Internet voting (Re: POC membership [was : Re: The Price of Admission is...])

["andi payn" <payn@null.net>]

Me:
>> I'd have a simple form listing all nominees (from the first step above)
and
>> a write-in slot. Each entity (person, commercial entity, etc., depending
on
>> class) gets one vote. There is some question as to how to very fair
>> elections in such a vote.

Jeff Williams:
>   Use what most election commisions use.

Me:
>> For individual voting, the issue is complicated by the fact that a single
>> person can have more than one email address (e.g., multiple screen names
on
>> AOL, Hotmail mailboxes, iName forwarding addresses, addresses from
multiple
>> departments in a University, even just multiple mail hosts on the same
>> network, etc.). It's just as possible to have multiple "verifiable"
>> addresses whether you're using signed PGP keys, First Virtual, etc. For
>> example, I have three different machines that I run Netscape from, and
each
>> one "uniquely" verifies me as a different person. I also have multiple
email
>> addresses and PGP keys for all of them. For that matter, if we use, say,
>> Netscape's unique verification, what happens to people who can't run
>> Netscape 3 or 4 (e.g., users of MkLinux, AmigaOS, DOS, etc.) or just
don't
>> have a copy?
>
>  Dos is not a problem as far as access.  The other two you mention
>are.  Not much you can do about this.  They are really not members of
>the Internet community at any rate.

I was referring to the issue of using something like Netscape's system of
uniquely verifying an individual, not just access to the Internet. I just
chose Netscape as an example; with that example, DOS is a problem. As far as
net access, none of those OSs are a problem, and I think it's had to claim
that Linux users, even non-386 Linux users, aren't really members of the
Internet community.

>  As to the multipul E-Mail addresses.  I can simplify this very
>easly.  "One man/woman one vote."  Those found using morethan one
>E-Mail address to vote multipul times will find their vote not
>being counted at all.  This can be determined very easely.

That's what I'm not sure about. How can it be determined easily? I'm pretty
sure I could send you email from five different addresses without you having
any idea that they were from the same person (except maybe by looking at the
style and content of the letters). And with AOL it's even more complicated.
My mother, grandmother, and sister share an AOL account. They each have
their own screen names (and hence email addresses), and my sister has two.
How can you tell that her two names are the same but my mother's is not?

Of course in most cases a person wouldn't vote multiply, but the type of
person who would would probably do it with as many names as possible. And I
could go to Hotmail right now and get 5000 names instantly, for free, and I
could go to iName and double those email addresses with forwarding
addresses. I could also give myself as many addresses as I wanted at one
ISP. For that matter, it's not too hard to forge thousands of fake email
addresses (just ask Spamford), although this last could maybe be detected
automatically.

I have a feeling that there is an acceptable solution that doesn't require
asking national governments to provide lists of their citizens and making
voters prove who they are, but I don't know what it is.

>> For organizational voting, there's a question of how to qualify
>> organizations. If I have a personal dba, does it qualify as a company?
What
>> if I have three? What if a company has multiple divisions that are
>> incorporated separately?
>
>  We are in the latter catagory you mention here.  Not only that we
>may even be in seperate groups.  (ISP and software development company).
>But be that as it may.  "One man/women one vote should suffice".

What does "one man/woman" mean in terms of companies?