AdamBrousseau

Yea I'm not sure how this is going to work. The javadoc is built from a branch of OpenJ9. Each automated PR would contain the changes from the previous build/PR....

To confirm my understanding, if we build the doc from an OpenJ9 release branch, it will only contain doc changes set to go out in that release?

To summarize, it sounds like we can have our regular cadence build pushing all changes to master and a second build at release time which pushes to a release branch....

To confirm, PRs held back at OpenJ9, and only delivered once Docs has split (not when openj9 has split).

Ok I think I'm sorted now. We will regularly build master, which will contain latest OpenJ9 with Javadoc changes. OpenJ9 will split for release early. We will continue to build...

I have not seen or used the log parser plugin before. It does look interesting but I worry that it is not very actively maintained and released so it may...

I'm not against trying the log parser plugin in a test env, like hyc-runtimes or better yet another internal instance (not openj9). If we have to do the analysis ourselves...

@johnoliver looking for feedback please. I have started down the path of copying from the heap_size code. I assume I'll need to add some metadata to the build json too.

Another possibility we thought of was to add 2 vendors. `ibm_oe` and `ibm_ce` for Open Edition and Certified Edition respectively. We thought that even though it may not be the...

Since this was closed (which is fine), I'll provide an update on what we did with the Semeru API. We ended up implementing using the multiple vendor method I mentioned...