Re: Continuous Development Process TPAC Slides

> On Sep 12, 2019, at 4:16, David Singer <singer@apple.com> wrote:
> 
> 
> 
>> On Sep 10, 2019, at 18:44 , Florian Rivoal <florian@rivoal.net> wrote:
>>> 
>>> I think the chances that someone will bring a suit because the Rec hasn’t yet issued and hence they haven’t yet granted a license is zero. The judge would laugh.
>>> 
>>> I think this might cover the case where someone proposed something, it was integrated, marked- at-risk, and removed before Rec.  Someone did an implementation (as requested) during CR. They have no formal license. 
>>> 
>>> But I suspect that any license grant we’ll get is for the IPR that became essential as a result of integrating the contribution into the spec. that existed at the time of the contribution. That’s the bulk of it; if later changes to the spec. made more IPR from that owner essential, it’s not covered, but that’s minor.
>>> 
>>> I still think it’s somewhat of a strange situation. “You were willing to grant a license to the IPR supporting your contribution if and only if it issued as a Rec., and it didn’t, and you’re now suing someone who made a good-faith implementation to prove the spec. while it was in development?"
>>> 
>>> Is that it?
>> 
>> I agree it is not terribly important, but I don't understand why you're having an issue with it.
> 
> I’m not sure I am having a hard time with it. I wanted to understand what problem it was fixing.
> 
>> The WHATWG has the exact same thing, and Apple agreed to that (and I'd assume, was part of drafting that).
> 
> Normal operating mode in the WHATWG is that it operates as an open-source-style project and expects contributions from non-group-members, so a contribution license was essential. For wjatever reason (I don’t recall), it’s applied to everyone, even those who will go into whole-spec. agreement at the next snapshot.

I'm not sure that's right, as I understand, the WHATWG does not distinguish between "group members" and external people. To make a contribution, you have to first become a Workstream participant, which binds you to the IPR policy as well.

>> The evergreen patent policy derives from the WHATWG's, and had that in it from the start, and you never had an issue with it. The proposed joint patent policy for REC track and evergreen keeps it too. Why is this a problem?
> 
> Clarity is what I need: when I talk to lawyers, if I am vague with them, I get vague replies…

Ok.
* It doesn't add anything to the long term view of a spec's lifecycle, but provides (some) clarity sooner.
* It gives peace of mind to implementers to implement and distribute things that have not yet made it to the stable state of the specification, and might get removed before it gets there. "at-risk" is how this things are described in the >= CR stage, but features are also added then removed in the WD stage as well.
* Given that other bodies (WHATWG, CGs) have it, having it for W3C standards as well removes a reason / excuse to argue in favor of developing the document elsewhere.

—Florian

Received on Wednesday, 11 September 2019 22:20:47 UTC