incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benson Margulies <>
Subject Re: [DISCUSS] [VOTE] HCatalog to Graduate and become part of Apache Hive
Date Wed, 20 Feb 2013 13:57:39 GMT
On Wed, Feb 20, 2013 at 5:31 AM, Ross Gardler
<> wrote:
> I'm brought to this thread byt he board report but my response here is as
> an IPMC member. My comment on the board report is quite different, it is
> "I've read the thread on general@ and feel that the IPMC should make a
> clear recommendation to the board in this and similar cases. The IPMC
> discussion seems to be healthy and productive."
> So, as a an IPMC member I have a few open questions [inline]...
> On 11 February 2013 18:20, Alan Gates <> wrote:
> Also, it has been agreed that each HCatalog committer will be provided with
>> a mentor from the Hive community to help him/her learn the rest of Hive,
>> with the goal of becoming a committer on Hive within six months.  The
>> submodule state is transitionary, not an end point.
> Why was this"mentoring" not done as part of the incubation process since
> building the right community structure for graduation (along with IP
> clearance) is the main role of the incubation process? Was Hive the
> sponsoring project for this proposal? If not why not?

Ross, my suspicion (and I haven't done the digging here on vacation)
is that HCatalog started incubation with the intention of becoming a
TLP, so their original sponsor was the incubator itself. The idea of
merging into Hive came up late in the process. So the Hive people had
no warning or reason to be part of the supervision.

Thus, your email seems to me to pose this question: "Should the IPMC
seal of approval be good enough for an existing TLP to grant committer
status?" I don't see why that should be. Projects have their own
culture and conventions, and if Hive was not participating in the
incubation process, why should those conventions be part of the
HCatalog incubation?

This brings me to my other obsessional point here. If the plan, from
the start, had been to import code to Hive, there would have been no
need for the IPMC to do anything except IP clearance. Hive could have
'incubated' via the usual mechanism of accepting patches. The notion
of 'incubate sponsored by project X' makes sense to me if the eventual
trajectory is some sort of autonomous subproject, and not otherwise.

> I ask these questions because HCatlog is making a very strong case that any
> other option for graduation is not appropriate. At the same time we are
> being told by the Hive PMC that the mentoring of the committers is
> incomplete since they have insufficient merit within Hive to be trusted to
> be full members of that project.
> it also concerns me that in this same month the IPMC board report says "The
> main concern of the incubator continues to be the quality and reliability
> of supervision... The supply of mentoring seems, still, to exceed demand."
> Why is it that the Hive PMC feels it is able to provide "mentoring" within
> their own PMC through the creation of what some people see as
> an umbrella project, but not here in the IPMC?
> Finally, why can't I find the HCatalog proposal in my mail client, markmail
> or the wiki (not had coffee yet, feel free to call me [insert adjective])
> Ross

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message