ietf-mailsig
[Top] [All Lists]

RE: Ways to proceed

2004-10-15 18:46:47

 
You point out the inherent weakness of the IETF scheme. The 
ineffectiveness of the mailing list discussions means that the real 
decisions get pushed out to a cabal and the rest of the group feel 
alienated and in the case of MARID went on to wreck the proposal.

Chaos leaves the decision making to a cabal.

Mailing lists and even face-to-face meetings require structure to run
effectively.  Most large face-to-face organizational run by Robert's
Rules of Order.  Similar processes could and should be created in an
e-mail forum.

I've seen a few well run e-mail groups (IETF and else where) work
through the decision making processes in steps that gradually, but on a
known timescale, converge on decisions. So, any form of communication is
fine, process is what's critical.  A conference call or a mailing list
needs a moderator that enforces some level of focus.  



Paul

-----Original Message-----
From: owner-ietf-mailsig(_at_)mail(_dot_)imc(_dot_)org 
[mailto:owner-ietf-mailsig(_at_)mail(_dot_)imc(_dot_)org] On Behalf Of 
Atul(_dot_)Sharma(_at_)nokia(_dot_)com
Sent: Thursday, October 14, 2004 10:41 AM
To: pbaker(_at_)verisign(_dot_)com; gmgross(_at_)nac(_dot_)net
Cc: william(_at_)elan(_dot_)net; ietf-mailsig(_at_)imc(_dot_)org
Subject: RE: Ways to proceed



While it is justified that equal opportunity not be given to 
all, as only 4-5 max people are actually effective. But it is 
not justified that these 4-5 effective people make the 
decision. You justify exclusion and equal opportunity in same breath.

I do agree that functioning at IETF slows down, if not 
totally prevent, acceptance of new ideas, equal opportunity 
to new comers or those who are not part of cliques. But what 
you suggest does not sound any better to me. 

Having audio confernces with 80+ people is going to be a 
chaos, without being able to relate voice to a face/ID. In an 
email or chat format at least you can tell who is saying 
what, helping in organization of the ideas as a team. There 
will be a log to go back to, which is more effective and less 
expensive than recording audio conferences and distributing them.

Just my thoughts on the issue.

Atul


-----Original Message-----
From: ext Hallam-Baker, Phillip [mailto:pbaker(_at_)verisign(_dot_)com]
Sent: Thursday, October 14, 2004 12:59 PM
To: 'George Gross'; Hallam-Baker, Phillip
Cc: Sharma Atul (Nokia-ES/Boston); william(_at_)elan(_dot_)net; 
ietf-mailsig(_at_)imc(_dot_)org
Subject: RE: Ways to proceed



Hi Phillip,

On Thu, 14 Oct 2004, Hallam-Baker, Phillip wrote:

For the WS-Security WG we started off with 80 people on 
the calls.

not relevant. By their nature, teleconferance calls have
the following
non-starter characteristics:

1) they are inherently exclusionary, as there is a finite 
number of 
teleconference participants who must show up at the
scheduled time, in
contrast to the e-mail process that is 7x24 and scales to
any size of
group.

And the IETF meetings would be what then?

Getting the job done is inherently exclusionary. Only four or five 
people maximum can be effective editors on a spec, the rest 
can only 
really comment if the job is going to be finished.


2) teleconferences generally prevent and/or inhibit 
anonymonity, you 
must identify yourself to participate in the call. For some 
conference services, who ever runs the call can retrieve call 
originations from the teleconference billing data.

There is not and has never been a requirement for IETF 
process to be 
anonymous. Suggesting that this should be a requirement is 
just silly.


3) it is problematic whether there will be a publically 
visible and
*accurate* record of the proceedings stored at an archive web 
address. it entirely depends on the minutes taker's capacity to 
hear, capture to text, and identify who said what. then post the 
results for
review/comment.

That is why in every business meeting the first order of 
business is 
always to review the minutes of the last meeting.

The Victorians managed to make the process work.


4) it places at a disadvantage those participants who 
have English 
as a Second Language, or reside in disparate time zones from the
majority.

The proceedings of the IETF have always been in English, 
and all the 
specs are in English. If you are not fluent in English you 
are going 
to be unable to make more than a limited contribution.


5) it incurs a recurrent economic and time scheduling expense to 
organize/participate.

Rubbish.

In some cases, working groups may delegate subsets of 
their work to 
design teams that work offline from the main group and 
then report 
their findings in an Internet Draft. However, it is not 
clear to me 
that process does not have some of the same exclusionary 
characteristics found in the teleconference proceedings. at what 
point does the design team have too many cooks? who decided team 
membership eligibility?

You point out the inherent weakness of the IETF scheme. The 
ineffectiveness of the mailing list discussions means that the real 
decisions get pushed out to a cabal and the rest of the group feel 
alienated and in the case of MARID went on to wreck the proposal.


The IETF is run in exactly the way that most engineers 
would like to 
work, free from deadlines, accountability and with infinite 
scope to 
achieve technical perfection. The results are pretty predictable.









<Prev in Thread] Current Thread [Next in Thread>