Subject: Re: Questions to OSI Board quorum
From: David Barrett <>
Date: Thu, 17 Nov 2005 16:10:37 -0800

Ernest Prabhakar wrote:
> On Nov 16, 2005, at 9:01 PM, David Barrett wrote:
>> The question is who's value judgment runs the OSI, and for that I'd  
>> say it must be the board.  And given that the board consists of  many 
>> people, the only way to sample that judgment is by vote.
> To be sure, I don't dispute your desire for your 'day in court.'    But 
> if the opinions on this list are overwhelmingly on one side or  other 
> other, I think that's a reasonable indication of how the Board  would 
> likely rule.

I agree, it's a good indicator.  But it's still not a decision.  And I'm 
not convinced it's as overwhelming as you imply.

So far as I can tell, the OSI was formed to bring objectivity to the 
vague notion of open source.  That objectivity was codified into the 
OSD.  I wholeheartedly endorse this group and its purpose.

But the reason the OSI has a board is so there can be some closure to 
these discussions, for all participants.  Witness how fast the OVPL 
debate ended once the board finally rejected it.  Had they done so six 
months ago, think of how much debate would have been saved?

If it is truly the shut-out that you claim, and if there is truly no 
support whatsoever for an opt-out contributor agreement, if it is truly 
and clearly in violation of the OSD, then the board's discussion should 
take less time than it takes to delete this and all subsequent emails 
until a decision is made.

And if it takes any more time than that, it was a discussion worth having.

Either way, it seems in everyone's best interests for the board to make 
a decision, explain why, and bring this topic to a close.

So again, I'd like to know in what way the three "open" tactics:

- Copyleft
- Contributor agreements
- Optional provisions

When combined, somehow become "non-open".  If it's ok to use copyleft to 
ensure community changes are brought back in, and it's ok to use 
contributor agreements to enable project managers to better manage, and 
if it's ok to use optional provisions to put the contributor in control, 
in what way does the combination of all three somehow introduce a 
greater risk of non-openness?

- Every example of the risks of an integrated contributor agreement 
applies equally to standalone contributor agreements.

- And every argument against the unfairness of making an integrated 
agreement optional applies equally well to making anything optional.

The only consistent argument against I can come up with is "contributor 
agreements are inherently non-open, and we only tolerate their risks 
because they are cumbersome.  If a contributor agreement becomes too 
easy, those risks are magnified, and that crosses the threshold for 

If that's the board's position, just let me know -- I'd accept and 
understand it.  But nobody has said that yet.  Indeed, up until the OVPL 
was rejected out of the blue, the board hadn't even expressed concern 
with it's *mandatory* contributor agreement.  And for all the email 
going back and forth, there are very few speakers, so I have a hard time 
thinking we're statistically significant.

I just want some closure to the topic, and the OSI board exists to 
provide precisely that.  If they say it's ok, I will do the expensive 
and time-consuming legwork necessary to draft and submit the license for 
formal approval.  And if they say it's not ok, I'll shut up.  But 
without reaching either of those conclusions, it's hard to just let it drop.