I disagree with this suggestion. Personally I vote for only those I find interesting and useful to me, but I don't put an response for every talk listed. I only respond on those I'm interested. Everyone else gets 0 points. I would expect that others do this, too. Katherine's suggestion also puts an burden on those who are legitimately participating while doing nothing to prevent those who are misbehaving.

I like Edwards's suggestions, which are easy to implement and don't really impact the process that much.

Personally, I believe that the proper response to this is to:

1. Publicly shame those who are participating in this. :) 
2. Delete their votes, or at least those you can identify.
3. Disqualify the person who is receiving illegitimate votes. See #1.
4. Eliminate voting altogether and have a committee of 10-15 people from the community select from the proposed talks. Isn't this what other conferences do?

In the end, the conference organizers can invite whoever they want to speak. The voting ends up being a courtesy to the rest of us.

--Joel

Joel Richard
Lead Web Developer, Web Services Department
Smithsonian Institution Libraries | http://www.sil.si.edu/
(202) 633-1706 | [log in to unmask]








On Dec 1, 2011, at 8:06 AM, Lynch,Katherine wrote:

> I was actually going to suggest just this, Kåre!  Another way to handle
> it, or perhaps an additional way, would be give a user's votes a certain
> amount of weight proportionate to the number of sessions they voted on.
> So if they evaluated all of them and voted, 100% of their vote gets
> counted.  If they evaluated half, 50%, and so on?  Not sure if this is
> worth the effort, but I know it's worked for various camps that I've been
> to which fall prey to the same problem.
> 
> Sincerely,
> Katherine
> 
> On 12/1/11 6:55 AM, "Kåre Fiedler Christiansen" <[log in to unmask]>
> wrote:
> 
>>> From: Code for Libraries [mailto:[log in to unmask]] On
>>> Behalf Of Michael B. Klein
>> 
>> <snip>
>> 
>>> In any case, I'm interested to see how effective this current "call
>>> for
>>> support" is.
>> 
>> Me too!
>> 
>> Could someone with access to the voting data perhaps anonymously pull out
>> how many voters have given points to only a single talk or two?
>> 
>> If the problem is indeed real, perhaps simply stating on the page that
>> you are expected to evaluate _all_ proposals, and not just vote up a
>> single talk, would help the issue? It might turn away some of the "wrong
>> voters". Requiring to give out at least, say, 10 points, could be perhaps
>> be a way to enforce some participation?
>> 
>> Best,
>> Kåre