cases 2 and
>>3 above). Is such a distinction worthwhile? Thanks.
>
>I find pointing out the distinction between VoteHere's
>"look-but-don't-touch-and-sign-the-NDA" and OVC's EVMPL to be quite
>important. You've done a good job of describing the licensing space
>above.

Let me give a strawman example. Suppose some commercial vendor
wanted to create a voting machine based on UC-developed software that
(1) was compatible with the BVA, BRP, etc., (2) included a
proprietary voice recognition system that allowed visually impaired
voters to "speak" their choices into a microphone, and (3) copied the
code from the EVM for creating the ballot XMLs, writing to the CD-R,
reading the ballot definition files, etc. Suppose that this vendor
got their system certified for use with the rest of our system. Is
this something we want to allow or prohibit? GPL would discourage
such innovation. A "FreeBSD"-type license would allow it.

Strawman example 2. Suppose that the OVC deployment model turned out
to be a failure, but the software turned out to be everything we want
it to be. Would it be better to let commercial vendors take over, or
allow it all to die? Or would the "FreeBSD"-type licensing make it
harder to succeed because commercial vendors could take over?

Best regards,
Arthur

-- 
-------------------------------------------------------------------------------
Arthur M. Keller, Ph.D., 3881 Corina Way, Palo Alto, CA  94303-4507
tel +1(650)424-0202, fax +1(650)424-0424
==================================================================
= The content of this message, with the exception of any external 
= quotations under fair use, are released to the Public Domain    
==================================================================
Received on Mon May 31 23:17:29 2004

This archive was generated by hypermail 2.1.8 : Mon May 31 2004 - 23:18:16 CDT