BOF Summary (Perpetuating Stereotypes?)

["William A. Rowe, Jr." <wr...@rowe-clan.net>]

Thank you to Tuscany, Wicket, Felix, Lokahi, Ivy, stdcxx, RAT and
others who I forgot to scribble down and were in attendance at
the Incubator BOF (Birds of a Feather) program in Amsterdam.
We had great feedback!

I raised these concerns as a "10 things we hate about incubation"
lightning talk, and it certainly raised a mixed reaction.

I've heard this is pretty much just what incubating projects wanted
to sound out about - and I had presented it so the entire ASF community
could hear these issues and (we hope!) step in to help solve some.

I've heard that it's perpetuating stereotypes - but the fact is that
*these perceptions are current*.  This isn't ment to say that the
incubator isn't working - it's ment to say the incubator can work better
and here's what we want to see change.

Anyways, love it or hate it, I shared this at the Lightning Talks and
the issues are on the table.  Some we are already doing something about
so they don't need a response.  But others haven't been discussed very
recently - if you want to discuss one of these themes, please start up a
***NEW, SPECIFIC SUBJECT*** and post away :)

-----------

Here's what they had to say...

We wished we had longer than an hour, we flew through introductions,
raised the issues to our project right now (or in the past), and found
our hour had flown by as the last project was introduced.

We especially wish we saw more of our mentors (four of them were there)
and iPMC members to talk to at the BOF.  Here's what we raised as our
issues for you to consider.

=== our issues ===

defining process.  We get confused because there are multiple disparate
definitions of the very same process in several places (and some of us
are also trying to refer back to our sponsoring PMC's policies too!)

too many process changes.  We get further confused by so many apparent
policy changes during our project's incubation lifespan.

learning mailing list participation.  What seems simple to participants
of other ASF projects isn't so obvious to those of us from corporate
projects, or those that were developed mostly in irc, or those with some
more structured/hierarchical organization.  Teach us.

creating releases.  Although we are glad there's been so much progress,
there still seems to be a lot of ambiguity (see 1 and 2 above).  We are
also still unclear if we must, or should, or shouldn't, or can't.  Some
explanation of why and when would be helpful.

too many votes.  (See specific confusion below).  What should be a vote,
what should be a poll, what's just consensus building?

why no -1's?  Although these are rumored to be common, we only see a -1
when there is a specific release/licensing/labeling issue in a release.
When is it appropriate?  When is it not?

how can we get reactions on general@?  Sometimes it seems we are posting
into the void.  We already know ESR's How to ask smart questions? but we
don't know how to get a rise here.

how many mentors do we have, who should they be, and what is it they do?
Yes - this is pretty well spelled out, but it comes up again.  Can we
expect a certain level of activity/response from them?  Should we ask
for new mentors when ours go AWOL?

inconsistent policies.  Yes, this is probably said ten times, but we brought
this up as many times at the BOF.  We are really frustrated!

documentation about incubator.  Buzzword bingo confuses us.  PPMC, IPMC,
PMC - which is which?  Who are on them?  Who should be on them (specifically
the PPMC)?

publicity and attracting committers/users.  How can we do this before we
have a release, how can we do this with a release while branding ourselves
as 'incubating' all over the place?  How can we convince folks to come to
us when we are telling them we are in limbo?

---------

Here are things we got into discussing the specifics of...

rat.  we like rat.  We are glad to be incubating rat, too, soon, but would
like some more guidance on using it, or to be simpler to set up.  Robert
promises us a way to identify 'excused artificats' that won't get Apache
License text for a good justification.

we want kickstart templates!  There are too many documents which always
look exactly the same, and there's no really comprehensive collection of
them in one easy-to-find place just yet.

too many votes.  Many of us believed we had to "vote" in a person to be
the release manager, or "vote" before we can make a tag, and some say
it is contrary to the ASF way.  http://httpd.apache.org/dev/release.html
says a bit about why the RM isn't defined in the interest of allowing
"the best code to win" and not letting one (or small group of) individual(s)
to get in the way of progress and releaseS.  There is an even better and
more complete doc somewhere, I can't find it anymore.  The upshot -
"If you don't like what an RM is doing, start preparing for your own
competing release." isn't spelled out very well @incubator.

maven is a pain.  Good to see allot of activity on figuring out the
relationship between incubation and maven, we are optimistic.

external releases.  We need more guidance when an incubating project
continues to persist outside of the Incubator (e.g. established works
that are being incorporated into the ASF.)

why join.  We don't see enough information about what compels (or maybe
should compel) a group of developers to start at/move to the incubator.

incubator pages.  What do we read first?  There are too many pages that
are spread out between www.apache.org/dev/, incubator.apache.org/ and
even elsewhere and this is overwhelming to a new learner (and sometimes,
to those who've been here a while and looking for specific answers.)


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: AWOL Mentors? (was: BOF Summary)

["William A. Rowe, Jr." <wr...@rowe-clan.net>]

Niclas Hedhman wrote:
> On Tuesday 08 May 2007 04:04, William A. Rowe, Jr. wrote:
>> how many mentors do we have, who should they be, and what is it they do?
>> Yes - this is pretty well spelled out, but it comes up again.  Can we
>> expect a certain level of activity/response from them?  Should we ask
>> for new mentors when ours go AWOL?
> 
> I find this to be an interesting tidbit (yeah, I just got back from my long 
> trip in Europe...)
> 
> How does one know that the Mentor(s) go AWOL??
> Since there is no 'heart beat', podlings could tag along quite a while without 
> knowing that there is no oversight. Of course this is more prevalent in 
> projects with one Mentor than those who has 3, but still...
> 
> Should we perhaps require Mentors to ACK ("Yes, this looks accurate.") the 
> monthly/quarterly reports, and failure to do so mean the Mentor is AWOL??

That seems quite rational to me.  Even if they ack things late (as in, we
find a tag to update the report-after-the-report has been filed.  This would
already help me answer jukka's query about stdcxx that this situation he'd
identified has been acknowledged and is being addressed.)  Better late than
silent?

I don't think anyone expects the same level of involvement for all 3x or more
of a projects' mentors.  But adding 'another mentor' just to have three names
does a disservice to the podling.

The only ones who really can identify an absent (as opposed to a quiet) mentor
are the other mentors.  New podlings don't have any idea of how much feedback
they should expect from their mentors.  (Next to none, when everything is
humming along splendidly :)  But I think as mentors, we know.

So what's the threshold?  If I know a fellow mentor has gone AWOL, should I
bring this up with them after a month of silence?  Three months?  We should
definitely, as mentors, ping each other privately first with a 'what's up?'
before bringing an issue to the incubator PMC.  But let's just say things
stay silent, the mentor doesn't respond to that ping, doesn't indicate that
they want to step down on their own.  How soon to bring things up with the
incubator PMC?

All chewy stuff to munch on a while, I don't think there are clear one-size
fits-all solutions to this issue.  Thoughts?

Bill


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: BOF Summary (Perpetuating Stereotypes?)

[Niclas Hedhman <ni...@hedhman.org>]

On Tuesday 08 May 2007 04:04, William A. Rowe, Jr. wrote:
> how many mentors do we have, who should they be, and what is it they do?
> Yes - this is pretty well spelled out, but it comes up again.  Can we
> expect a certain level of activity/response from them?  Should we ask
> for new mentors when ours go AWOL?

I find this to be an interesting tidbit (yeah, I just got back from my long 
trip in Europe...)

How does one know that the Mentor(s) go AWOL??
Since there is no 'heart beat', podlings could tag along quite a while without 
knowing that there is no oversight. Of course this is more prevalent in 
projects with one Mentor than those who has 3, but still...

Should we perhaps require Mentors to ACK ("Yes, this looks accurate.") the 
monthly/quarterly reports, and failure to do so mean the Mentor is AWOL??


Cheers
-- 
Niclas Hedhman, Software Developer

I  live here; http://tinyurl.com/2qq9er
I  work here; http://tinyurl.com/2ymelc
I relax here; http://tinyurl.com/2cgsug

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org