[CF-metadata] standard names committee

From: Pamment, JA <J.A.Pamment>
Date: Fri, 16 Feb 2007 15:06:42 -0000

Dear Jonathan, Roy, Nan and Tom,

sorry for the slow response on this - I am working only a few hours this week
due to school holidays.

When I raised the subject of reaching decisions I certainly did not mean to
suggest that we should impose rigid or arbitrary time limits on the process of
agreeing new standard names. Thus far I have adopted a pragmatic approach in
which threads that are clearly actively under discussion have been left to run
until, hopefully, a consensus is reached at which point the names have been
listed for inclusion in the table. I see no particular reason to alter that
approach. Indeed, there can be very good reasons why a discussion takes a long
time, for example, to allow for consultation with experts outside the CF
community as is happening currently with the aerosols and chemistry proposals.
I feel very strongly that it would be a mistake to cut short efforts to agree
names on a sound scientific basis purely in the interests of speeding up the
process.

A problem arises when a thread has been dormant for a while, say two or three
weeks without any postings. In that case it is tempting to think that the
discussion has run its course and that people are happy to proceed with whatever
position was reached before the postings stopped. However, I have learned that
it is dangerous to assume under those circumstances that anything has actually
been agreed! To a great extent I think the responsibility for restarting
stalled discussions lies with me, through prompting for posts on the mailing
list and/or contacting proposers directly. I agree with Roy's suggestion that
we could have a "guillotine date" where, after prompting and if no further
comments have been received, we can either proceed on the basis of what has
already been discussed or agree to drop the proposals from further discussion.
The committee would need to decide which course of action to take. If
proposals were dropped there would be nothing to prevent them being reintroduced
at a later date but they would then have to go through the whole process again
from the beginning.

Another major benefit of having the committee comes if a discussion has reached
a point where there is no overall consensus and no clear way forward. This
situation does arise on occasion, often because there are only two or three
participants in a particular conversation each advancing a different opinion.
Under those circumstances it does not seem appropriate for me, acting as an
individual, simply to impose a decision on the community by opting for one
particular point of view. That is especially true if the topic under discussion
is outside my own area of scientific knowledge. I do think it is likely that
there will be a need to ask the committee to vote from time to time.

Having looked after standard names for some months now I am gradually learning
through experience how to manage more smoothly at least some aspects of the
process. I have learned that I do need to be pro-active in keeping things
moving and publishing the schedule of planned updates is one step towards this.
Another step is to announce in advance of an update the names that are planned
for inclusion - this alerts people to the fact that a change is about to take
place and also gives them one last chance to air any objections to the names
going forward. The last point is important because names, once included in the
table, are not taken out again so there is a clear need to avoid making
erroneous or only semi-agreed additions. Based on recent experience, I think
that names planned for inclusion should be announced two weeks in advance of an
update. In the past I have given a week's notice but giving the extra time
makes some allowance for the possibility of people being on leave, attending
conferences, etc.

Regarding the use of trac, I agree with Jonathan's comments. Certainly my hope
is that having separate tickets will make it easier to monitor progress on
different issues. Life may get complicated if people change subject headings or
raise new tickets part way through a conversation as sometimes happens now on
the mailing list. There is bound to be an associated learning curve. However,
I think the timescales for the standard names process are definitely something
that can only be arrived at by discussion and my gut feeling is that they will
vary between one set of proposals and another, even if that is within an agreed
framework.

There are other comments I could make regarding the standard names process, but
I think that is enough from me for now. I am pleased that we are having this
conversation and I do hope that we can at least agree some guidelines for how
the committee will work.

Best wishes,
Alison
Received on Fri Feb 16 2007 - 08:06:42 GMT