Hello Streams Community, I finally put some more detail around each of the key project priorities on the confluence wiki:
https://cwiki.apache.org/confluence/display/STREAMS/Roadmap I invite your review, edits, and discussion on wiki or on this thread. Also, there are 5 open pull requests I’d appreciate reviews for. I think we can begin the release of 0.6.1 immediately there-after. Steve Blackmon On Jan 20, 2019, 10:51 AM -0600, Steve Blackmon <[email protected]>, wrote: > Thanks for starting this thread. Responses in-line. > On Jan 19, 2019, 5:07 PM -0600, Joey Frazee <[email protected]>, > wrote: > > I don’t think it should be super surprising to anyone to say that Streams > > could be having a bigger impact if we could kick our activity up a notch. > > The hard thing is figuring out how to get us there, so I wanted to open a > > discussion on that. > > > > I think the following would be good places to start: > > > > a. If we had a formalized roadmap with more starter or bite-sized tasks, I > > think it’d make it easier for new or existing contributors alike to spend > > time on the project. Yes, some of this is in JIRA, but I think making it > > more of a document too would be useful. > > > > > We’ve accumulated a short list of roadmap initiatives, each a good deal of > work potentially, with potentially high impact on project capability or > quality, enumerated here: > > https://github.com/apache/streams/blob/master/src/site/markdown/roadmap.md > > Maybe a good place to start would be to work through the specifics of each on > the wiki, with the goal of defining success for each, and breaking up into > actionable stories. > > > b. There’s also the current backlog of course. As part of (a) I think we > > can talk about what in the backlog we really want to do. > > Our backlog clearly needs signficant culling to make room for the community’s > current priorities. > > Related, we have modules (some with open tickets) which have seen no activity > since graduation or longer > > Any that aren’t essential to the projects core mission of data > interoperability I think we should discuss retiring and deleting. > > > > c. We can’t ignore community development of course. Craig Russell has > > encouraged a lot of projects to look at some of what Beam did [1] and I’m > > pretty sure most of us have read it. It’s probably worth having a second > > look at though. > > > > > This is true; every board report we’ve acknowledged that attracting new > users, committers, and PMC members is important - but we’ve lacked a > coordinated approach and the discipline to succeed at this. > > > d. If we wrap all this up into some quarterly goals, we’ll know what we’re > > working towards. I think if we got (a) done that’d be a great place to > > start. > > > > > I’d love to see us retire dead code, get a binary and source release out, > clean-up the backlog, and spec out a year’s work of specific tasks (at > current team size) by April. This is achievable if we rally as we have > before. > > > What do you all think? > > > > -joey > > > > > > 1. > > https://lists.apache.org/thread.html/373e1c1427f16142a3ffd7c2f9b3639bf3c343c80e9c893bd861a740@%3Cdev.beam.apache.org%3E
