My 2 cents:
Have everyone mark the logos they like, that is don't restrict to a top
5 or top 4 or whatever.
Extract from this the top 3 most liked logos. If there is a tie extends
to top 4, top 5 or whatever
Do a second run, where voters can only choose 1 of the 3 remaining logos.
btw, who is going to vote, commiters?
Greg Reddin a écrit :
On Aug 17, 2006, at 2:13 PM, James Mitchell wrote:
For example (3 votes per person):
Craig Sean James Greg Wendy
Image 1005 x x x <- tie
... except I think a better listing would be:
Image 1005 3 binding, 15 non-binding.
Image 1006 2 binding, 10 non-binding.
Image 1008 3 binding, 10 non-binding.
... in which case Image 1005 and 1008 are still tied, but we *might*
choose one based on non-binding votes or have a runoff.
You can see how these 2 method differ. Both examples show the same
images being voted the same, the difference is in the ranking. I
prefer the 2nd method because it more accurately reflects how people
feel, rather than just being the common one that most people _didn't_
exclude. It's our way of saying "I really REALLY like this one or
that one", yet still giving equal voting all around.
I see your point, but I actually like the first method better. This
is simply because it will be easier for me to choose my top 5 and not
necessarily pick an order than it will be for me to decide on a single
one I like best. There are too many that I like a lot to decide which
one is "best". But I could probably narrow it down to 5 that I like
enough to vote for.
Either way we go, I'll be surprised if there is actually a tie. What
are the chances that multiple concepts will actually get the same
number of votes? But we may get two or more that are very close and
choose to do a runoff based on that. Or we might see a definite
preference for one from the non-binding community and choose to have a
runoff.
Once the voting is final. This will divide the pool of entries into
2 groups.
A) those with a ranking
B) those not yet ranked
At this point we *should* know who won or tied (requiring a run-off),
but until that author provides both the source (Photoshop .psd, or
whatever) and files a CLA, they aren't really the winner yet. We
should decide on an appropriate time frame to allow people to file
this paper work. 1 week? 2 weeks?
I'm cool with the above. It could take much longer than 1 or 2 weeks
for a CLA to go through if we use the "normal" channels. I think mine
took about 6 weeks, and recently some have taken even longer. I'm not
sure the best way in which to handle this.
In the event that the chosen author cannot or does not file the CLA
and provide the image sources, that author will be moved to an
unranked status and the next runner-up will be asked to provide the
image source and file a CLA. Hopefully, there won't be any issues
getting the paperwork filed and such.
How does this sound? Does that accurately represent what you guys
are thinking?
Close enough :-)
Greg
