Also, as efforts become more interconnected, working groups have
"customers" even within the IETF. It's not good if working group A can't
proceed in publishing a spec because of normative references from
working group B that can't get its act together. In that sense,
milestones are a contract between a working group and the larger
community. In particular, it's a contract between the spec
authors/editors and the larger community. A spec in a WG work item
generally gets some amount of exclusivity, in that the working group
isn't going to consider a similar spec by a competing set of authors.
(There are exceptions in highly contentious working groups, obviously.)
However, that only works if the authors deliver on time. Otherwise, an
effort should have been made to find (cajole, entice, browbeat,
sweet-talk) authors that have more time.
Milestones that are enforced are also useful to make resources
available. Speaking from recent working group experience, if a deadline
matters, employers are suddenly that much more willing to let people
work on the spec. Managers tend to understand the notion of a deadline.
Michel Py wrote:
Dave Crocker wrote:
and, by the way, there is plenty of experience suggesting that
time pressure often improves quality. it focuses the group
and emphasizes near-term utility. within discussions about
project management, it is usually recognized that milestones
are not merely for measuring progress but also for motivating
it.
I agree. In the case of the IETF, the worst part is that the milestones
and/or deadlines are self-chosen. I have complained myself about some
marketing bozo fixing arbitrary or unrealistic deadlines just because the
product has to ship that day. This is not the way the IETF works. First, we
can choose our own deadlines and second we can move them if we
under-evaluated the time it would take. What is wrong is the generalized
feeling that missing a deadline by two years (for a 6-month project) is ok.
And the same is true for WG charters. Why do we have charters and milestones
if nobody tries to meet them? Is it the way IETFers work at their day-time
job?
And please, don't give me the crap "we do it for free". I do it for free,
too. Quality has a little to do with money and a lot with engineers being
interested, motivated, intellectually challenged, that kind of thing.
Michel.
-
This message was passed through
ietf_censored(_at_)carmen(_dot_)ipv6(_dot_)cselt(_dot_)it, which
is a sublist of ietf(_at_)ietf(_dot_)org(_dot_) Not all messages are passed.
Decisions on what to pass are made solely by Raffaele D'Albenzio.