RE: write-ins

["Ken Clark" <ken@dieboldes.com>]

> From: owner-rcr@dieboldes.com [mailto:owner-rcr@dieboldes.com]On Behalf Of
> Scruggs, Phil
>
> We are asking for a way to accommodate groupings of not only
> mis-spelled names but also variations of the same name. For example:
> Bill Clinton
> William Clinton
> William J Clinton

Yes, the motivation is clear here.

> Can we get a level-of-effort to incorporate this feature into GEMS?

That depends on the "how" part of this RCR, which is still pending.

A write-in candidate can only win if the total of *all* write in candidates
is higher than the number of votes for the winning candidate.  I have never
seen this happen, though I am sure counterexamples exist.  In the unlikely
event that it ever did happen, I would think that the election administrator
would want to go over the write-in report with a fine tooth comb.  If the
idea here is that they go through all write-in names and group them
manually, then the obvious way to do that is with a pen and check marks on
the write in report.  The "level of effort" to approximate this in GEMS
would be higher than with the pen.  Not as user friendly either, methinks.
I can't think of any "automatic" algorithm that I would always trust.  Does
"Slick Willy" count in your example?

Bottom line, lets get the "how" part defined.  We can do just about
anything.  If this boils down to a deal-breaker for Harris, then the level
of effort really isn't an issue.  I would be very surprised (and
disappointed) to find this a deal-breaker though.  If it is, then we can
always write it into the contract as a line-item.

Ken