Going to Last Call

Hi,

Part of our call today will be dedicated to planning for moving Media
Capture and Streams to Last Call Working Draft.

Last Call in the W3C process is the signal that the Working Group
believes the relevant specification addresses the requirements it was
set to fulfill, and assesses that there is no open issue about the said
document.

To determine when we are ready to go to Last Call, the Chairs will be
looking at our issues list; when that list gets empty, the Chairs are
entitled to call for consensus to go to Last Call.

Given our intent to go to Last Call, Working Group participants should
start to make sure the issues they know have been identified as such. In
an ideal world, Working Group participants no longer need to raise issue
once Last Call has started.

When we go to Last Call, we will need to identify W3C Working Groups and
Interest Groups from which we specifically want to draw attention on the
spec; I think for Media Capture and Streams, the following groups would
make sense:
* Web Apps WG
* Audio WG
* Privacy Interest Group
* Web & TV Interest Group

We should also request feedback from non-W3C groups that have a
particular stake in the document; I think the IETF RTCWeb WG and the
ECMA TC39 (responsible for JavaScript) would make sense.

The Last Call period lasts a duration determined by the Working Group,
but no less than 3 weeks; depending on other factors (e.g. summer time),
we might want to make that window wider to ensure we get feedback from
as many stakeholders as possible.

We will need to process formally each and every comment we get during
Last Call, determining in particular if the resolution we reach on the
said comment satisfies or not the commentor; if it doesn't, we will need
to determine whether the said person wants to file a Formal Objection.

Once we have integrated all the feedback from Last Call comments in the
spec, we can either:
* go back to Last Call if the changes brought to the specification are
deemed substantive
* request transition to Candidate Recommendation to the W3C Director; in
that case, we will need to justify to the W3C Director that we followed
the process as expected, in particular with regard to comments handling

Depending on the amount and difficulty of the comments we get, handling
all the comments can range from trivial to very time and energy
consuming; given that we already have relatively good implementation
experience, I'm rather on the optimistic side of this scale though.

Once we go into Candidate Recommendation, we will need to switch our
focus toward gather implementation feedback, esp in light of test suite
results; in particular, we will need to further develop our test suite
to ensure we have good confidence that all of the spec is implemented
correctly at least by two independent implementations.

Dom

Received on Tuesday, 7 May 2013 14:59:30 UTC