Re: PEP TALK -- SPEAK UP!

[Michael Eager <eager at eagercon dot com>]

Ron 603-884-2088 wrote:
> 
> Last month following cancellation of our 9 December meeting, I asked Michael
> why he had canceled the meeting when it seemed that we had at least two
> topics that could usefully be discussed:
> 
>   - My 001130.1 alternative to 001101.1 regarding the multiple section
>     problem for location lists and discontiguous scopes (recall that 001130.1
>     resulted from the discussion of 001101.1 at the 28 November meeting).
> 
>   - The V3 vs "conditional V2" issue.
> 
> His reply concerning the first was basically that he'd seen no discussion of
> it on the e-mail list and so had little reason to believe that it was ready
> for formal meeting consideration.

Ron --

Here is  the remainder of my email.  I think it is more relevant than the
very small portion you quoted:

    I am very concerned that we attempt to repair flawed proposals at 
    the committee meeting.  This is very inefficient and poor use of the 
    committee's time and energy.   More, I don't think that we do a good
    job.  Support for this view is that we have had multiple proposals about
    several topics and proposals which amend prior proposals.  This suggests
    that we are considering proposals which are not fully considered but
    are more like trial balloons. 

    I believe that we have shown that it is very difficult to address issues 
    with complex interactions during the meetings.  My opinion is that much
    of the discussion should proceed on the email list with a (hopefully) wider 
    audience, rather than with the few people at a committee meeting, with
    time limits.  

    I'd much rather see an implementation of a complex proposal such as yours
    which will provide support for claims that this is the "right" proposal.
    Consider what Dave Anderson did with his proposal on reducing the frame
    description in which he was able to present experiential results.

    Unless I see some consensus of opinion that your third proposal, despite
    lack of empirical evidence, is the best approach, I will remain concerned
    about whether is ready for discussion.

I'll note that you have rejected my suggestions that you provide a demonstration
implementation of your proposal.  

In the past we have included features in Dwarf that the proposer claimed were
absolutely necessary, perhaps with the same fervor that you espouse.  We have
seen that these features remain unused and only serve to complicate the standard.  
I'd rather not repeat this.

In every standards process there are proposals which are "left on the table"
as the process reaches an end.  The assumption is, and I believe this to be
well founded, that the important matters have already been addressed at the
beginning of the standards process and that issues raised late in the process
are secondary or tertiary.

A standards committee does its best work when validating current practice.  It 
does its poorest work when it tries to create mechanisms on the fly, with little
investigation, based on little more than conjecture, assertion, and hyperbole.  
We have done much less of the former than I prefer.  Accordingly, I do not 
feel that blue sky brainstorming, as you have suggested, is appropriate for 
the Dwarf standards committee.

I would (again) recommend that you and Compaq implement your proposal as a 
user extension and provide us and/or a future Dwarf standards committee with
experience in implementing and using the method you propose.

> If no one speaks up, then there will little reason to hold the next planned
> meeting on 9 January (next week!) because nothing will have changed. Keep
> that up a time or three more, and DWARF V2.1 Draft 5 will be the last ever
> produced. Perhaps Draft 5 could enjoy the same no-formal-status but be a none-
> the-less useful document like the DWARF V2 spec? Maybe that is good enough?
> I think we can do better...

We are moving forward to final public release of a proposed Dwarf standard
and issues are being satisfactorily addressed.  We are moving toward 
participation in IEEE-ISTO.  We are moving forward on the schedule that we 
discussed some months ago, prior to your proposal and its multiple revisions,
with only minor deviations to this schedule.  Your contribution in maintaining
the standard document has been a valuable contribution to this effort.

As time goes by, errors in any standard become apparent.  I am quite aware
of a number of errors in the Dwarf 2 standard.  My hope has been to avoid
making additional errors in this revision, a goal which I am afraid has not
been achieved.  One way to avoid errors is to not make significant changes
to the standard at the end of the process, especially without adequate review
and consensus.  In this area, I would much rather err on the side of caution.

Again, thank you for your contribution to the Dwarf standards effort.

-- 
Michael Eager	 Eager Consulting     eager@eagercon.com
1960 Park Blvd., Palo Alto, CA 94306  650-325-8077