Very useful.
On Tue, Jun 4, 2019 at 9:22 PM Adrian Cole <adrian.f.c...@gmail.com> wrote: > Hi, all. > > Through Zipkin's incubation, I noticed that knowledge of state of the > art is not equally distributed. Some voting approaches already in use > aren't known. Also what you can use to automate isn't known. As > podlings don't always know the right questions to ask, it might be > beneficial for future podlings if some advice about state of the art > was taught instead. This is especially important as many projects will > likely want to import multiple repos. I will inventory some. > > * "parallel votes" is a technique to reduce the lag between dev@ and > general@ by starting the IPMC vote slightly after, but before > conclusion of the PPMC one. This may have only been tried once (in > Zipkin) and met resistance. > * "bundled votes" is a technique where multiple repos are sent in a > single vote. This is much more efficient for coupled repos than > pipelining. OpenWhisk uses this and don't appear to have met > resistance. > * not all repos need to actually release prior to graduation. Both > Dubbo and OpenWhisk had unreleased repos when they called for > graduation (OpenWhisk has dozens). It isn't intuitive for projects to > know they don't have to release every repo, so they should be told > this. > * Jenkins is most ASF, but Travis is ok, but CircleCI is not, and > maybe Appveyor is ok. All we know now, is that CircleCI is not ok, > which is personally fine, but basically people should be able to know > what CI tools are allowed to be used. > * You are allowed to automate release process, even with Travis. It > can be confusing to know what you are allowed to automate, and whether > it must be done with ASF infra etc. The most elaborate example is > https://github.com/apache/incubator-openwhisk-release > * ASF tools like RAT and the release plugin are barely maintained in > comparison with Incubator policy. It is unintuitive that tools > projects use to audit themselves don't evolve lock-step with general > software, policy and discussion. This feels bad whenever told, but > earlier is better. > > Anyway, figured I would send off these notes so that folks entering > the incubator don't end up with a lot of unnecessary grief, irritation > due to ignorance about state of the art, or the disparity between > project-specific ASF tools and general ASF tools. > > -A > > --------------------------------------------------------------------- > To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org > For additional commands, e-mail: general-h...@incubator.apache.org > >
