[EM] Intro to list (etc)

[Rob Brown]

At 02:56 PM 10/26/2003, Adam Haas Tarr wrote:
>Looks like a promising idea, Rob.

Thanks!

> >Here is a UI I am working on for doing for ranking
> >candidates:    http://weblogz.com/voting/2000pres.html
> >This demo is of course based on the 2000 presidential election, and allows
> >you to rank candidates with a (hopefully) friendly UI.  I avoided having
> >people manually assign numbers to candidates  (after all, they are sitting
> >in front of a perfectly good computer which can do that sort of thing
> >well!), and I tried using a little animation, which seems to help in making
> >it easy for voters to follow what they are doing.
>
>It looks good.  I'd suggest replacing the words "unpick" "move up" and "move
>down" with an X, an up arrow, and a down arrow.  I'd also suggest 
>implementing
>drag-and-drop of the actual names (drop McCain bewteen Bush and Gore,
>automatically sliding down the lower of the two).  Just my suggestions, of
>course - as I said it looks nice

Hah, yeah, just had this discussion with ezboard people friday. :)  The 
ezboard graphic designer guy will make arrows and icons for pick/unpick.  I 
like the idea of X for unpick (and if unpick is an X, i suppose pick could 
be a check mark, right?)

As for drag and drop.....I agree completely.  I messed with it a bit, 
realized it was pretty hard to get right, and put it on the back 
burner.  It will happen though, I'm sure.  (whether or not people will be 
able to figure out you can drag things around in a web page ....don't know)

>I don't mean to open a huge can of worms here, but are you tabulating 
>results by
>winning votes or by margins?  If less strategy is your goal, then you 
>should use
>winning votes.  You can search the archives for debates on this.

Yes, I absolutely want to go by winning pairwise elections, and less 
strategy is most definitely my goal.

>Here is what I would use.  Compute the results using beatpath, aka 
>Schulze, as
>oppose to Ranked Pairs (aka Tideman).  Then display the winning candidate, 
>along
>with the number of votes on his weakest beatpath.  Then display the remaining
>candidates in order of strongest beatpath against the winner.
>
>This will guarantee that the winner has the most votes, and it will give the
>remaining candidates in a reasonable decreasing order.  No candidate can get
>more votes than the number of voters (easier to swallow than Borda), and only
>one candidate can have more than 50% of the vote.  Finally, if the voters all
>vote for only one candidate this reduces to a plurality count, which is a 
>nice
>property to have.

Wow....ok, sounds about perfect....

> From a technical perspective, Beatpath and Ranked Pairs are so similar 
> that it's
>probably not worth worrying about.  If you want to use Ranked Pairs to 
>determine
>the winner, but display the beatpath counts, that probably wouldn't present a
>problem.  I'm merely proposing using Beatpath because it provides a count 
>that
>is useful to you for presentation purposes.

Thank you!  This seems exactly what I'm looking for.  I have no preference 
for Ranked Pairs over Beatpath.  I don't know how beatpath works yet, but 
it sounds like it will do exactly what I need.  I'll give those articles a 
good read....

-rob