-------- Original Message --------
Subject: RE: [chairs] When is a TCs work done?
From: "Dennis E. Hamilton" <
dennis.hamilton@acm.org>
Date: Fri, July 16, 2010 6:38 pm
To: <
robert_weir@us.ibm.com>
Cc: <
chairs@lists.oasis-open.org>,
"'David RR Webber (XML)'"
<
david@drrw.info>,
"'Martin Chapman'" <
MARTIN.CHAPMAN@ORACLE.COM>
Oddly, I can't find an example of any closed TC being closed
because it missed a deliverable.
All of the cases I see are usually because of lack of
activity and ability to maintain an active membership and
hold meetings. It tends to be rather stark when a TC runs
out of gas, even when there is a heart-beat post to the list
every month despite there being no list activity
representing work of the TC itself.
In these cases, it appears it is not that a deliverable is
late but that a deliverable is not happening and there does
not appear to be anything in place to have there be a
different outcome.
I am sure that appeals are possible and that might happen
(be happening) with any of those TCs which are currently
identified as candidates to be closed.
Going back to the post that began this thread, I find it odd
that there is any expectation that OASIS is responsible for
providing life support. It seems to me that a viable TC will
have ample means to demonstrate its sustainability. It is
not a very harsh criterion to require an ability to hold
regular meetings and maintain a minimal quorum with
published minutes as evidence of continuing health.
Determining that there is meaningful forward progress is
harder. Evidently the defection of participants from an
unproductive effort provides an easier and more-decisive
indicator, making trickier and subjective inspection for
forward progress less necessary. I have not seen a deadline
reached on time in any TC I am involved in, and I am yet to
see any raised eyebrows from OASIS so long as it has been
obvious that we have been moving forward with a prospect of
convergence on a result.
I am currently an official of a TC that is borderline in one
aspect: we have exactly 5 Voting Members and there have been
times when there were fewer. But there have been regular
quorate teleconferences at least monthly as long as I have
been a member. At the same time, we completed a 60-day
Public Review of three inter-related and mostly-new
documents in May, we are processing comments and making
important improvements to those documents for a second
15-day public review later this summer, and I expect we will
achieve Committee Specification status shortly after that.
In the short time I have been a member, I don't recall
seeing a TC Closure e-mail from the TC Administrator.
I'd say that the case I have in mind is the least
sustainable size for an OASIS TC activity and having such a
small community of active participants is less than ideal.
We are probably a marginal and borderline effort, whatever
one might consider to be the worthiness of the work. I say
it is noteworthy that OASIS is so tolerant of something so
specialized and of limited attention in the world that five
of us are able to soldier on and provide material
deliverables that pass public scrutiny.
The cut-off conditions that OASIS TCs operate under are, in
my analysis and experience, quite generous.
- Dennis
-----Original Message-----
From:
robert_weir@us.ibm.com
[
mailto:robert_weir@us.ibm.com]
Sent: Friday, July 16, 2010 14:31
To:
dennis.hamilton@acm.org
Cc:
chairs@lists.oasis-open.org;
'David RR Webber (XML)'; 'Martin Chapman'
Subject: RE: [chairs] When is a TCs work done?
We are very fortunate that we have some very nice and
reasonable TC Admins
today, and they are are not closing TCs just because they
missed a
deliverable by a week. I'll grant you that point. But you
must grant me
that the process, as written today, explicitly permits TCs
to be closed
purely for that reason. That is my concern.
By all means we need a way to prune out abandoned TCs. I
have nothing
against that. But personal I wonder whether the criteria, as
written are
overly-broad and the checks and balances absent? I think
this would be
fairly easy to address in some future revision of the TC
Policy. If a TC
has truly been abandoned or is moving in circles or
aimlessly, then it
probably doesn't hurt anyone if we had some safeguards that
turned this
from a one-person kill clause to a lightweight procedure
that guaranteed
the opportunity for stakeholder and member feedback over,
say a 30-day
period.
Nothing urgent, just something to consider.
-Rob
"Dennis E. Hamilton" <
dennis.hamilton@acm.org>
wrote on 07/16/2010
04:32:24 PM:
>
> Well, before we get too carried away about the
speculative hazards
> of tyrannical TC Administrators, I have some anecdotal
data-points on
this.
>
> There are a number of OASIS TCs that are definitely
inactive in the
> sense that there is nothing happening with documents,
there is no
> activity on the list, and there is no indication of
meetings in a
> considerable time. This also includes situations where
the chair
> keeps scheduling calls that don't happen for one reason
or another
> (and cancelling others for one reason or another).
>
> I have also reviewed mailing lists (as part of wanting
to know the
> status of work that I was checking on) and observed
inquiries from
> the TC Administrator to determine whether the committee
is really
> active or is it ready to be closed. My impression is
that there is
> nothing precipitate about the closing of a TC and that
it is not
> undertaken lightly. Also, these inquiries don't happen
until there
> has been a prolonged period of inactivity - more like
years, not
> months, and action is not immediate or precipitate even
then. It
> appears that there not only has to be no life in the
corpse, it has
> to be overgrown with weeds and returned to dust. Even
for
> committees which, it seemed to me, were/are really
nothing but one-
> man bands (zombie TCs?), the TC administrator is very
cautious.
>
> I conclude that there is no trip wire or time bomb by
which a TC,
> once constituted, flushes down the drain like a "dunk
me" target at
> a summer amusement park. That's true of all of the
cases I've
> researched. I have at one time or another reviewed
every OASIS TC
> that has produced an OASIS Standard in this century,
along with many
> that never will (because the need the standard was
intended to serve
> has disappeared or been satisfied another way, because
the standards
> are published in a different venue, or because the TC
has a purpose
> in which achieving an OASIS Standard is not the primary
goal).
>
> - Dennis
>
> -----Original Message-----
> From:
robert_weir@us.ibm.com
[
mailto:robert_weir@us.ibm.com]
> Sent: Friday, July 16, 2010 12:23
> To: Martin Chapman
> Cc:
chairs@lists.oasis-open.org;
David RR Webber (XML)
> Subject: RE: [chairs] When is a TCs work done?
>
> Just looking at the relevant clause:
>
http://www.oasis-open.org/committees/process.php#closingTC
>
> In full it says:
>
> "The TC Administrator may close a TC that fails to
conduct at least one
> Quorate Meeting or conduct any Specification Ballots
during any six
month
> period; whose membership falls below the Minimum
Membership; which has
not
> completed its deliverables within the schedule listed
in its Charter; or
> which has failed to show progress towards achieving its
purpose as
defined
> by its Charter."
>
> I'd wonder what % of OASIS TCs actually complete their
deliverables
within
> the schedule they predicted at the time they chartered
it? I bet it is
> quite small. Nothing special about OASIS. Schedule
estimation is hard
in
> general and we all tend to be overly optimistic. But
regardless, one
> man's "lack of progress" is another man's "deliberate
pace with
consensus
> building". Some of the best breakthroughs come after
"lack of
progress",
> including good stuff worth waiting for, Hopefully we're
not triggering
TC
> closures based on those criteria very often.
>
> I realize that this is always going to be a judgement
call, and you want
> to look at the totality of the facts and circumstances.
So I'm
surprised
> that this decision can be made by a single TC Admin. I
would have
thought
> that there would be at least some minimum notification
time, comment
> period, access to a Board appeal, etc. Closing a TC --
absent approval
of
> the TC members -- does not feel like an administrative
action to me on
par
> with sending out the announcement of a public review,
or conducting a
> committee specification ballot. It sounds like a far
graver action,
which
> should have some more checks and balances behind it.
>
> My personal opinion, of course.
>
> -Rob
>
> Martin Chapman <
MARTIN.CHAPMAN@ORACLE.COM>
wrote on 07/16/2010 12:22:59
> PM:
>
> >
> > So what is the exit strategy for your TC?
> > IMHO, the only way to avoid these automatic
shutdowns is for every
> > charter to be explicit enough so that TC members
themselves can
> > declare victory and shut themselves down. Also the
criteria isn’t
> > that onerous: maintain minimum membership (I’m
sure temporary lapses
> > are tolerable), and hold a quorate meeting every
six months!
> >
> > Cheers,
> > Martin.
> >
> >
> > From: David RR Webber (XML) [
mailto:david@drrw.info]
> > Sent: 16 July 2010 12:58
> > To:
chairs@lists.oasis-open.org
> > Subject: [chairs] When is a TCs work done?
> >
> > I'm reminded of the maxim in any good initial
business plan asking -
> > what is the exit strategy?
> >
> > Seems that OASIS has criteria based mainly around
number of emails
> > posted, who's posting them (apart from the TC
chair) and how many
> > meetings and minutes you have posted.
> >
> > As TC chairs however - I think we deserve more
support than OASIS
> > hitting our TC with FUD messages to bolster
continuation of the
> > technical work.
> >
> > Several members I recently canvassed told me they
would like to do
> > more than observer but their company is
restricting hours and
> > requiring formal manager approval and
justification for any new TC
> > related work - even just reading emails or joining
a group. Given
> > those types of challenges its little wonder that
typical TC work is
> > being driven by just a handful of individuals.
> >
> > OASIS needs to therefore do more in terms of
assisting garnering
> > support for our work. So for example - one simple
thing I notice
> > that is misleading - is that Kavi only shows
voting members - what
> > should be shown also is the total number of
observers (just the
> > count), and non-voting members underneath that
also on the whole
roster.
> >
> > Clearly TC chairs have a huge role in continuing
work of a TC. In
> > the lifecycle of a standard it is way more than
just calling
> > meetings, writing specifications and publishing
schema.
> >
> > Rather than FUD messages from OASIS staff to our
TC - we need more
> > informal coordination to help with members who may
be contemplating
> > contributing - or just testing the pulse = looking
to help get more
> > involvement and so on by working with the TC
chairs and reaching out
> > to potential new resources.
> >
> > Also - chairs usually know way more about what is
really going on.
> > The mailing list only tells one small part of the
picture - in terms
> > of what is external parties are doing, or planning
to do with a
> > specification, or additional potential resources
to advance new work.
> >
> > The current administrative door slamming by OASIS
seems to be based
> > solely on reducing the number of TCs to some
acceptable lower number
> > - rather than any rationale based on the
importance of work - and
> > need to actively foster and help TC chairs gain
support either
> > within their TC or with external industry groups
or academic
> > institutes who may benefit or contribute further.
> >
> > Everyone is burned out of course on standards work
- and its now
> > layers of burn out over burn out. Now in tough
economic days it
> > seems that bean counting and ROI have totally
taken over the
> > equation of specification development - rather
than anything else
> > relating to technical value and incubating
potential groundbreaking
> > or interesting XML capabilities within OASIS.
> >
> > Ironically the small independent members would
appear to be those
> > that have the most flexibility to continue OASIS
work and yet OASIS
> > itself it set to penalize them for trying!
> >
> > Thanks, DW
>