Re: [eclipsecon-na-program-committee] Talks selection process

[Mikael Barbero <mikael@xxxxxxxxxxx>]

It sounds very reasonable to me. Thanks Jay!

Le 26 oct. 2015 à 16:07, Jay Jay Billings <jayjaybillings@xxxxxxxxx> a écrit :

I wouldn't make it a function of the number of submissions because there might be a lot of talks that don't receive any votes. I would make it a function of the number of talks that received votes. Then, I would calculate the standard deviation of the distribution of community votes and say that one, two, three, etc. times the standard deviation below the mean number of votes is the threshold for where community votes count. It will be much easier to pick the number of standard deviations to go below the mean because of the way the distribution falls off and it will prevent us from guessing on how many votes we will get at this point.

Jay

On Mon, Oct 26, 2015 at 10:44 AM, Mikael Barbero <mikael@xxxxxxxxxxx> wrote:

> Le 26 oct. 2015 à 15:10, Maximilian Koegel <mkoegel@xxxxxxxxxxxxxxxxx> a écrit :
>
> I understand both your positions but I am in favor of treating this as a test balloon, therefore I think we should limit the publicly announced influence of the community voting, since we will not be able to take it back easily.

Yes, that's my point ;)

> Another question about this came into my mind: If we opted for x% influence by community voting, do we define a lower bound in the number of votes overall in community voting before we even consider it in the voting result?

I think we should define this lower bound, however setting it is very difficult! Maybe we can define this minimum number of votes as a function of the number of submissions (and let's define 1 vote = 1 grade for 1 talk by 1 user):

lower bound = X * number of submissions.

Any proposal for X? 2, 5, 10, 100?

(these maths should be simple enough so I won't screw it ;))


_______________________________________________
eclipsecon-na-program-committee mailing list
eclipsecon-na-program-committee@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipsecon-na-program-committee

--

Jay Jay Billings

Oak Ridge National Laboratory

Twitter Handle: @jayjaybillings

_______________________________________________
eclipsecon-na-program-committee mailing list
eclipsecon-na-program-committee@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipsecon-na-program-committee

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail