single-winner ballot with restricted rankings?

[Steve Eppley]

John Taplin wrote:
> With the technology known to me in the early 80s I advocated a
> square array with candidates as rows and preference for columns
> with the voter filling in the intersecting squares as a
> "mark-sense" card.  I pointed out this would allow indifference at
> any level and that it need not be compulsory to put a mark against
> each candidate.  The voter need only be sure she did not put two
> marks in any row.

That's what I have in mind, except for the squareness.  With a lot of
candidates there won't be room for a square array because the ballot
probably isn't wide enough.  My conjecture is that restricting the
number of distinct rank positions won't cause bad results. 

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)