<div dir="auto">It’s necessary to count N(N-1) pairwise vote-totals. …2 for each pair.</div><div dir="auto"><br></div><div dir="auto">What if there are lot of candidates?</div><div dir="auto"><br></div><div dir="auto">The trick-software could be in the part that initially increments the N(N-1) vote-totals, even if that’s immediately done at the voting-machine.</div><div dir="auto"><br></div><div dir="auto">The only thing that the voter checked for accuracy was his ranking. There’s no guarantee that the pairwise vote-totals were incremented honesty.</div><div dir="auto"><br></div><div dir="auto">So a handcount-audit would require doing the whole exhaustive pairwise-count, from the raw rankings.</div><div dir="auto"><br></div><div dir="auto">25 candidates? 600 pairwise vote-totals to count.</div><div dir="auto"><br></div><div dir="auto">Of course there’d be an effort for security, but, undeniably, it would be easier & more able to be counted-on with approvals.</div><div dir="auto"><br></div><div dir="auto">To maximize count-security, of course it’s Approval.</div><div dir="auto"><br></div><div dir="auto">It’s getting late ⏰, so I’d best continue this topic in the morning.</div><div dir="auto"><br></div><div dir="auto">In. Vermont it’s 6:00 a.m. I hope you’re getting up early, & not staying late !</div><div dir="auto"><br></div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sat, May 18, 2024 at 01:06 robert bristow-johnson <<a href="mailto:rbj@audioimagination.com">rbj@audioimagination.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;padding-left:1ex;border-left-color:rgb(204,204,204)"><br>
<br>
> On 05/16/2024 2:04 AM EDT Michael Ossipoff <<a href="mailto:email9648742@gmail.com" target="_blank">email9648742@gmail.com</a>> wrote:<br>
> <br>
> …<br>
> Count-fraud is a problem. Condorcet’s humungously computation-intensive count ridiculously facilitates count-fraud.<br>
<br>
I consider Condorcet to be precinct summable with no more than N^2 number of tallies. I think 16 tallies for 4 candidates is feasible. Even 25 tallies for 5 candidates.<br>
<br>
The N(N-1)/2 pairwise comparison are done as the ballot is scanned by the tabulator machine. That part is opaque, but the rest of it is completely transparent and the complexity is small.<br>
<br>
Unlike IRV has been, Condorcet RCV can have election results on election night.<br>
<br>
> …<br>
> You want to do a handcount-audit of a Condorcet count?<br>
<br>
Hand-counting Condorcet is processing the pile of ballots N(N-1)/2 times. Hand-counting IRV is processing the pile of ballots N-1 times.<br>
<br>
> …<br>
> Additionally, the count-program itself is easier to hide or add fraud-code in.<br>
<br>
But, this is only at the tabulator level. When the ballot is inserted into the tabulator.<br>
<br>
No different than we have now with FPTP. That part of the data trail is opaque to protect the Secret Ballot. But the rest of it can be transparent with Condorcet RCV, as it already is with FPTP.<br>
<br>
> …<br>
> As a general principle, then yes it’s much better to have the voters do it for themselves rather than having a complicated fraud-prone automatic-machine do everything for them.<br>
<br>
Pick a precinct and hand count it. Not much worse than IRV.<br>
<br>
Results are summable. Not so with IRV.<br>
<br>
This is pretty transparent.<br>
<br>
--<br>
<br>
r b-j . _ . _ . _ . _ <a href="mailto:rbj@audioimagination.com" target="_blank">rbj@audioimagination.com</a><br>
<br>
"Imagination is more important than knowledge."<br>
<br>
.<br>
.<br>
.<br>
----<br>
Election-Methods mailing list - see <a href="https://electorama.com/em" rel="noreferrer" target="_blank">https://electorama.com/em</a> for list info<br>
</blockquote></div></div>