[EM] Meek's Method/approval hybrid

[James Gilmour]

> raphfrk at netscape.net> Sent: 27 May 2007 15:55
> 
> I was thinking about Meek's method and the possibility of 
> combing it with approval.
> 
> There doesn't seem to be a definition of Meek's method on the 
> wikipedia.  Does it already allow equal rankings ?

You'll find an algorithmic description of Meek STV as item 4 on the "Voting matters" Resources
webpage at:
  http://www.mcdougall.org.uk/VM/RESOURCE.HTM

You'll find reprints of Meek's original papers as the first items in Vm March 1994 at:
  http://www.mcdougall.org.uk/VM/MAIN.HTM

I didn't immediately see a reference for it, but I am fairly sure someone has programmed a version
of the Meek algorithm that does allow for equal rankings and used it in real (non-public) elections.

I have no idea whether Meek STV and Approval are philosophically or practically compatible.  Meek
STV is a highly iterative procedure, with surplus votes transferred to already elected candidates,
which of course, creates further surpluses that have to be transferred, and so on, and so on,
...........  Meek STV also starts the whole count again every time non-transferable ballot papers
are encountered.  The quota is reduced by the amount of the non-transferable votes and the count
begins afresh with the reduced quota.  This is not iterative, but it can go round many cycles.

James Gilmour