Dave Crocker / IMC wrote:
>
> (a re-submit; earlier submission got bounced. /d)
>
> At 08:51 AM 12/10/97 -0800, Dan Busarow wrote:
> >The point that bothers me most is the one requiring members to choose
> >which of the above categories they belong to and voting only that.
> >This effectively break PAB into separate bodies. I think it's a
> >bad idea.
>
[...]
>
> The simplest (and, I think, best) way to decide who a candidate represents
> is to let each candidate declare it themselves and then let people voting
> decide whether they agree. They do that by voting.
>
[...]
>
> The proposal posted by David Maher requires partitioning the PAB membership
> into the different constituencies and restricts the voting according to the
> chosen category.
>
> I think there are two serious problems with this aspect of the proposal.
>
> The first is that many PAB members legitimately belong to more than one
> category.
>
> The second is that the process of partitioning the membership strikes me as
> an extra step which will prove difficult, causing delay, and also seems
> likely to be pretty contentious.
>
> Hence I think we should strongly consider an alternative approach for doing
> the voting.
>
> The proposal posted by Robert Shearing (see the gtld-mou RFC archives)
> suggested giving each PAB member one vote for each slot. Hence, everyone
> would vote on every representative. This results in exactly the problem of
> "incompetent" votes, with people voting for constituencies about which they
> are not familiar.
>
> A refinement which was later suggested (I think I heard it privately so
> it probably is not on a mailing list archive) was to reduce the number
> of votes each member may cast, perhaps giving every PAB member a vote
> for
> one-half of the number of candidates.
Well, I think I was the one suggesting that, and, yes, you got it in a private
e-mail ;-)
>
> This means that those voting must decide which categories (constituencies)
> they consider to be most important to their organization. I believe this
> refinement is an extremely clever way to obtain a high degree of
> "competent" voting without requiring an procedural complexity. I believe
> it also will tend to cause those voting to think a bit more carefully about
> their choices.
My reasoning was the following:
* Concern: avoiding that any more or less coherent/orgasnised "group" within
PAB (be that ISPs, CORE Registrars or Catalan lawyers) "kidnaps" the whole PAB
representation on POC by voting their preferred reps in an organised way.
* Solution. Step 1: This will be very likely if PAB appointees to POC are not
devided into different "interests" (the "majority" will elect all reps). This
is the logic behind the creation of separate slots intended to represent
different interests/groups/stakeholders. This will be also the case if any
company would vote in all "interest-defined" subset of seats to be elected.
* Concerns with Step 1: How to deal with separate slots? If we allow every
PABer to vote in avery category, we are back to case 1. If we oblige PABers to
choose a given (and only one) category to vote on, we will face enormous
difficulties and will end with artificial divisions (my mechanism was basd on
the so-called Searing proposal, wich claims for a fair number of different
slots: ISPs, telcos; software industry, IP law... In which category will you
put say IBM or MCI?).
Solution to problem 2: Allowing each PABer to vote for, say, (up to) half the
seats to be elected. If, for instance, PAB is to elect 10 seats (two for ISPs,
two for IP law-related issues...) IBM, NetNames, FCR or Deutsche Telekom could
only cast, say, five votes. They would "tend " to spend the votes with the
nominees that are closer to their real interests. Grouping is then
self-selecting, not imposed from outside. Allowing each company to vote hjust
for up to half the actual seats we could efectively prevent any given group to
dominate POC. And we would also prevent splitting PAB into separate Chapters.
My figures were temptative. The number should depend of the actual
categories we define and the seats allocated to them. But I believe that each
PABer should at least be able to vote for the whole largest category (3 seats
uner the Rough Consensus/Maher proposal), and still be able to vote for at
least another seat (be that one "at large" seat or form another specified category).
Now some points specifically directed to the Rough Consensus proposal:
* I think that we should add a new category: Trademark law/interests. Like it
or not, TM concerns are hugely involved in any decision about domain names.
The reson to add such a category is to avoid these views represented
exclusively through INTA (one of the naming organisations under the curretn
proposal). INTA is widely respected, but by no way the only voice in this
area. Adn we don't like monopolies, don't we? ;-) (remember that the WIPO role
will be, as a non voting Observer, basically devoted to corridnation with the
Mediation and Arbitration Center charged of administering the ACPs)
* As *any* division we could imagine (be that the more detailed Sheraing-like
or the more vague-Maher like) will always be at least parially unsatisfactory,
I favor having some (two or a maximum of three) "at large" representatives,
not bound to any given interest group loyalty.
* I think that the three geographical slots are not the ideal solution. This
process has shown a high degree of geographical diversity in appointing
members to IAHC, iPOC, POC, PAB ExCom and CORE ExCom. The USA and Europe are
fairly well (over)represented). Japan and Australia have their fair slot, as
well. For this reason I favor excluding at least the USA and Europe from those
seats. and reducing these seats from three to two. I favor having one "at
large" seat without any geographical reference and reserving the other seat
for people coming from non-OECD countries.
* According to what I stated aofve. if PAB elects 9 seats, every PABer should
have 4 votes. If we are to elect 10, then each PABer would have 5 votes.
Amadeu, following the thread
This archive was generated by hypermail 2b29 : Sun Jan 30 2000 - 03:22:15 PST