On Thu, Oct 25, 2012 at 11:58 PM, Ate Douma <[email protected]> wrote: > On 10/25/2012 07:44 PM, Chris Geer wrote: > >> On Thursday, October 25, 2012, Jasha Joachimsthal wrote: >> >> Hi Chris, >>> >>> it has been more quiet indeed on the mailinglist, but there is still a >>> lot >>> of development going on. We now have 2 branches and a sandbox for the >>> model >>> split, MongoDB and content services. I hope they will merge into trunk >>> soon, so it's more visible that new functionality is developed, and it >>> will >>> decrease the chance of merge problems. >>> >> >> >> You are right, maybe "momentum" was the wrong word choice. Maybe >> collaboration would have been better. I'm with you, I look forward to >> getting some of these major overhauls integrated. Just in time for the >> next >> round of major overhauls :) >> >> Right :) > > Let me first say I fully agree with the feedback from Ross and the > thoughts about the practicalities of a roadmap. I'm all for cleaning and > clearing up our longer term goals. Some of which maybe are outdated or no > longer desired. And reaching agreement on a short term and practical > feature set for a 1.0 release seems good to strive for. >
At the end of the day, I also agree having a long term roadmap isn't useful, I just picked it as a term because everyone understands the concept and we have one on our wiki (kind of). I guess my intent was to get people thinking about the big impact changes we wanted to see before 1.0. > > I also agree having multiple major overhauls in separate branches doesn't > seem to work well from a collaboration and community perspective. > Especially not if they are too widely scoped, experimental or time > consuming, like what I think the content integration sandbox turned out to > be. > Maybe we should strive for not spawning off more than one active branch at > a time? And keep them as short-lived and concretely scoped as possible. > Well said, if I could just take parts of everyone's responses and redo my email I think it would have been much better :) Maybe after we get the model-split branch merged, and content-services wraps up we can really dig into any major things we want to see over the next couple months. > > With respect to our sandbox work for content services integration: I'm in > the process of wrapping that up, technically I think it is about done (for > a sandbox) and IMO worked out quite well actually. But it also gave us some > new insights causing us to pause and ponder a bit more about how to proceed > from here. And that very much does relate to the roadmap and how we want to > leverage and integrate Rave in our environments. > Not going into detail here now, I will come back on this next week and > start the discussion about if or how to merge and integrate some or all of > these features into the trunk. I'm looking forward to seeing this. > > Ate > > >>> On 25 October 2012 18:55, Raminderjeet Singh <[email protected]<* >>> *javascript:;> >>> >>> wrote: >>>> >>> >>> Thanks Chris for taking a lead to identify the problems with Rave >>>> activities and Roadmap. I think everyone will agree with you here. I >>>> >>> admit >>> >>>> that i missed your email Model Split email. We have requirements where >>>> we >>>> want to customize user model so i will take a look at the branch this >>>> >>> week. >>> >>>> I volunteer for the feedback. >>>> >>>> My 2 cents on development activity is we need to set some Roadmap for >>>> >>> Rave >>> >>>> 1.0( a wish list). We should not keep going with 0.xx. That will bring >>>> everyone back with their choices about Rave 1.0. We can setup some >>>> developer hackathon using Google hangout etc between now and ApacheCon. >>>> >>> We >>> >>>> made a great presence in ApacheCon 2011. We need to do the same for >>>> ApacheCon 13. ApacheCon can be set as a milestone for Rave 1.0 (just an >>>> idea). To achieve that milestone community need to start talk about 1.0 >>>> >>> and >>> >>>> document it on wiki [1]. We need to expand on little details on >>>> architecture documents also for new users to start. >>>> >>>> >>> The next ApacheCon starts in 10 days ;) Unfortunately I can't be there, >>> but >>> would love to join a hangout in one of the evenings. I know some of the >>> other Rave committers will be there. >>> >>> Jasha >>> >>> >>> >>>> I agree with Ross that we can better spend time implementing than >>>> documenting but have a wish list for 1.0 can give a better direction to >>>> >>> the >>> >>>> community. >>>> >>>> 1. >>>> http://wiki.apache.org/rave/**RoadMap<http://wiki.apache.org/rave/RoadMap> >>>> >>>> Thanks >>>> Raminder >>>> >>>> On Oct 25, 2012, at 12:36 PM, Ross Gardler wrote: >>>> >>>> I'm a little out of touch with Rave as I'm busy elsewhere, as you >>>>> surmise many people in your thoughtful mail below. >>>>> >>>>> In general though it is normal for a community to go quiet >>>>> periodically. Things tend to happen in waves. When one individual gets >>>>> busy they motivate others to get busy simply through their actions. >>>>> >>>>> I'm never much of a fan of roadmaps being formally defined by a PMC. A >>>>> projects roadmap is set by the people with time to work on it at that >>>>> point in time. Putting a list of things that the PMC considers >>>>> important on a website doesn't make those things happen. That being >>>>> said, a little periodic re-evaluation of objectives and updating of >>>>> the website certainly doesn't hurt. For example, I noticed that the >>>>> home page still reflects the very high level roadmap that was defined >>>>> during at proposal to the incubator. Some of the items on the "future" >>>>> list have been implemented. Of course, I could have updated the home >>>>> page when I noticed this, but I didn't because I was too busy at the >>>>> time (and still am). >>>>> >>>>> If you have the time and the desire to define a roadmap that you think >>>>> is important and you wish to ask for feedback on roadmap then go for >>>>> it. However, personally, if I find I have some time for Rave I'll >>>>> probably spend that time just implementing what is at the top of my >>>>> personal agenda. >>>>> >>>>> Finally, with respect to your mail that has had no response, lazy >>>>> consensus means that no objections means agreement. I've reviewed your >>>>> mail but don't have any opinion on it since I have not reviewed or >>>>> tested your code. But you are a committer so I'm +0 on supporting your >>>>> recommendations (not +1 as I don't have the time top properly review). >>>>> If you want a little more explicit support for the proposed merge then >>>>> post a reply to your own mail saying "no feedback so I assume all is >>>>> good. I'm going to merge in the next few days". That will bring it >>>>> back to the top of peoples inboxes and possibly prompt more review. >>>>> Finally, once you have done the merge if it all goes wrong or someone >>>>> wants to object for a solid technical reason the changes can be rolled >>>>> back. >>>>> >>>>> In summary, you are a committer. We operate a commit then review >>>>> policy here so don't be afraid to just get on with it :-) >>>>> >>>>> Ross >>>>> >>>>> On 25 October 2012 17:13, Chris Geer <[email protected]> wrote: >>>>> >>>>>> Over the past couple months there has been a growing worry of mine >>>>>> >>>>> that >>> >>>> we >>>> >>>>> are loosing some momentum/cohesiveness as a team. As a metric, if you >>>>>> >>>>> look >>>> >>>>> at the dev mailing list, the traffic for the past three months has >>>>>> >>>>> been >>> >>>> very low (September was awful). What finally spurred me to send this >>>>>> message was the fact that I sent an email [1] on October 13th, asking >>>>>> >>>>> for >>>> >>>>> some help reviewing a major change Matt and I had been working on and >>>>>> haven't gotten a single response. I know it's not a sexy change but >>>>>> >>>>> it's >>> >>>> something that people thought was a good idea when it was proposed a >>>>>> >>>>> while >>>> >>>>> back. What I don't know is if the lack of response is because people >>>>>> >>>>> are >>> >>>> too busy, they don't care or they don't support the change but don't >>>>>> >>>>> want >>>> >>>>> to say that. The third option would concern me the most since we >>>>>> >>>>> should >>> >>>> feel free to provide feedback, both positive and negative. >>>>>> >>>>>> My suspicion is that people are just swamped at their day jobs and >>>>>> >>>>> Rave >>> >>>> has >>>> >>>>> taken back (maybe far back) seat to normal life which is >>>>>> >>>>> understandable. >>> >>>> With that knowledge though, as a PMC we should probably spend some >>>>>> >>>>> time >>> >>>> really coming up with a priority list of items we agree need to get >>>>>> >>>>> done< >>> >> >> >
