incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <>
Subject Re: Short form IP clearance
Date Thu, 22 Oct 2015 04:29:21 GMT
On Wed, Oct 21, 2015 at 11:10 PM, P. Taylor Goetz <> wrote:

> Apologies for potentially coming out of left field on this…

hehe... I did too :-)

> But I think that IP clearance is currently a difficult road to travel, and
> I worry that we are graduating podlings that don’t even know when or how to
> go down that road. It’s all too easy to merge a github pull request without
> considering IP clearance.
> I’ll admit that I’m unclear on the policy, and try to err on the cautious
> side.

I expect anybody unclear to simply ask. That happens all the time. We have
press@ to help people with questions about marketing and outreach. We have
trademarks@ for branding questions. We have legal-internal@ and
legal-discuss@ to ask questions about IP clearance (and other legal-ish

We have a system of Trust and Independence for our TLPs. The (filed) forms
are there for a TLP to follow, to check off steps, etc. There is a guide
and a description on how to fill out and file the form. And what is needed.

All good.

My concern is the injection of the IPMC into the process, and subjugating
one TLP to its will. IMO, that just is not *possible/allowed* by the
structure we have set up at the ASF. Thus, step 5 needs some modification
and 7/8 need removal to align with the actual structure of the ASF.

> I’ve seen commits to TLP projects that made me think “WTF… how did this
> evade IP clearance?”.
> I may be overreacting, but it seems to me IP clearance is REALLY
> important, and I worry that it may be taken for granted.

You say it yourself: commits come from out of the blue. Nobody
second-guesses a TLP's series of commits. Nobody second-guesses (say) the
trust network they have set up in a KEYS file for their release artifacts.
The "IP Clearance" process is already *more* controlling than other
policies that TLPs must follow.

I'm not asking to omit the process, but (IMO) the notion that the IPMC has
control over our other projects is simply incorrect, so the doc needs


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message