Part 2 - TC Process RE: [tm-pubsubj] Declare victory and retire?

["Bernard Vatant" <bernard.vatant@mondeca.com>]

This is the second part of the answer, concerning TC process and status

> * Bernard Vatant
> |
> | - PubSubj TC has been very very long (almost 2 years now) to deliver
> | something that seemed simple to begin with, and that in its initial
> | charter was due more than one year ago (2002-Q1).

LMG

> Well, that is your opinion. It never did seem simple to me, and I am
> quite pleased with what we have done so far, though I agree that we
> have not made as much progress as I would have wished us to. I think
> the problem is largely that we have had too little resources for work
> on this outside the F2F meetings.

Yes, we keep saying we don't make progress enough because our resources are
so sparse, but never tackle the fundamental issue: why are our resources so
sparse, why did not we gather more task force since two years? When this TC
was founded, the declared roadmap was to attract potential adopters in the
industry, publishers, librarians, and so on, and work with them around real
use cases. Two years after, it turns out that those potential collaborators
are not here, at a few notable exceptions, and that the work is made by
basically the same core community than two years ago, also involved in
SC34, and that the whole expected scenario did not happen.

> | - The very slowness of the process, and/or the lack of bandwidth of
> | participants, and/or the lack of clarity or consensus on what this
> | TC really wants to achieve, and/or the lack of investment in this
> | process of real users, all those factors have been slowly leading
> | the TC activity, over the past six months, to some kind of dormant
> | stage.
>
> Agreed, and I think member bandwidth is the primary problem.

No, it's a secondary problem, the primary one being lack of interest and
involvment of more people. The lack of task force entails the lack of
bandwidth. We've not reached the needed critical mass.

> | The pile of core issues, in fact the very purpose of this TC :
> | bringing practical solutions for PSI widespread adoption in the
> | industry, remains in standby.
>
> Well, that's hardly surprising. How could we expect this to be adopted
> before the technology is ready? Before deliverable 2 is in place
> there's nothing to show people what they are supposed to do.

Sorry I was not very clear here. I meant that real users' viewpoint has not
been an effective input in the process, and that we have not really tested
the specification against real world use-cases. A notable exception being
what Mary has done with UNSPSC subjects. But this input has been also in
standby for a while.

> | PubSubj TC is not pushing for anything right now, it's sort of
> | stunned by the complexity of the issues it has encountered, and does
> | not seem to gather the energy to tackle them and move forward. This
> | is obvious from the lack of reactivity and investment of everyone
> | lately in this TC.
>
> I very strongly disagree. I feel we have done quite well in tackling
> the complexities, and that the main thing remaining now is to work out
> the consequences of what we've agreed on so far and to formulate a
> document (deliverable 2) that describes the result. The only problem
> is finding the time to do it, really.

Good for you to be so optimistic. Please make up a list of what you
consider consensus points for deliverable 2. All I see is the pile of
questions we put again on the table in London.

> | I'm pretty well convinced that I will not build consensus in this TC
> | around that viewpoint, and I have no intention either to fight about
> | it, or to endorse as TC chair options that I would not be deeply
> | convinced to be right.
>
> That's fair enough.
>
> | The practical consequences of all of the above is that the wisest
> | thing to do now IMO woul be (quoting Sam Hunting after the painful
> | delivery of XTM 1.0) to "declare victory and retire".
>
> In my opinion that would be the same as declaring defeat and
> retiring. You may do that if you wish, but I am not interested in the
> slightest.

Pretty clear.

> | That means practically
> |
> | - Wrap up Deliverable 1 (why this is not done yet, I wonder)
>
> Good question. It would be very useful if you could post a summary of
> its current status so that we can finish this job.

Will do ASAP in a separate thread.

> | - Wrap up the TC, explaining publicly why (impossibility to meet its
> | charter, for such and such reasons).
>
> Personally, I am very strongly against that. If you want to resign as
> chair that's up to you, but if you do my wish would be to find a new
> chair and carry on.

Understood.

> | - Let publishers interpret Deliverable 1 at will, and see what
> | survives.
>
> Nothing would survive. Deliverable 1 is not enough to make this
> usable.

You're probably right here. It figures we've not achieved much yet.

> | If the majority decides otherwise and wants to move forward
> | altogether, I will retire and let someone else endorsing the
> | responsibility for it. This is not an easy position, but I have
> | engaged way too much time and energy in the past two years for this
> | TC, for quite a poor result indeed.
>
> I'm not pleased with our progress either, but I believe the solution
> is simply for us to do better in the coming months.

Wishful thinking. But I'm interested to know why and how we could do better
in the coming months than in past six months or so.


Bernard Vatant
Senior Consultant
Knowledge Engineering
Mondeca - www.mondeca.com
bernard.vatant@mondeca.com