incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marko Rodriguez <>
Subject Re: [Proposal] TinkerPop: A Graph Computing Framework
Date Mon, 29 Dec 2014 17:33:50 GMT

Here is how TinkerPop current runs it "TinkerPop-Contributors."

	1. If you are a vendor, you get one engineer from your organization to be on TinkerPop-Contributors
who speaks on behalf of your organization/product. (~15 people)
		- e.g. "That API addition will be extremely expensive for us to implement." "Can we get
the next release out within the next month because we are about to release our product and
want the latest features?", etc.
	2. If you are working on core TinkerPop day-in and day-out you are part of TinkerPop-Contributors.
(~3 people)
		- e.g. On the Google Hangouts, nit-picky about documentation, committing code and maintaining
your code…the people who baby the source.
	3. If you have worked on TinkerPop core at some point. (~5 people)
		- e.g. You were hot and heavy on the codebase for a month+ straight but then lost interest
but still want to hang around. Perhaps it looks good on the resume why you don't want to leave…
who knows?

Most people are in group 1 or 3. I was told by our champion (David Nalley) that you can't
have people be on the board because of who they work for. Thus, the concept of "one engineer
per company" is not acceptable. Next, I don't think it is smart to just have everyone in group
3 mapped over. I think its best to start with the minimum requirement of people and grow from
that core. Who are these three people?

	Marko -- has been coding on TinkerPop "day in and day out" for the last 5 years. 
	Stephen -- has been coding on TinkerPop "day in and day out" for the last 3 years.
	James --  has been an evangelist for TinkerPop for the last 5 years and is getting the TinkerPop
book effort underway.

These are the most "hardcore" members of TinkerPop. Now, once (and if) TinkerPop goes Apache,
I'm sure more engineers (either currently in "TinkerPop-Contributors" or new to the scene)
will want to make concerted efforts to be apart of TinkerPop. Through the years I have become
too painfully aware of "commit and split"-contributors ("here is a big ball of features, merge
it…oh, I can't work on that anymore, my boss doesn't care about graphs anymore. good luck
maintaing that code." --- many group 3 people are in this camp). Once we realize someone is
here "for the long haul" and truly cares about the project, we are happy to have them join.

"Lets start small and grow" is the philosophy behind the 3 initial committers.


Thank you,

On Dec 25, 2014, at 3:12 PM, "Dennis E. Hamilton" <> wrote:

> -- with reply below --
> From: Marvin Humphrey [] 
> Sent: Thursday, December 25, 2014 13:39
> To:; Dennis Hamilton
> Subject: Re: [Proposal] TinkerPop: A Graph Computing Framework
> On Thu, Dec 25, 2014 at 1:06 PM, Dennis E. Hamilton
> <> wrote:
>> I am puzzled by the TinkerPop proposal identifying 3 initial committers and
>> yet there is this long list of affiliated folks.
>> What is that separate list intended to signify if none of them are worthy to
>> be initial committers?
> If it's like other projects, I speculate that the people on that longer list
> have exhibited varying levels of activity over time.  I don't think it's a bad
> thing if the initial committer list is only a subset (though 3 is arguably too
> small even to start).  So long as the podling busies itself with the task of
> voting new people in, it should be OK.  I think the way Marko put it bodes
> well:
>    Thus, typically finding those people is the difficult part, not the
>    accepting of those who do so.
> <orcmid>
>   That list is noise, then, especially with no accompanying explanation of
>   how it supports the proposal.
>   With regard to finding committers, I notice that folks request 
>   being added to the initial committers list of a proposal and 
>   those are vetted one way or another by the proposer/champion.
>   Has the proposal been publicized to that group and any 
>   interest in being initial committers (and especially filing 
>   Apache iCLAs) elicited?
> </orcmid>
> Marvin Humphrey
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

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