The date on the Jakarta CDI 3.0 release is bogus and needs to be updated.
I would prefer to see something in the description that describes what the release involves. If it includes DI, then IMHO that should be indicated here. This is no different from what I'd expect (both as an EMO and PMC representative) for any other project: a concise description that tells me the key features of the release. In the case of a specification project, the key features are the names and version number of the specifications included in the release.
If a project team wants to split them up and have separate release records, they're more than welcome to do so. In general, we've done this in the past with Eclipse MicroProfile because MicroProfile releases individual components on separate schedules.
On my first read, it seemed these two answers didn't align.
Wayne says the release record must be created before the Ballot
begins. In the case of DI, that would imply this must be done
before we can begin a ballot for DI -- (conversely, it can't wait
until CDI, which is further down-stream).
Perhaps, Kevin you are saying that splitting the release record
into multiple parts is optional.
In the CDI project, there is a Release 3.0, with a date of June
15, 2020. Perhaps this is sufficient and I can just check this off
even though the date seems to be the original date and there isn't
anything indicating both Dependency Injection and Contexts and
Dependency Injection are included -- perhaps that is implied --
and I'm fine with that.
Here is the governance
link for CDI project. I'm sorry, I don't know if there's
anything specific to check for, other than, it seems to exist and
says 3.0 (presumably that's CDI version) and there's no reference
to DI (2.0) -- and that wasn't identified in the previous release
either.
We have the
top
level release record for Jakarta EE 9, but I also created one
for Web Profile
9 and Managed Beans 2.0. This is a similar practice as what we
do
for MicroProfile. I agree with Wayne that this may not be
necessary
for the actual reviews and EMO processes, but it does help with
communicating
the intent.
So, if I was
reviewing
Dependency Injection, I would suggest that they create a
separate Release
Record for DI 2.0 under the CDI project. We can't state that
it's
a requirement (per Wayne's note), but it's a nice to have.
---------------------------------------------------
Kevin Sutter
STSM, MicroProfile and Jakarta EE architect @ IBM
e-mail: sutter@xxxxxxxxxx Twitter: @kwsutter
phone: tl-553-3620 (office), 507-253-3620 (office)
LinkedIn: https://www.linkedin.com/in/kevinwsutter
The release record serves as a means
of communicating release plans to the community. As such, it
should be
created as early in the development cycle as possible. It also
serves as
the anchor for reviews and EMO processes around reviews and
must
be created before we start the ballot.
Release records are for projects. As
such, a single release record is sufficient for a single project
that is
releasing multiple specifications at the same time. The
description field
in the release record should be used to describe the nature of
the release
(e.g., list the specification releases included).
Wayne
On Sat, Jul 18, 2020 at 12:52 PM Ed
Bratt
<ed.bratt@xxxxxxxxxx>
wrote: Hi, In the Specification committee Operations
Guide, under
the section "Creating
a Final Specification," step 8 indicates that a formal release
record
must be created. I'm currently reviewing Dependency
Injection
(wave 1) which is included in the Context and Dependency
Injection project.
Presumably, a release record won't be created until the CDI spec
(wave
3) is readied for release. I presume this is okay, but could
someone
confirm? This implies that all the formal EMO
and PMC work won't be done, before this specification has
completed its
ballot. Assuming this is accptable, will these notifications
(for CDI)
need to include references for both DI and CDI? (I think a
consequence
of this is that all the EMO and PMC validation work will be done
after
the DI ballot has been completed. I don't know if that is
acceptable, or
not.) Thanks, -- Ed