License Proliferation

[Ernest Prabhakar]

Hi Russ,

On Sep 1, 2005, at 8:19 PM, Russell Nelson wrote:
> The very most basic question (which you don't ask) is this:
>
>   Does OSI certify all licenses that comply with the OSD?
>
> or
>
>   Does OSI certify all good licenses that comply with the OSD?
>
> In other words, do we take into account the interests of the
> community?  Or do we act as a conditional rubber stamp -- conditional
> *only* upon compliance with the OSD?

Well put. Thanks for asking. :-)

> I suggest that the answer is not obvious.  It's not obvious what the
> right answer is, and it's not obvious what the community wants us to
> do.  Ian Lance Taylor was complaining earlier that we don't ask people
> questions.  Presumably, now that I'm asking, he'll have an opinion

  I completely agree that this is the vital question.  The thing that  
I personally found frustrating was the sense that this question had  
*already* been answered, but we (the list) hadn't been told exactly  
what the answer was.  I am glad we're able to hash this out in the  
open, and I commend you for that.

However, part of the answer *does* seem obvious.  I am NOT a layer,  
and I don't speak for anyone but me, but my understanding is that the  
*legal* obligation of the OSD trademark is that the OSI *must*  
eventually certify all qualifying licenses that are submitted.  Is  
that true, or isn't it?

My personal preference is that we (this list, representing the OSI)  
provide *recommendations* for good drafting procedures and non- 
duplicativity, but always be clear that these are not  
*requirements*.   I suspect it is the confusion between those two  
terms that has been the source of much angst.

That is, if someone submits a valid but ill-written license, we could  
an should provide *suggestions* for improvement.  But, whenever - 
they- feel they'd had enough suggestions and decide to submit it, we  
would still approve it (if it conforms to the OSD).  Let the license- 
proliferation team sort it out after that -- that's *their* job, not  
ours.

If that's NOT what you want to do, then I personally would feel much  
more comfortable if the OSD was amended to reflect what the hard  
criteria actually *are*, because as it is now the additional 'soft'  
constraints far from transparent.

My $.02.

-- Ernie P.