I would add two things to the list of potential graduation blockers that an IPMC member reviewing the project is likely to pick up on
1) Off-list activity There is still seems to be lots of off-list activity that happens. While you've gotten a lot better at writing emails to the list that discuss what was discussed the fact is there shouldn't be extended discussions happening off list as much as they appear to do. Also often reading the threads it comes across as being fait acompli - "We [the PMC] discussed this thing in private and here's this decision(s) we made" - while you do ask for comments from the wider community the fact that the main discussion has already happened (and people don't have all the context that was in that discussion) makes it much harder for other people to feel like they have a say and can be involved in the decision making process. The recent thread on the JIRA review is a prime example of this and in this case you even had someone from the community complaining that you held that discussion in private 2) Vendor Focus A lot of your discussion still seem to focus on vendors as being a key motivator in decisions that are being taken e.g. the discussion about limiting changes to the proposed mutation APIs to avoid breaking Titan too much. As I and your mentors have said in the past Apache is based on communities of individuals and it is not a vendor organisation. It is perfectly fine and normal for people to work for vendors and expected that they will advocate for approaches that work best for them but at the same time community members should consider the health of the broader community ahead of their specific special interests. For example a PMC I'm involved in (Jena) recently moved to Java 8 only going forward, my employer relies heavily on this project in some of our commercial projects but for various reasons we cannot move to Java 8 yet. However for the project community moving onto Java 8 was the best move and I was happy to support that despite the fact that it will create some extra work for us in terms of back porting important fixes to the discontinued Java 7 branch. Personally I'd want to see more of this kind of attitude and less of the "we can't do that because we'll upset vendor X" - remember community != vendors Rob On 27/08/2015 17:35, "Daniel Gruno" <[email protected]> wrote: >In general, there are a few rules of thumb: > >- You need AT LEAST 3 active PMC members at all times. With 4 in the >PPMC (apart from mentors), this could be a problem. >- You need (preferably) a good Pony Factor (see David's blogs/talks on >that issue, or look at the derivatives (Elephant factor, Bus factor etc)) >- You need diversity in the PMC to prevent strongarming >- You need AT LEAST one ASF member in the PMC when you graduate. > >For some board members, few is okay, and for some, a larger number than >4 is non-negotiable. If you want, you can ask either David or Rich >personally, as they are both on the board. Take a look at the public >board meeting minutes from the past, and look at the number of >committers that graduating podlings generally have. > >With regards, >Daniel. > >On 08/27/2015 06:30 PM, Marko Rodriguez wrote: >> Hello Daniel, >> >> When you say "raise eyebrows" do you mean "no" or do you mean, we have >>to provide a rationale argument for why TinkerPop is currently the way >>it is. >> >> Perhaps more generally, who makes the decision and which problems do >>they specifically have such that we can address them either via >>fulfilling their want or changing their mind through discussion. >> >> Thank you, >> Marko. >> >> http://markorodriguez.com >> >> On Aug 27, 2015, at 10:26 AM, Daniel Gruno <[email protected]> wrote: >> >>> Community, community, community. >>> >>> If you were to graduate now, the board would raise their eyebrows at >>> there only being 4 committers, and would insist that all mentors join >>> the PMC (we are sort of free to _not_ join it normally when you >>> graduate). and even then, 8 people is not a whole lot of community. >>> >>> Get your non-mentor community to at least double the size (preferably >>> more!), and we can discuss some more detailed goals. >>> >>> With regards, >>> Daniel. >>> On 08/27/2015 06:21 PM, Marko Rodriguez wrote: >>>> Hello Rich (cc: others), >>>> >>>> What still remains for TinkerPop to get out of incubation? >>>> >>>> Thank you, >>>> Marko. >>>> >>>> http://markorodriguez.com >>>> >>>> On Aug 10, 2015, at 5:59 PM, Rich Bowen <[email protected]> wrote: >>>> >>>>> Thanks. I've signed off. Perhaps before the next cycle we need to >>>>>evaluate >>>>> what still remains to graduation, and make plans of how to reach it. >>>>> On Aug 10, 2015 4:35 PM, "Marko Rodriguez" <[email protected]> >>>>>wrote: >>>>> >>>>>> Hello, >>>>>> >>>>>> I appended TinkerPop's section to the end of the document. >>>>>> >>>>>> I believe that each mentor needs to [x] sign off. >>>>>> >>>>>> Next, I don't know what else we need to do to graduate so I just >>>>>>said >>>>>> "Grow Community." (seems nebulous). >>>>>> >>>>>> Thanks David, >>>>>> Marko. >>>>>> >>>>>> http://markorodriguez.com >>>>>> >>>>>> On Aug 10, 2015, at 2:20 PM, David Nalley <[email protected]> wrote: >>>>>> >>>>>>> Hi folks: >>>>>>> >>>>>>> There currently isn't a Tinkerpop report in the larger incubator >>>>>>> report. We still have a short period of time that we can add it, >>>>>>>and >>>>>>> it would be nice to do so if at all possible. >>>>>>> >>>>>>> http://wiki.apache.org/incubator/August2015 >>>>>>> >>>>>>> Who is going to take charge on accomplishing this? >>>>>>> >>>>>>> --David >>>>>> >>>>>> >>>> >>>> >>> >> >> >
