On Fri, Nov 17, 2000 at 08:26:29PM -0500, John D. Giorgis wrote:
> There are no provisions for well-defined guidelines in Florida
> law, and thus, each county has been left to their own devices in
> establishing such standards.
Which, of course, does not in any way prevent the manual counters in
each county from following some set of well-defined guidelines, defined
by the county canvassing board.
> This was widely rumoured in the days after the election, but the most
> recent available data suggests that this is not correct. See the
> following link: http://orlandosentinel.com/elections/lost.htm
That link has nothing useful to say on this particular issue. The data
that are relevant are the difference between the number of uncounted
votes when counted by machine and when counted by person. That site only
gives the number of "blank or spoiled" ballots as counted by machine.
> The data from this link is highly suggestive that more votes are lost
> whenever ballots are counted at a central location in the County,
> rather than at the precincts.
Yes, I agree.
> This further suggests that greater handling of ballots *may* actually
> decrease accuracy.
For a MACHINE count, maybe. It says nothing about a manual count.
> At any rate, it is clear that when ballots are taken to central
> counting locations, Optiscan technology does not miss significantly
> fewer ballots than Punchcard technology.
No, you have read too much into the data. As I said above, the data does
not compare the relevant quantities.
--
"Erik Reuter" <[EMAIL PROTECTED]> http://www.erikreuter.com/
