incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From jan i <>
Subject Re: [PROPOSAL] Apache Atlas for Data Governance on Hadoop
Date Wed, 15 Apr 2015 15:18:44 GMT
On 15 April 2015 at 15:10, Justin Erenkrantz <> wrote:

> On Wed, Apr 15, 2015 at 8:55 AM, Ted Dunning <>
> wrote:
> > I hate to be a pissant, but why does it follow that new code has to be
> > implemented in private?
> >
> > Building community can start with the first semi-colon.  It doesn't have
> to
> > wait until the code is frozen.
> We were having this conversation with a bunch of folks last night here
> in Austin.
> One thing Brian mentioned in his keynote this week is that our culture
> borrows from the IETF for "rough consensus, running code".
> It's extremely hard to come to Apache when you have a vague idea and no
> code.
> The Incubator is simply too high a bar for folks who just want to play
> with an idea and aren't sure where it is heading.
Incubator has this bar for good reasons, we could use project LABS for
experimental projects.

That is the perfect sandbox for that. At the moment particiants needs to be
committers in advance. We could change that so a LABS project must be
created by a committer,
but other project members become LABS committers.

I happen to be both IPMC and PMC in labs, so if there is interest I would
be happy to carry this idea

jan i.

> So, I wouldn't hold it against people - turn it on ourselves - how can
> we create a process to facilitate people who aren't ready yet?
> FWIW, my opinion from last night is that is exactly what GitHub is
> for.  (Apache Labs was that place at one time, but it's only for
> members - not external folks.)  Once you have that rough consensus and
> running code and have a fledgling community of stakeholders, then it's
> time to bring it to Apache via the Incubator.  To be fair, there were
> others who disagreed; but, the rough consensus (ha!) was that the
> Incubator hasn't yet cracked that model.
> My $.02.  -- justin
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

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