Rob Richie contradicting himself on comments page

[Clay Shentrup]

http://lookingatdemocracy.org/submissions/14935-the-united-states-needs-approval-voting

>voting 

>for your second favorite can be a problem. Why? 

>Because it counts equally with your favorite 

>and, as a result, can lead to the defeat of your 

>favorite candidate. That means anyone who think 

>their favorite as a chance to win are going to 

>be wary of voting for their second choice -- and 

>we're heading back toward plurality voting dynamics.

So, people are going to bullet vote. Which is only a problem if they do it even when it's not in their best interest. So Rob must be worried that people will bullet vote, even when it's in their best interest not to.

 

>Another example was at U-Colorado this year. In 

>the 2-candidate race for student government 

>president, the result seems to have been 51.9% 

>to 51.6%. The margin was 22 votes, with 246 

>students apparently voting for both candidates.

I.e. voters chose not to bullet vote, even when it was in their best interest to bullet vote. And the number of voters who did this was far greater than the margin of victory, meaning it easily could have made a difference to the outcome.

Rob has done great job refuting himself here.

[Michael Ossipoff]

On Wednesday, May 15, 2013 1:16:48 AM UTC-4, Clay Shentrup wrote:

http://lookingatdemocracy.org/submissions/14935-the-united-states-needs-approval-voting

>voting 

>for your second favorite can be a problem. Why? 

>Because it counts equally with your favorite 

>and, as a result, can lead to the defeat of your 

>favorite candidate. That means anyone who think 

>their favorite as a chance to win are going to 

>be wary of voting for their second choice -- and 

>we're heading back toward plurality voting dynamics.

 

 

Wouldn't it be nice if voters would vote only for their favorite!  :-)  If Richie thinks that's how people vote, then one must wonder what planet Richie lives on.

 

And, given that millions of voters now (based on what they themselves say) aren't voting for their favorite, then why does Richie believe that they'd start doing so if given the option to vote for both Compromise and Favorite (when now they vote only for Compromise)?

 

At first, we could have generously, charitablly, attributed Richie's confused statements to ignorance. But these things have been explained to him so many times, that it must be something more than that. Dishonesty suggests itself as an explanation.

 

Richie and FairVote are an example of what happens when promoters with their head up their a _ _  somehow acquire the means to mass-promote their ignorance and confusion (and dishonesty?).

 

As Clay mentioned, sometimes it's in your best strategy to approve only one candidate, when only one candidate is acceptable. Then, how does Richie think that's a disaadvantage of Approval? But, when there are several acceptable candidates, and some unacceptable ones, then it will be in your best interest to approve all of the acceptable candidates.

 

Sure, there can be a chicken dilemma among the preferrers of those acceptable candidates, but we've spoken of how that can be dealt with. I've posted a list of reasons why the chicken dilemma won't be a problem--only a nuisance that can be dealt with, to the extent that it even happens.

 

In the Green scenario, IRV's MMC compliance and freedom from chicken dilemma would make it a good method. But it's necessary to distingush between different conditions, and recommend the different conditions that are good under different conditions. IRV is no good for current conditions.

 

Even in the Green scenario, Benham, Woodall and Schwartz Woodall would probably be better, in the sense that they wouldn't be vulnerable to replacement by a dis-satisfied majority. Those IRV Condorcet hybrids, by always choosing the CW,  and always choosng from the Smith set, will never dis-satisfy any majority, however constituted.

 

 

 

 

 

 

 

[Abd ul-Rahman Lomax]

I just spent days writing a deconstruction of the FairVote position
paper on Approval Voting. Yeah, I'll understand if Clay hasn't read
it. I send my drafts, which include a great deal of dicta (consider
them footnotes that haven't yet been distinguished, or they could
become references to other documents), to the CES list. Maybe someone
will comment, and then, theoretically, a tight, condensed examination
of the arguments can be constructed.

Yes. I do assume Richie is the main author -- or only author -- of
that position paper, the arguments are very much his arguments, the
approach is very much his approach, as it has been for years.

Richie treats "heading back toward plurality voting dynamics" as if
that were "unworkable." Actually, it's simply a fallback mode that
guarantees that Approval won't make things worse. And the FairVote
document never looks at where Approval could make a big difference.

Richie translates "wary" -- which is a normal response to the
situation for voters who *do* have a strong perference for their
favorite -- into "won't add additional approvals," neglecting that
*many voters* will add such, and for many different reasons. How many
will vary with conditions.

Someone who thinks their favorite is a frontrunner may well not add
approvals, that's normal strategic voting. But there are forces in
the other direction, that depend on the particular contest and on
voter positions. Adding an additional vote is *also* strategic
voting, where the voter prefers electing one of a set to another
frontrunner possibility.

Approval voting *requires* making strategic decisions, it is not
merely "vulnerable to strategic voting." Strategic voting makes Approval work!

The truly bizarre use of langugage that arose among voting systems
experts, once they wanted to find ways to criticize Approval voting,
which had been announced by Brams as "strategy free," was to redefine
strategic voting to include, as the most prominent "strategic voting"
example, a voter simply voting for their favorite.

Yes, that's "strategic" in the ordinary sense of the word, *under
some conditions.* But it is not "strategic voting" in the former
sense, which *always required* preference *reversal,* not merely
suppression of preference, by choice of the voter. So, bottom line,
FairVote is calling simple, common, fully sincere votes -- bullet
votes -- "strategic." That requires an understanding that the voter
"really" also approved of another candidate, but didn't vote this
"sincere approval." It's a direct contradiction, and it plays on the
name of Approval Voting. That implies that voters vote for all
candidates whom they approve in some sense.

No. Voters choose where to set their approval cutoff, and Approval is
"strategy free" because they may *always* vote for their favorite,
there is no "strategic" decision to be made in that. This is, in
fact, the principal benefit of Approval, which FairVote doesn't even
mention in their position paper, no surprise. Where the voters set
their approval cutoff will depend entirely on their expectations, it
is not, as implied, a set condition. It's simply a choice.

And then we also need to know that most voters, in most environments,
will bullet vote, naturally. FairVote is correct in their examination
of the history, but they utterly fail to recognize the cause of the
phenomenon and its significance. Yet Dodgson knew it in the 1880s.

Were the CES video intended as a balanced view of Approval Voting, it
could be considered misleading. It was not intended as that, and we
invited people to "join the conversation," which was brilliant. We
will not, in our full presentations, neglect the details, including
the situations where Approval voting may fail to improve a specific
outcome. But there is *no* evidence that Approval will *ever* make
things significantly worse, and that is why voting systems experts
came to consensus that an excellent first reform is Counting All the Votes.

That was my slogan for Approval, and I was actually gratified to see
that FairVote picked it up. It's obvious to me that I've been read,
at least occasionally.

I will write a separate post critiquing the video from a
more-FairVote perspective, and then showing how contrary conditions
were asserted in the video that would be likely to cause, in a
real-world election similar to the video perspective, enough multiple
approvals to still fix the vote-splitting problem, and this could be
made even more reliable by the candidates. I did cover that argument
in my deconstruction of the FairVote document.

[Warren D Smith]

>>Another example was at U-Colorado this year. In
>>the 2-candidate race for student government
>>president, the result seems to have been 51.9%
>>to 51.6%. The margin was 22 votes, with 246
>>students apparently voting for both candidates.
>

> I.e. voters chose *not* to bullet vote, even when it was in their best
> interest *to* bullet vote. And the number of voters who did this was far

> greater than the margin of victory, meaning it easily could have made a
> difference to the outcome.
>
> Rob has done great job refuting himself here.

--Richie's arguments change with the phase of the moon.

But anyway, was that data from U-Coloradio RR cited, actually valid?
I saw a U-colorado web page

http://cusg.colorado.edu/content/spring-2013-election-results-cusg

about this election.
I do not see how Richie deduced 246, 51.9, and 51.6. He must have
access to data
not available on this web page? (In fact, 51.9 and 51.6 both CONTRADICT
what is stated on this web page, while 246 is not deducible from it,
albeit I do not
see a contradiction for 246.)

So then I emailed some people in U-Colorado asking for info on the matter, but
no response came back.

Keep in mind, Richie did not cite any source nor provide any
independently verifiable data.

[Warren D Smith]

Also, this U-Colorado election had only 2 candidates. So it is kind
of weird to
have approval voting in a 2-candidate election, and even stranger to
pretend/assume that
such an election has any relevance to the quality or lack thereof, of
approval voting.



--
Warren D. Smith
http://RangeVoting.org <-- add your endorsement (by clicking
"endorse" as 1st step)

[Clay Shentrup]

On Wednesday, May 15, 2013 9:46:05 AM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

But anyway, was that data from U-Coloradio RR cited, actually valid?
I saw a U-colorado web page

   http://cusg.colorado.edu/content/spring-2013-election-results-cusg

about this election.
I do not see how Richie deduced 246, 51.9, and 51.6.

I remember this Twitter post:

"Just 20 minutes until polls close for #CUSG Spring 2013 election. As of 5 p.m. and impressive 6,424 votes had been cast. UNITE of INSPIRE?"

Maybe more voters showed up. I don't know. But with this, we'd have:

3354/6424 = 52.2% for Unite

3332/6424 = 51.9% for Inspire

So 262 people voted for two candidates. I wish I had a more reliable figure than the 6,424 though.

In fact, 51.9 and 51.6 both CONTRADICT what is stated on this web page, while 246 is not deducible from it, albeit I do not see a contradiction for 246.

It looks like 16 more people voted after the Twitter post, and somehow Richie got the final official number.

[Warren D Smith]

On 5/15/13, Clay Shentrup <cl...@electology.org> wrote:
> On Wednesday, May 15, 2013 9:46:05 AM UTC-7, Warren D. Smith (CRV
> cofounder, http://RangeVoting.org) wrote:
>
>> But anyway, was that data from U-Coloradio RR cited, actually valid?
>> I saw a U-colorado web page
>>
>> http://cusg.colorado.edu/content/spring-2013-election-results-cusg
>>
>> about this election.
>> I do not see how Richie deduced 246, 51.9, and 51.6.
>
>
> I remember this Twitter post:
>
> "Just 20 minutes until polls close for #CUSG Spring 2013 election. As of 5
> p.m. and impressive 6,424 votes had been cast. UNITE of INSPIRE?"
>
> Maybe more voters showed up. I don't know. But with this, we'd have:
>
> 3354/6424 = 52.2% for Unite
> 3332/6424 = 51.9% for Inspire
>
> So 262 people voted for two candidates. I wish I had a more reliable figure
> than the 6,424 though.

--well, the web page says 3354 = 50%
which assuming 50% means <50.5%, means #voters >= 6642.
It also says 3332=50%, which assuming 50% here means >49.5%,
indicates #voters <= 6731.

We also have 3354+3332=6686.
6686-6642=44 suggesting at most 44 voters voted for both.

Also your twitter post was at 5pm but the elections closed at 8pm.
This was, note, NOT 20 minutes later.

[Jameson Quinn]

The 50% on the page may well be fraction of total approvals, not fraction of voters approving.

2013/5/15 Warren D Smith <warre...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Warren D Smith]

On 5/15/13, Jameson Quinn <jameso...@gmail.com> wrote:
> The 50% on the page may well be fraction of total approvals, not fraction
> of voters approving.

--I do not believe that, because in the 2nd election on same page,
3150+3137+2968+2743+2996+2968+2758+2974+834+1012=25540
as compared to the "total number of voters" = 27457 (which must mean, in all
3 elections) and this was by far the most voted-election.

It would be amazing for approval voters to bullet-vote this often and
multi-approve this little in a 10-canddt election, I've never seen it
happen, so it must have been voters, not approvals.

[Warren D Smith]

http://cusg.colorado.edu/content/vote-count-spring-2013

gives votes at various times. The 5pm twitter total 6462 is confirmed.
But the polls closed at 8pm (it also says).
Extrapolating based on the rate of voting that day = 104 per hour, it
seems plausible the
final vote count would be about 6774.
As you recall my lower & upper bounds based on the "50%" on the
results web page, were
6642 and 6731. For Richie to be correct, the number of votes would
have had to be
3354/.519=6462 and 3332/.516=6457.
Assuming Richie's 51.6% meant "<51.55%" the largest the number of voters
could possibly have been would have been 3332/.5155=6463.6, i.e. at
most 6463 voters.

That would have meant at most ONE voter could have voted between 5 and
8 pm. That is absurd.

CONCLUSION:
Richie lied, just like he usually does.

[Warren D Smith]

> Assuming Richie's 51.6% meant "<51.55%"

--(Typo: I meant >51.55.)

[Clay Shentrup]

On Wednesday, May 15, 2013 3:48:36 PM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

--well, the web page says 3354 = 50%
which assuming 50% means <50.5%, means #voters >= 6642.

They divided by the total number of votes, not ballots.

Also your twitter post was at 5pm but the elections closed at 8pm.

5pm in California is 7pm in Colorado. Could that explain anything?

[Abd ul-Rahman Lomax]

At 07:30 PM 5/15/2013, Warren D Smith wrote:
>http://cusg.colorado.edu/content/vote-count-spring-2013

>
>CONCLUSION:
>Richie lied, just like he usually does.

Richie's comment was ad-hoc, about a recent event, not carefully
researched, and he did qualify his statement ("apparently"). Let's
stop this "lied" thing. We are still, ourselves, figuring out what
happened. I will separately look at what Richie wrote, and at what
information we have. Richie erred, perhaps, and so did some of us. I
don't see any cause for thinking anyone lied, except pure expectation
("usually") which is highly subjective.

It's time we stop this ****. <--- self-censored highly subjective
judgment of our behavior.

[Warren D Smith]

On 5/16/13, Abd ul-Rahman Lomax <a...@lomaxdesign.com> wrote:
> At 07:30 PM 5/15/2013, Warren D Smith wrote:
>>http://cusg.colorado.edu/content/vote-count-spring-2013
>>
>>CONCLUSION:
>>Richie lied, just like he usually does.
>
> Richie's comment was ad-hoc, about a recent event, not carefully
> researched, and he did qualify his statement ("apparently"). Let's
> stop this "lied" thing. We are still, ourselves, figuring out what
> happened.

--Richie attacked the CES video, perhaps costing you tens of thousands
of dollars, using bullshit he made up. If you believe the bullshit
was accidental on his part, then you still conclude he used bullshit.
There are "confirmed facts" and there is "bullshit." Richie and you
both ought to know the difference.

And also, his bullshit was just a priori highly likely to be absurd.

[Jameson Quinn]

Richie's complaint is that approval showed a couple hundred votes for both candidates. Remember that plurality or IRV don't magically make it so voters never do this; it just calls those ballots "overvotes" and refuses to even count them. 3% overvotes is perhaps a tad on the high side for plurality, but almost certainly on the low side for IRV.

2013/5/15 Warren D Smith <warre...@gmail.com>

Also, this U-Colorado election had only 2 candidates.   So it is kind

[Abd ul-Rahman Lomax]

I believe I commented on this before. However, because of subsequent
discussion, I'm looking at some issues here a bit more deeply.

In this analysis, I make a bonehead error, not one but two. I don't
correct them in the text, in situ, because I want to make the point
about how easy it is to do this, to make some incorrect assumption.
It's corrected at the end. I hope!

But I will give the summary here: Richie, for the Colorado CSUG
election, used the 5 PM total, not the final total at 8 PM, which we
don't know, but he then calculated the percentage of votes for each
candidate based on the final (8 PM) individual results. Using a a low
total vote count, then, he overestimated the number of dual
approvals. He did stated this as "appears," which must mean "appears
to me." He *also* appears to have made a math error because the 5 PM
total vote figure would give 224 dual approvals, not 246 as he stated.

At 12:16 AM 5/15/2013, Clay Shentrup wrote:
>http://lookingatdemocracy.org/submissions/14935-the-united-states-needs-approval-voting

[Rob Richie commented there:]

> >voting
> >for your second favorite can be a problem. Why?
> >Because it counts equally with your favorite
> >and, as a result, can lead to the defeat of your
> >favorite candidate. That means anyone who think
> >their favorite as a chance to win are going to
> >be wary of voting for their second choice -- and
> >we're heading back toward plurality voting dynamics.
>
>So, people are going to bullet vote. Which is only a problem if they
>do it even when it's not in their best interest. So Rob must be
>worried that people will bullet vote, even when it's in their best
>interest not to.

However, Richie's comment is correct. It *is* a problem. And Clay is
also correct. "Best interest" is not crisply defined, but this was a
quick interchange. I just went over FairVote's attempt to analyze
Approval Voting more deeply. It's a fairly impressive page,
coherently organized, that, however, presents conclusions, as if
proven, that do not follow from the evidence.

FairVote did not succeed in showing any example of voters voting
against their interest with Approval or the more-effective Bucklin,
which provides limited Later-No-Harm protection and thus likely
encourages the addition of more approvals.

Not Approval, not Bucklin, not IRV, in fact, will cause voters to add
additional approvals when they don't know the candidates except for
their favorite. With Approval methods, this type of bullet voting
will cause majority failure, and that's why we are looking at using
Approval in runoff voting primaries. With IRV, it causes majority
failure as well, in spite of FairVote's attempts to redefine
"majority." Hence IRV, used as a single-ballot method, commonly
results in majority failure in nonpartisan elections.

With partisan elections, voters know how to vote without knowing the
candidates. They vote for the party, and if, say, they are inclined
toward the Green Party, they may well vote for the Democrat in second
rank. Hence IRV *does work* with partisan elections, under minor
party spoiler conditions. It would work with certain other partisan
conditions as well. FairVote then completely confuses the types of
elections, and cheerfully recommends IRV where failure is quite likely.

We are suggesting that they stop doing this, and that they reassess
their position on voting systems, entirely. They could create, or
facilitate, an advisory body to study the matter more deeply, or
could cooperate with the Center for Election Science. If FairVote's
positions fully considered election science -- which does not mean
that they need to forget about "psychology," which is properly a
*part* of election science -- then we could -- and would -- fully
support FairVote, and I don't care what they did in the past. That's
History, not Election Science.

(Though the study of the history of *elections* is definitely a field
of interest, or our "science" becomes purely theoretical, an
Aristotelian thing of logical beauty from ungrounded assumptions, and
sometimes dead wrong.)

> >Another example was at U-Colorado this year. In
> >the 2-candidate race for student government
> >president, the result seems to have been 51.9%
> >to 51.6%. The margin was 22 votes, with 246
> >students apparently voting for both candidates.
>
>I.e. voters chose not to bullet vote, even when it was in their best
>interest to bullet vote. And the number of voters who did this was
>far greater than the margin of victory, meaning it easily could have
>made a difference to the outcome.
>
>Rob has done great job refuting himself here.

If someone actually does this, it would mean that they are not
filtering the evidence, or, alternatively, they are laboring under
illusions that lead to expectations divorced from reality. Richie
appears to think that there is something wrong with voting for both
candidates in a 2-candidate election. Indeed, one of us opined that
there was something strange about using Approval in a 2-candidate
election. I'd say that we have, ourselves, overlooked something.

Approval is very simple: Count All the Votes.

So if a voter votes for both candidates in a 2-dandidate election,
should those votes be counted, or should they not be counted?

Under default Robert's Rules, which *assume* that overvotes are voter
errors, the votes are still counted as in the basis for a majority,
but not as being for either candidate. Approval Voting had not gained
enough currency, actual usage, for the manual to cover it. It's an
obvious and harmless tweak. What these overvotes would cause, then,
in a 2-candidate election, where it was close like this, is election
failure. Robert's Rules would have the election be repeated.

However, Robert's Rules is looking for any result to be approved by a
majority. In public elections, it is well-established and is actually
practice, in the case of alternative conflicting ballot questions,
the situation is quite like a 2-candidate election under Robert's
Rules, except that the questions are separate. If no question gains a
majority, both fail. Even if the decision is "very important." And
nobody questions that. And if both pass, both have majority Yes, then
the question with the most Yes votes wins. Not the one with the
highest percentage Yes.

Approval Voting.

The Robert's Rules position is technical, not substantive, they never
justify not assigning overvotes to the candidates they support except
as "error," which is circular. If we have an election that is
difficult to repeat, accepting a multiple majority result, awarded to
the candidate with the most votes, is obviously a reasonable
compromise, compared to insisting that the majority make a clear choice.

*If a majority is required*, then voting for more than one, even if
there are only two, has an obvious purpose: The voter is supporting
both candidates. This gets obvious when write-ins are allowed. Voting
for both candidates, then, is easily interpreted as a message: I
prefer either of these to holding a runoff. It is only when write-in
voting, or other alternatives, are locked out, and when a majority is
not required, but is *forced* by the conditions, that it makes no
sense except as a kind of abstention that approves of both candidates.

My opinion has long been that a fair number of voters imagine that
all the votes will be counted.... so they will, on occasion,
overvote, laboring under that illusion. The illusion comes from
"common sense," actually. After all, what is wrong with Counting All
the Votes? These people simply are not aware of the counting rules.
They do see the Vote for One on the page, but some people still think
that the votes will be counted.

It would be of interest to do a survey on this. But, in fact, I'd
prefer to make that moot, and establish the tradition -- and rule --
of counting all the votes.

So what happened in this election?

The election report is punk, newbie naive. Whoever prepared that has
not looked at reports of public elections. The data necessary to
understand the election, even in a rather minimal way, is missing. We
have only total vote counts for the candidates, and under Approval,
that is generally larger than the number of ballots. We have better
reports from elections in the 1910's and 1920s, where Bucklin was
reported. The number of ballots cast is always reported. Generally we
see, also, the number of ballots that are spoiled in some way.
Sometimes blanks are separated out.

http://cusg.colorado.edu/content/spring-2013-election-results-cusg
What we have here is
>Total Voters: 27,457
>
>Votes Percentage
>
>Chris Schaefbauer, Ellie Roberts, Marco Dorado,
>UNITE 3354 50% Elected
>Logan Schlutz, Alexis Scobie, Colby Schwartz, INSPIRE 3332 50%

"Total Voters" is not the total number of voters who voted on this
question. I'm not sure what it is. It may be the total number of
*votes*, but not in this particular election. Notice it is the same
as the "total voters" as for the Representative-at-large election and
the Ballot Question. The Ballot Question would have been restricted
to Yes/No votes, and the total there was 6386 votes. Total voters
might have been the total eligible to vote, in which case they are
reporting turnout, and turnout is low. It's unclear because it could
also represent all voters who cast a single vote in the process. And
there were other elections in the process than the CUSG elections. It's a mess.

From the Representative-at-large election we can see that reported
percentages are percentages of total approvals, not voters. When we
have a preset conclusion in mind as a possibility, we can get very
confused; the preset conclusion here was "Richie lies." I don't get
that, certainly the evidence we have so far, on this number of dual
approvals issue, doesn't support it. And a math or logical error does
not equal "lie," or we have some liars right here in River City. It
just means that some people write first and think later, or "think
out loud" in writing.

Hey, I've done that! Not infrequently, in fact. In fact, most of my
writing is "thinking out loud," and what I find later may contradict
what I found earlier, and I don't always go back and fix stuff.
Hopefully, in finished documents, they have been thoroughly reviewed,
and not just by me.

Notice: this is a party-list election. So half of the election, the
determination of two "candidates" -- lists -- happened elsewhere, in
party process. Was that by Approval? Or by some advanced ballot
system? To understand what's really going on we need to know.

What we'd expect to see, from public election precedent, is

Eligible voters
Total ballots in this election
Blanks
Spoiled
Votes for each named candidate
Write-ins not individually totalled

With some voting implementation, Blanks and Spoiled may be locked
out. Write-ins may be locked out.

(It is common that Write-ins are totalled together, and they only get
counted -- which is a pain -- if all write-ins together could affect
the result. "None of the Above" is a write-in vote, usually. It
actually makes a difference if a majority is required. It has exactly
the named effect.)

So, how many total ballots were there?

http://cusg.colorado.edu/content/vote-count-spring-2013

This has 6462 votes as of 5 PM. The polls closed at 8 PM. Someone
pointed out the time difference between Mountain Time and Pacific
Time. That's irrelevant, the page was reporting, it's clear, in local time.

Richie reported 246 dual approvals. With a total of 6686 votes,
subtracting the 123 additional votes, this gives 6563 ballots with a
legal vote (1 or two approvals). This would require an additional 101
ballots be cast in the last three hours of the election.

Warren seems to think this impossibly high. I don't know why. Many
people will leave an action to the last minute, and will then rush to
do it. There was at least one tweet warning that voting was about to
close, Clay reported it.

Clay appeared to confirm the Richie calculations in a later post
where he reported the tweet. Howvever, the tweet gave no information,
even though reporting 20 minutes to go, other than the 5 PM figure.
So how Clay came up with "16 more people voted after the Twitter
post," I don't, know, except to suspect he made a math error or
misunderstood something, because that Twitter post only gave the 5 PM
total, not the "current vote" as of the Twitter time, which was
obviously later than 5 PM. It said "20 minutes" left, so it was at
7:40 PM, but ... it did *not* give the current vote total, probably
the sender did not know that. They had the 5 PM report, just like we
all have now.

Suppose the ballot total were 6462 (i.e., no more votes cast before
polls closed, an extreme). What would be the percentage of dual
approvals if there were no more votes, i.e., using the final vote totals?

The percentages reported on the CSUG result page are obviously
percentages of total votes, not ballot percentages. Someone tell them
how nuts that is with Approval voting. People want to know about
majority of voters approving, not majority of votes.

Unite: 3354/6462, 51.90%
Inspire: 3332/6462, 51.56%

Total of approvals: 6686. Amount over ballot count: 224 extra votes.
The ballot count Richie used must have been higher. Above, I
back-calculate from Richie's figure and come up with 6563 ballots,
requiring 101 ballots in the last three hours.

I just realized that I misread Richie's statement. I may have read it
as what he intended, not what he actually wrote. I need to look at this again:

>>The margin was 22 votes, with 246
>> >students apparently voting for both candidates.

The margin was indeed 22 votes, so he was likely looking at the final
results as to individual totals, but it is also possible that the
margin did not change with the additional votes, they would *tend* to
be equally for both parties, if we assume the final voters are a
sample of the voting population, which isn't a bad assumption.

IF there were 246 dual approvals, this would be 246 additional votes,
not 123. I had misread this. Chalk it up to not enough coffee. So
this would imply 6440 ballots. But there were more than this at 5 PM,
there were 6462 ballots.

Richie erred somewhere. Above, I show how easy this is to do. All it
takes is a moment's confusion, not recognized. Of course, maybe I'm
particularly dizzy. Let me check everything.

6686 total approvals, from result page. minimum total ballots: 6462
at 5 PM. As total ballot count increases, the number of calculated
dual approvals declines. I.e., if total ballots were 6682, there
would be no additional approvals.

With the 5 PM total ballot count, there would have been 224 dual
approvals, not 246 as Richie stated. The final ballot count was very
likely to be higher.

Richie's figure is too high. How high depends on the number of voters
between 5 PM and the end of the election at 8 PM. It is not
impossible that there were *no* such multiple approvals, but that is
also unlikely. Those kinds of votes get cast, realistically.

Richie's *implied* conclusion, "something wrong here" was unwarranted.

Shallow analysis, drupped on our video comment page. We'd invited
response by commenting on Instant Runoff Voting there. Comment pages
are *typically* knee-jerk, quickly written, not carefully considered.
For Richie to immediately think there was something wrong with an
Approval election would be, shall we say, not surprising.

As Warren pointed out, this was a 2-candidate election and tells us
nothing about Approval behavior in more complex elections.

How many voters would have had to vote in the last three hours for
there to be *no* multiple approvals?

224 voters, out of a total of 6686. 3.3%. Warren looks at this as if
the rate would be stable over the voting period, which is a radically
unwarranted assumption. Many people wait till the last minute to do
things like this. 3% is not unreasonable.

So, to review my own work, I go back to Warren's actual post, not my
memory of it.

>http://cusg.colorado.edu/content/vote-count-spring-2013
>
>gives votes at various times. The 5pm twitter total 6462 is confirmed.
>But the polls closed at 8pm (it also says).
>Extrapolating based on the rate of voting that day = 104 per hour, it
>seems plausible the
>final vote count would be about 6774.
>As you recall my lower & upper bounds based on the "50%" on the
>results web page, were
>6642 and 6731. For Richie to be correct, the number of votes would
>have had to be
>3354/.519=6462 and 3332/.516=6457.
>Assuming Richie's 51.6% meant "<51.55%" the largest the number of voters
>could possibly have been would have been 3332/.5155=6463.6, i.e. at
>most 6463 voters.
>
>That would have meant at most ONE voter could have voted between 5 and
>8 pm. That is absurd.
>

>CONCLUSION:
>Richie lied, just like he usually does.

Warren is more correct than I had thought. Not about Richie lying,
that's unwarranted.

Whatever Warren wrote about the 50% figure was bogus, because that
figure was clearly based -- or basable -- on the total approvals. It
*might* be based on total voters, but we know from below on the
results page that they are reporting percentages based on total
approvals, that's what they did with the representative-at-large
election. So, unless they decided to report differently for this
election from the other, which is unlikely, those percentages tell us
little about the dual approval rate.

I've written that last-minute voting can be higher in rate than at
other times. However, toward the end, coming from a declining pool,
it can *also* decline in rate. Essentially, it can be very hard to
predict, it can depend on many factors. A fascinating game on TV?

"Damn! Forgot to vote! Ah, well, doesn't really matter, these two
parties are Tweedle-dum and Tweedle-dee."

The last voters *might* have a higher percentage of dual approvals,
maybe. Maybe not.

The number of voters in the last three hours was somewhere between 0
and 224 voters. 0 is extremely unlikely. Warren is correct as to that.

It could be more if blanks (votes for neither of the two candidates
-- or write-ins --) were allowed.

Richie apparently was using a total of 6462 voters. That would give
the 51.9% and 51.6% that he reported. That was the 5 PM total, it's
not complicated. And that's all that happened. Simple error.

What *was* the real rate of dual approvals? We don't know yet, unless
someone gets that total ballot number.

An estimate, though, could assume constant voting rate for the last
day. I've pointed out how that could be way off, but it's better than
nothing, and just for the exercise, it would be 6774 votes total.
Total approvals: 6686. Are there ballots with no vote for either
candidate party?

There is no evidence here for dual approvals, but that is also
unlikely. I'd expect to see some. If the system didn't allow blank
ballots to be cast, then we can also infer that the voting rate
slowed down toward the end. Also not unreasonable.

[Abd ul-Rahman Lomax]

It's fraction of total approvals, almost certainly. In that 2nd
election, the percentages reported are indeed the individual
candidate totals divided by the total approvals. So those percentages
are understated. The winning candidate was approved by *more than* 12%.

The "total number of voters" is likely the total number of students
eligible to vote, nothing else make sense to me. This was extremely
poor reporting. We wish for more data from public elections, but it
is typically much more than this. If it is the total eligible number
(I think I looked at this before and thought it reasonable), this was
poor turnout for an internet election. Unite is not uniting the
students and Inspire is not Inspiring them. At least not to vote.
Given the way that the election is set up, as a *partisan* election,
voters may not perceive a difference between the two parties, and so
have little incentive to vote.

It may have been, if we assume 6600 total votes, as a rough estimate,
48%, which is a much better indication of "winner" than "12%." This
was a 5-winner election, so an approval threshold of interest could
be 1320 votes (6600/5). 8 candidates had much more than that, and
could have been considered fair as representatives. Inspire and Unite
parties dominated the vote, being almost equally balanced. All 8 of
the candidates with such a "quota" were the party candidates, i.e.,
four from Unite and four from Inspire. However, of the four declared
winners, three were from Inspite and one was from Unite. The other
seat was tied between an Inspire and a Unite candidate.

I think the Inspire and Unite parties are a game. And the students
want to play that game. The independent candidates (2) got bupkis.
roughly 25% of the support of the party candidates, i.e, roughly 15% at best.

Raw Approval voting is a bit rough in multiwinner elections.... I'd
be interested in an analysis of this election, if we could get raw
ballot data, as to how, say, party-list STV works. As an actual
experiment, Asset would be spectacular. Given the candidate set, the
result declared is probably not far from equity. But it won't be
thoroughly representative of the whole student body. It may be
imbalanced as to party affiliation, and that's a problem. This would
require 3 Inspire winners and 2 Unite winners. Instead there are 2
Inspire winners and 1 Unite, with one seat tied. If the tie goes to
Unite, it's fair. And that's obvious from the results, because,
really, these two parties are almost equal in proportion.

What would be of great interest would be the multiple approval rates
correlated with candidates voted for. We would expect such rates to
be higher for those who voted for the Independent candidates, *if
they knew the likely outcome,* i.e., that those candidates were hopeless.

Certainly Approval allowed those candidates to approve their
favorites and to also vote in the *real election.* But whether or not
they cared enough to even bother is something I can't predict.

I haven't looked at exactly how students voted (i.e., the ballot
design and technology used). It's not impossible that the voting
system only reported individual vote totals, which would be punk, and
poor design. Anyone know what they used?

[Warren D Smith]

On 5/16/13, Abd ul-Rahman Lomax <a...@lomaxdesign.com> wrote:

> I believe I commented on this before. However, because of subsequent
> discussion, I'm looking at some issues here a bit more deeply.
>
> In this analysis, I make a bonehead error, not one but two. I don't
> correct them in the text, in situ, because I want to make the point
> about how easy it is to do this, to make some incorrect assumption.
> It's corrected at the end. I hope!
>
> But I will give the summary here: Richie, for the Colorado CSUG
> election, used the 5 PM total, not the final total at 8 PM, which we
> don't know, but he then calculated the percentage of votes for each
> candidate based on the final (8 PM) individual results. Using a a low
> total vote count, then, he overestimated the number of dual
> approvals. He did stated this as "appears," which must mean "appears
> to me." He *also* appears to have made a math error because the 5 PM
> total vote figure would give 224 dual approvals, not 246 as he stated.

--now you're getting it.

[Abd ul-Rahman Lomax]

My full response on this issue noted that it would explain nothing.
The tweet was not at 5 PM, it was obviously after that, because it
reported the 5 PM total vote count from an official web page.

Clay was correct, it was total votes, not ballots. While those *could
be the same*, that's not likely. There would be, with this many
votes, at least a few dual approvals.

Things were said in this sequence of posts that were just plain
wrong, including the claim that the tweet was at 5 PM.

If one is looking at a page with information, it takes very little to
put up the link. That way, if a mistake is made, it can easily be
corrected. When we just report what we have seen without pointing to
the source, maybe someone can find it anyway, with a web search, but
not always, and it's a lot more work possibly for many people. Adding
the link is very little work, and only for one person.

Maximizing social utility, then, suggests adding links when it isn't
difficult. Certainly that's what we all want, right? Maximizing social utility?

So, what was that tweet? Clay was probably looking at it, because he
quoted it accurately. But not the full information with the tweet.
Easy to find it with the Twitter.com search tool on the text Clay had
given before.

https://twitter.com/coloradodaily/status/322524361422696448

><https://twitter.com/coloradodaily>Colorado Daily
>@coloradodaily<https://twitter.com/coloradodaily>

>
>Just 20 minutes until polls close for

><https://twitter.com/search?q=%23CUSG&src=hash>#<https://twitter.com/search?q=%23CUSG&src=hash>CUSG

>Spring 2013 election. As of 5 p.m. and impressive 6,424 votes had
>been cast. UNITE of INSPIRE?

>6:39 PM - 11 Apr 13

Okay, polls closed at 8PM Colorado time. The 5 PM total was 6,424,
from the site that was posting voter counts ("votes" is not correct
here, a lot more votes than that were counted, but, hey, this is just
a tweet.).

I remember watching all the flap about the Boston Marathon Bombing,
and conspiracy theorists were using tweets to "prove" that there was
a bomb set off a few minutes before the bombing, by the police. In
fact, they had confused tweet times because of time zone differences,
and the controlled demolition was of suspected packages,
substantially after the Marathon Bombing.

I'm not sure how tweet times are displayed, but may be in the server
local time zone, which appears to be Pacific time. That would be
earlier than Colorado time by one hour. So the tweet time would be
7:39 Colorado time, consistent with the claim of "20 minutes" until
the polls would close.

This tweet gave no more information than was on the official web
page, because this was *not* a new report of voter total.

Comedy of errors. Warren derived "5 PM" from Clay's post, which did
not say that at all. Clay, then, seems to have accepted the 5 PM
idea, and tried to explain it with a time difference, which was
incorrect. PDT and MDT differ by one hour, not two. Clay had not
looked at the actual tweet time, which is not displayed in the first
screen you see, unless you press a button. It looks like Clay also
assumed 5 PM from the mention of the 5 PM total, when obviously the
tweet would likely have been after that.

Now, I found, looking at related tweets, this article:

http://www.cuindependent.com/2013/04/15/close-student-government-election-closes-thursday-night/44710

This article gives little additional information. It reports the
results as being 50%-50%. It says that an election was never that
close before. Of course it was really not those percentages, the
percentages as votes are normally reported in approval elections were
higher, a bit, almost certainly. I'm amused to see the winners of the
Presidential election speculating that being out "more hours" caused
this. Maybe. But they didn't get that result in the Representative
elections, why not? The fact is that this was an extremely close
elections, and such things go different ways depending on the
weather, or maybe the phase of the moon. Maybe literally!

The article is misleading a bit. It says it that voters could vote
for as many or as few candidates as possible, regardless of
affiliation. That was true for the Representative election, but not
for the Executive election, which was for a slate of candidates, two
slates. Not individual candidates. It's for "three Presidents."
Apparently they are a little short on political history. Not the
first time this has been tried. Might work in that context. This will
tend to dilute responsibility. This is the first time, according to
the article, that this has been done.

Reformers tend to get a bit over-enthusiastic. "Let's try two reforms
at once." That way, if one fails, maybe we can blame it on the other
reform. That happened with Bucklin in Oklahoma. Hey, not only
Bucklin, but fractional votes for lower ranked votes (Range voting!).
And then not only that, but mandatory lower-ranked voting if there
were more than two candiates. Ooops! Unconstitutional, it was ruled
*and the entire procedure was tossed.*

Could happen to us in Arizona: IRV option (possibly unconstitutional
there) linked to Approval option (almost certainly constitutional).
Two eggs in one basket. Oops?

>In spite of their big win in the executive, the self-described
>"progressive ticket," Unite, claimed just one of five
>representative-at-large positions with candidate Juedon Kebede.
>Inspire's Ashley Prince, Brianna Majewski and David Bretl won three
>of the at-large seats. Each of the elected candidates won 12 percent
>of the vote.
>
>The final at-large position was a tie between Inspire's Joe Putnik
>and Rosana Rodriguez, each with 2,968 votes.

A victory by one slate over the other by 22 votes is a "Big win"?
Hello? It's only Big if the Executive is Big. Is it? Can the
Executive order that students be shot? Can it fine them Big Bucks?
Somehow I think not. No, it was an extraordinarily narrow win. Of
course, the reporter did not do research on that. She quoted someone
as saying what they don't know. ""I don't know that CUSG has ever
seen an election that's this close," Representative-at-Large Scobie
said. "The percentage was the exact same."

Brilliant. The word "exact" now comes to apply to rounded-off
calculations, and "percentage" only refers to integers. The reporter
is a journalism student, not a math student. But she *is* also a
politial science student, but ... voting systems are largely not
taught in political science courses. I've seen political science
texts that don't seem to be aware that there are options.

And the reporter, who also appears to be working as a real reporter,
repeats the claim that the victory was with 12 perecent of the vote.
That's highly misleading. We can't know exactly, without knowing the
total number of votes, but the winner was approved by, it seems, 48%
of the voters, in a 5-winner election. That's very good. 12% looks
punk, until one knows what it means. It's 12% of all approval votes,
but "total approval votes" is a meaningless statistic. Unless you are
the Minnesota Supreme Court trying to figure out a way to dump
Bucklin. Then, suddently, it's a big deal. Shocking! More votes than
voters! Unfair! Unconstitutional! (And totally meaningless. Only one
vote actually counts to pick a winner, ever, in a single round
single-winner election with an approval method.)

Now, actually, I liked the article. It did cover stuff of interest.
But critical thinking was missing. And ... nobody seems to have
looked at the issue of the tied election. I saw no further reporting on it.

She has an email address. Warren said he wrote to someone. Was it
her? I'd rather expect her to be responsive, but Warren may sometimes
ask questions in ways that will not inspire that. Speculation on my
part. I think there would be a way to, indeed, inspire someone in
Colorado to at least find out if the information is avialable that
would be needed to assess this election more fairly. This appears to
be in Boulder, Colorado.

[Abd ul-Rahman Lomax]

At 07:55 AM 5/16/2013, Warren D Smith wrote:
>On 5/16/13, Abd ul-Rahman Lomax <a...@lomaxdesign.com> wrote:
> > At 07:30 PM 5/15/2013, Warren D Smith wrote:
> >>http://cusg.colorado.edu/content/vote-count-spring-2013
> >>
> >>CONCLUSION:
> >>Richie lied, just like he usually does.
> >
> > Richie's comment was ad-hoc, about a recent event, not carefully
> > researched, and he did qualify his statement ("apparently"). Let's
> > stop this "lied" thing. We are still, ourselves, figuring out what
> > happened.
>
>--Richie attacked the CES video, perhaps costing you tens of thousands
>of dollars, using bullshit he made up.

Didn't cost me a penny, Warren. It may have cost CES some serious
money, but the lessons learned could be invaluable. I.e., don't start
a debate on the voting page. This started with what would easily be
read by Richie as attacks on Instant Runoff Voting. Entirely the
wrong place to do that. This invited a response, and his response was
mild. His response would not have cost CES the prizes. The action
that may have done that was the debate itself, and especially a
certain intemperate comment. Not Richie.

> If you believe the bullshit
>was accidental on his part, then you still conclude he used bullshit.

He used defective arguments that he probably believes. He also
praised CES for the first time I've seen. Warren, you've been
shooting yourself in the foot, for quite some time, with your
peevishness. How is it working for you?

> There are "confirmed facts" and there is "bullshit." Richie and you
>both ought to know the difference.

There are confirmed facts and there are common intepretations and
there are indiocyncratic interpretations. The world is far more
complex than a division into "confirmed facts" and "bullshit," and
much of what you write is *not* "confirmed fact," so what is it? By
your usage, it must be "bullshit." I don't think so.

>And also, his bullshit was just a priori highly likely to be absurd.

Right. But that says much more about you and your history with Rob
than it does about his *actual comments.* He wrote things there that
were *true* in the sense of being "common interpretations," i.e,
reasonable and only misleading if applied in unreasonable ways, and
that wasn't "bullshit."

Voters with a favorite who might actually win *will* be "wary" of
adding additional votes, if they have sufficient preference strength.
If your simulations don't take that into account when determining
voting strategy, they are ... aw, geez, I'm tempted to call them
"bullshit." In fact, they would merely be incomplete. Your
simulations are among the most valuable tools we have in
understanding voting systems, *even though they are obviously not
fully complete.*

What goes around comes around. Look at what Richie says as
"bullshit," instead of actually addressing the arguments, that *will*
come back to you, and he and his supporters will see your work in the
same way. It's *normal human communication,* Warren. You might
consider that you have something to learn about that, I certainly do,
and I can do a lot of what I'm now doing because I went for the
training and confronted my own very bad habits, a process that has
just begun, so it is not complete.

Most of what I'm doing now is not visible here, I have many other
interests that might be more important. And breakthroughs are now routine.

Meanwhile my training with regard to community projects also tells me
to turn over leadership, and I'm quite pleased by CES leadership as
it is, even though I might criticize what I see as missing, because
I'm a stand for CES *excellence*, not merely for being "good enough."

The CES video was *excellent* for purpose. All you could see, Warren,
was that it wasn't what you would do.

Yes, we knew that. We too have history and may expect you to behave
as in the past. And we can drop that right now, if you say so.

In fact, *I say so.* You are not limited to your past states. How about it?

Now, was the video what I'd do? No, probably not. It's probably
*better* than I'd do, that's what I think.

Ah, okay, maybe I'd do something as good. However, I'd do that by
involving others and *listening* to them. Which is exactly what I
want CES to do, to become *much more* than a small collection of
people on a board. And that is *totally* possible, and I just going
to say it is happening already.

[Abd ul-Rahman Lomax]

Yes. And you, Warren? Your previous posts did *not* "get it." Richie
did not just make up the total number of voters, he definitely used
the 5 PM totals, from shortly before the polls closed. That error
gave him the percentage votes, accurately, as he stated them. He
seems to have made a *small error* in the calculated number of dual
approvals. Fingers stumbled on a calculator, and the numbers didn't
*look* radically bad. Is there really a big difference between 224 and 246?

It would, then, make no sense for him to deliberately make a math error.

All of us were confused about this election at various points, there
is a whole history of misinterpretations, including mine. I only
found them by thoroughly documenting it, and reviewing the
documentation several times and noticing contradictions. That, in
fact, is why I wrote such tomes, I *document my research.* You see my
learning process. A few do benefit from that, who care to invest the
effort. That investment is a small fraction of what I put into it. I
make it easy, that's my goal. But people still have to think, and
maybe to check my sources.

Yes, because Richie has certain beliefs about Approval Voting, and
saw something that vaguely confirmed his vague notions, and he used
it, he didn't check it carefully. He's simply betraying what we
already knew about him: he is attached to a position and thus behaves
like people attached to positions behave. Helping him wtih this
propensity -- which can cause damage to his own goals -- is probably
something we can only do if we build rapport. It will never happen in
a million years if we keep calling him a "liar." That, in fact is the
general and normal behavior of ...

... people who are attached to a position.

[Abd ul-Rahman Lomax]

At 08:56 AM 5/16/2013, Jameson Quinn wrote:
>Richie's complaint is that approval showed a couple hundred votes
>for both candidates. Remember that plurality or IRV don't magically
>make it so voters never do this; it just calls those ballots
>"overvotes" and refuses to even count them. 3% overvotes is perhaps
>a tad on the high side for plurality, but almost certainly on the
>low side for IRV.

Eh? Richie raised the point with an implication that there was
something wrong, though we have to infer that from his comment.

Here is the comment:
http://lookingatdemocracy.org/submissions/14935-the-united-states-needs-approval-voting

It should be understood in the context that arguments against Instant
Runoff Voting or Ranked Choice Voting had been presented. Notice how
Richie leads. He's skilled. But I also think that his lead was
sincere. If not, so what? Authenticity will out, but there is also a
saying, "Fake it till you make it." If he's faking cooperation, maybe
it will turn into real cooperation. It's ancient wisdom, in fact.
From the Qur'an: "If they seek peace, seek peace. If they have a
plan, know that God has a plan, and God is the Best of Planners."
"God," in my glosses, is a synonym for Reality. That whole Qur'anic
message, to this one who reads it in Arabic, is essentially, "Trust Reality."

Those who don't trust that Reality is enough are in trouble, they
will inevitably be in suffering.

><http://challengepost.com/users/robrichie>Rob R 8 days ago
>
>Advocates of different methods of voting for single winner offices
>have a history of rather fractious behavior, but it's improving.
>Congratulations on that front to those seeking to work more cooperatively.
>
>That said, here's the core difference of opinion. Aaron Hamlin is
>right that "you can always vote for your favorite in approval
>voting." That's a good thing, and a good reason to like the system.
>But problem is that voting for your second favorite can be a

>problem. Why? Because it counts equally with your favorite and, as a
>result, can lead to the defeat of your favorite candidate. That
>means anyone who think their favorite as a chance to win are going
>to be wary of voting for their second choice -- and we're heading
>back toward plurality voting dynamics.
>

>Let's be concrete. Among the few examples of contested elections
>with approval voting are student elections for president the past
>couple years in Dartmouth. In both of them, the winner earned the
>approval of less than than 40% of voters. Perhaps that means voters
>just didn't know much about the candidates. But perhaps it means
>many didn't want to dilute their vote for their favorite by voting
>for their second favorite also.

>
>Another example was at U-Colorado this year. In the 2-candidate race
>for student government president, the result seems to have been
>51.9% to 51.6%. The margin was 22 votes, with 246 students

>apparently voting for both candidates. Did those 246 students really
>mean to do that? Did they have a difference of opinion that might
>have tipped the balance if only voting for one ticket?
>
>Ranked choice voting and traditional runoffs both don't have that
>property, which may explain why advocates of those approaches have
>been far more successful in having their reforms implemented and
>used in meaningfully contested elections.
>
>But kudos for having a conversation about reform and it's great to
>focus on the fact that plurality voting is simply unacceptable -- as
>this video successful demonstrates.
>
>Finally, let's keep in mind that if we want broader and fairer
>representation for more of us, we have to have more elections where
>more than one person wins -- and those winners can represent
>different opinions and voters. A big focus for some reformers is
>changing winner-take-all voting rules to fair voting plans like
>these ones FairVote has drawn for Congress at
><http://www.fairvoting.us>http://www.fairvoting.us

Richie's criticism of Approval here was *mild.* Yes, it's based on
some misunderstandings. I just spent several days deconstructing the
FairVote paper on Approval, written in 2011, apparently, probably by
Richie. It is full of shallow understandings, but Richie has never
been a deep thinker, as far as what I've seen. So what? He's a
tireless worker for what he believes is a necessary reform. He's
partially mistaken, that's all, and we can stand for his education;
if he isn't willing to grow, he'll become irrelevant rapidly, that's
all. That's just what happens. If we get stuck in what we believed
previously, we get old and become irrelevant.

His U-Colorado example was just a recent snap judgment. He made an
interpretive error, a simple one. He did say "appears." I note that
both Clay and Warren made similar interpretive errors on that
election, and I did, as well, myself. People make mistakes. As a
writer, I depend on others to notice them. It's rather irritating
when I've been writing something for years, many times, and it was
incorrect, and nobody said anything. I have to take responsibility
for that, though. I write too much to expect people to read it
carefully. So the truth comes out when I boil it down as polemic, and
someone with a different point of view, offended by my position,
points out the error. It's much more rare for a true friend to pull
me aside and say "Pssst... that argument was bullshit. Because see
[this link]." So mostly I depend on *enemies* to help me out. They
have serious motivation. And to benefit from this, I have to stop
assuming that everything they write is "bullshit," even though much
of it might be, so to speak.

Very bad habit. It actually disempowers us.

Richie pointed to a proportional representation plan for Congress. I
have not looked at it in detail. I'm quite sure that we could figure
out what is wrong with it. I'm suggesting that we not start there.
What is *right* with it? What is it standing for? And then, how could
we hellp FairVote be more successful in their stand? Maybe some
details of their plan could be improved, from the point of view of
election science, our field.

The basic identification of the "problem," Richie ended with this,
was single-winner elections. Most of us agree. So let's build on that
agreement.

[Clay Shentrup]

On Wednesday, May 15, 2013 5:18:51 PM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

--I do not believe that, because in the 2nd election on same page,
3150+3137+2968+2743+2996+2968+2758+2974+834+1012=25540
as compared to the "total number of voters" = 27457

I'm pretty sure that meant that there were 27457 eligible students. I.e. "active accounts enabled" in the voting system. But less than 7000 actually voted.

[Warren D Smith]

Usual Terminology:
An approval "vote" is the set of approvals emitted by a voter. Each
voter casts one "vote." No voter ever casts more than one vote.
"Non-voters" cast zero votes.

In this terminology, it is an inherently false statement that "the
number of voters differs from the number of votes."

> I'm pretty sure that meant that there were 27457 eligible students. I.e.
> "active accounts enabled" in the voting system. But less than 7000 actually
> voted.

--that must be true.

Let us review. There are several official web pages about this
election, all badly done:
A. http://cusg.colorado.edu/content/vote-count-spring-2013
says election 8am 8-apirl to 8pm 11-april, gives "vote counts" at various times.
Extrapolating to closing time, the total vote count exceeded 6462 and
was probably about 6774.

B. http://cusg.colorado.edu/content/spring-2013-election-results-cusg
gives alleged final results of 3 simultaneous elections.
They (badly) use "votes" to mean "#approvals."
The "percentages" evidently are as a fraction of summed #approvals (strange).
They may or may not be using approval voting.
For example, in the "rep at large" election, there were either 4 or 5 winners
in a field of 10 candidates. But classical AV only has 1 winner.
In the "exec" and "decision" elections there were only 2 choices. But
classic AV is usually only done in elections with more than 2 choices.
They have an asinine video about AV: http://www.youtube.com/watch?v=5D9xbzidIxc
which among its many flaws does not actually say which elections are using AV.

C. http://cusg.colorado.edu/content/elections
describes elections & who to contact. I tried to contact both as well as
Lora Roberts, director of communications for CUSG. All 3 attempts
failed to get any response.

D. http://cusg.colorado.edu/content/past-elections
They used a computerized-input & output voting system.
Past elections:
s2010: http://cusg.colorado.edu/content/Sprg2010-election-results
f2010: http://cusg.colorado.edu/content/cusg-election-results
["Vote Count = 3,692 (Highest In 20 Years!)"]
s2011: http://cusg.colorado.edu/content/cusg-spring-2011-election-results
vote counts by year: http://cusg.colorado.edu/content/vote-counts
min=739, max=10407.

CONCLUSIONS:
I see no evidence at all, that even a single voter voted for both candidates in
either 2-choice election. Richie's claims that
(a) 246 students apparently voting for both candidates
(b) the result seems to have been 51.9% to 51.6%
are both false. We know this because for (b) to be true, the
total #voters would have had to be in [6456, 6463]
which it was not. For (a) to be true, the total number
of voters would have had to equal 6440, which it did not.
But Richie's
(c) the margin was 22
was correct.

OK? So Richie lied as usual.

[Jameson Quinn]

OK. The conclusion seems to be that Richie was wrong. He should be told. Charitably, we could say that telling him would help him correct his mistake. Uncharitably, we could say that it would let us call him out if he repeats the lie.

I think that one person should tell him, and that it should be done politely. Even if he is lying deliberately, we gain nothing by saying that to his face, or by swamping him with multiple letters.

I would be willing to be the person. I'd also be happy if it were anyone on the CES board. 

Opinions?

Jameson

2013/5/17 Warren D Smith <warre...@gmail.com>

[Warren D Smith]

This page says the turnout was 25%.

http://www.colorado.edu/news/features/your-student-government-preliminary-election-results-announced

This means the total #voters was in [6727, 7001]
since the set of eligible voters was 27457 and since 25 is above 24.5
and below 25.5.

This is entirely compatible with the hypotheses that zero voters approved
both alternatives in either 2-choice election. It also is compatible
with my extrapolation
that the total #voters was "about 6774."

To my knowledge Richie has never issued a retraction or correction of
any error I ever pointed out to him over an span longer than 10 years.
I also have encountered many other people who have pointed out
errors to Richie. None of them ever reported RR acknowledging or
correcting his lie. In the present case, Richie's goal was to
prevent CES from winning that money prize for the video, so he
invented a story to get the job done.

[Abd ul-Rahman Lomax]

At 10:38 AM 5/17/2013, Warren D Smith wrote:
>Usual Terminology:
>An approval "vote" is the set of approvals emitted by a voter. Each
>voter casts one "vote." No voter ever casts more than one vote.
>"Non-voters" cast zero votes.
>
>In this terminology, it is an inherently false statement that "the
>number of voters differs from the number of votes."

Aw, you can define words any way you want, and then make others wrong.

The Minnesota Supreme Court explicitly defined votes as marks on
ballots. A mark on a ballot is definitely a "vote."

While Warren's definition is possible, it makes mincemeat of common
usages. For example, with cumulative voting, voters may have more
than one vote, which they can divide up how they choose. They also
cast only one ballot. Is the ballot a "vote," or is it a device which
may cast one vote or more than one?

With an IRV ballot, a voter may case a series of votes, and only one
is active in amalgamation at a time.

On an Approval ballot, more than one vote may be active at one time,
and how can we count them in two separate candidate totals if voters
have only cast "one vote."

What Warren above calls the "usual terminology," then, is very much
not the usual terminology. We have defined Plurality as a one-vote
sytsem, which it is. We have defined Approval as a vote-for-as-many
as you choose system. If you cast one vote for A, and you choose to
vote for B, how many votes have you cast? One or two?

Yes, you can use definitions to say you only cast one vote. That is,
you cast one vote for (A+B). During the amalgamation process,
however, this will generally be counted as two votes, one for A and one for B.

"The number of voters differs from the number of votes" was a
reference to Brown v. Smallwood, where they first talked about, and
cited precedent for, what matters is the number of *voters*
supporting a candidate, not the number of "marks." "Marks" are votes,
in the ordinary meaning. You can define a set of marks as a single
vote (i.e, a single voting pattern) or you can defined it as multiple
marks. Which you will do depends on your purpose in communication.
What's your purpose, Warren?

the BvS court then proceeded to add up the "votes" for each
candidate, and, behold there were more such *votes* than voters. They
implied that there was something wrong with this. It appears to have
been a major basis for their decision. The other basis was that
somehow a voter who votes for more than one has an unfair advantage
as against a voter who bullet votes. That's a more complex argument.
It's highly defective, and no other court has confirmed this. And
they knew they were breaking precedent, and prevailing legal opinion,
and they didn't care. Actually, it was a three-judge court, if I
recall correctly, and there was a vigorous dissent from one justice.
And, as well, lots of complaints from legal authorities. that comes
out in the request for rehearing.

> > I'm pretty sure that meant that there were 27457 eligible students. I.e.
> > "active accounts enabled" in the voting system. But less than 7000 actually
> > voted.
>
>--that must be true.

It is *obviously* true. I researched this six ways till Sunday, and
reported it. It may or may not be "active accounts," it might be
"eligible students," but that is of little importance to us. That
turnout was low -- roughly one-quarter -- indicates something else to
me, dissatisfaction with the *system*. The system that would maximize
turnout, my theory, is Asset. Every vote counts to create
representation in the student government for every voter.

>Let us review. There are several official web pages about this
>election, all badly done:
>A. http://cusg.colorado.edu/content/vote-count-spring-2013
>says election 8am 8-apirl to 8pm 11-april, gives "vote counts" at
>various times.
>Extrapolating to closing time, the total vote count exceeded 6462 and
>was probably about 6774.

That's possible. I.e., it almost certainly exceeded 6562, but how
much depends on a universe of complex factors. The rate could vary
all over the map. The maximum number of voters, I already reported
all this, would be the vote totals with no double approvals. Double
approvals will indeed be very low, but we don't know how low, so it's
guesswork.

>B. http://cusg.colorado.edu/content/spring-2013-election-results-cusg
>gives alleged final results of 3 simultaneous elections.
>They (badly) use "votes" to mean "#approvals."
>The "percentages" evidently are as a fraction of summed #approvals (strange).
>They may or may not be using approval voting.

That's correct. I thought of that, but didn't have evidence either way.

>For example, in the "rep at large" election, there were either 4 or 5 winners
>in a field of 10 candidates. But classical AV only has 1 winner.

It's approval voting, because the number of approvals are not limited
to the number of winners. It looks like most voters did not add the
full complement of votes as in plurality-at-large. That could be why
even the top candidate didn't pass 50% approval, it seems.

>In the "exec" and "decision" elections there were only 2 choices. But
>classic AV is usually only done in elections with more than 2 choices.

Warren, this was approval-at-large, which uses approval voting
multiwinner. The same thing is proposed for Arizona: the primary is
two-winner. Vote for as many as you choose. If it were "vote for no
more than two," it would be plurality-at-large.

You *can* run approval in a two-winner election. It's a measure of
approval, which is separate from making a choice of candidate. A vote
for two candidates in a two-candidates does abstain from the choice,
but has a different obvious meaning than a vote for neither. In
systems where a majority is required, it's obvious the difference it makes.

>They have an asinine video about AV:
>http://www.youtube.com/watch?v=5D9xbzidIxc
>which among its many flaws does not actually say which elections are using AV.
>
>C. http://cusg.colorado.edu/content/elections
>describes elections & who to contact. I tried to contact both as well as
>Lora Roberts, director of communications for CUSG. All 3 attempts
>failed to get any response.
>
>D. http://cusg.colorado.edu/content/past-elections
>They used a computerized-input & output voting system.
>Past elections:
>s2010: http://cusg.colorado.edu/content/Sprg2010-election-results
>f2010: http://cusg.colorado.edu/content/cusg-election-results
> ["Vote Count = 3,692 (Highest In 20 Years!)"]
>s2011: http://cusg.colorado.edu/content/cusg-spring-2011-election-results
>vote counts by year: http://cusg.colorado.edu/content/vote-counts
> min=739, max=10407.
>
>CONCLUSIONS:
>I see no evidence at all, that even a single voter voted for both
>candidates in
>either 2-choice election. Richie's claims that
>(a) 246 students apparently voting for both candidates
>(b) the result seems to have been 51.9% to 51.6%
>are both false.

No. "Seems" was accurate as to what appeared to Richie. Warren, you
need to deepen your understanding of "true" and "false." Any
statement may be interpreted as true and any statement may be
interpreted as false. It is clear that Richie did not completely
understand what he was seeing. Or we could alternately interpret this
as a deliberate and wilful attempt to deceive people looking at that page.

I prefer the former interpretation, most of all because we look a lot
better making that intepretation. But it's also sincere. Richie is
not likely to make a bonehead mistake, easily exposed as an error,
unless he doesn't see it as one. Even if he is Evil Incarnate, the
Grand Panjandrum.

And he isn't. He's just a political activist, a hard worker, somewhat
successful, and not terrible careful about details. Lots of people like that.

There is *no direct evidence* that any voter double voted in the
executive election. I confirm that. It's likely, however, that some
did. The matter hinges on the total number of votes in the election,
which was not reported.

Richie was led astray by some of the very facts that confused, not
only him, but us. We made assumptions that just weren't true. He
wasn't far off, in fact, his basic error was taking the 5 PM total
and using it instead of the final result, which wasn't reported.
That's really frustrating, by the way. Why didn't they complete that
election report with the total votes? So simple.

I'll tell you why. These are students, and most are not trained in
complete integrity. That's all. It's actually normal. Now, can we
help them? To do that, we need to be able to *communicate* with them.
Part of the skill of communication is how to elicit responses.

Yes. Your attemps failed. Failed to get responses. So, how to get a
response? Want ideas?

> We know this because for (b) to be true, the
>total #voters would have had to be in [6456, 6463]
>which it was not. For (a) to be true, the total number
>of voters would have had to equal 6440, which it did not.
>But Richie's
>(c) the margin was 22
>was correct.
>
>OK? So Richie lied as usual.

Warren, I conclude that you are quite incautions with your language.
There is *no* evidence here for a "lie," which means a statement made
knowing it is false. That's actually very unlikely. There is some
usage of "lie" to mean a false statement with highly disliked
implications, but this usage demolishes the possibility of
cooperation. If you are in an adversarial situation, basic social
rule. Never back your opponent into a corner where his only
alternatives are severe loss of face (which is worse than death for
some people) or a direct attack on you. That's true for racoons and
it's true for humans.

Richie, writing about Approval Voting, has, I'd agree, commonly
erred. Were those "lies"? Well, I'd not even ask that question. First
of all, I'd correct errors, and especially where they can have a
public impact. If I cry "lies," the real-world effect is that I look
like just another political fanatic, I'll be less effective than if I
don't say that. I have made rare exception where deliberate deception
was clear from clear evidence.

I used to stop at that. (And I didn't always follow it. The constant
repetition of arguments by Richie, in particular, in internet fora,
where he ought to have known better, I thought, i.e, he was familiar
with contrary fact, I assumed, did get to me sometimes.)

My training now would be to look for Richie's "stand." What is he
actually out to accomplish? I.e., what is the *purpose* underneath
his activity, and we, when we look at this carefully, often uncover
layers. One layer is personal survival. Most of us have been,
effectively, trained to do this. We will do what makes us look good
or "right," we will make others wrong, and we will avoid domination.
This is *normal personal survival,* and the amygdala takes over
easily. However, humans are capable of something else, which is what
I will call collective communication. We actually seek the common
welfare, and there are lots of signs that this is instinctive. The
"stand" I'm looking for, then, is not the personal stand that wants
to "succeed" as an activist, i.e, to "win," but that is deeper and
greater than that.

And it probably is coming out in his suggestions of cooperation with
regard to proportional representation. He got caught in a "pro-IRV
stance" as a normal political-activist taking of a position, and our
system encourages such biting onto a goal like a bulldog and stopping
looking anywhere else. However, if he doesn't come up for air
periodically and reassess, this can take him into major failure, eventually.

Another part of my training suggests looking at my own errors. How
have *I* been ineffective in communication? Mostly, from a lot of
feedback I"ve been getting, I create a response that sees me as
arrogant, know-it-all, and insisting on being "right." And, again, my
training is to take responsibility for that. Instead of blaming them
for blah, blah, blah.) Technically, my training is to take
responsibility for every complaint I might have.

That's not about "truth," it's a "stand." It's called a "committed
assessment," in the training. The stand is taken because it is
*powerful*, it enables *transformation*, not merely "improvements."
And, I'll testify, it works.

You are saying that Richie "lied," a highly inflammatory statement,
which might as well be designed to perpetuate a conflict, without any
evidence that he knew the truth and wrote what he wrote, anyway. That
says far, far more about you than about him.

You can take advantage of this opportunity to recognize that -- I'd
say it's simply logical from the record, and using the ordinary
meanings of words, i.e., how people actually interpret them -- or you
can blow it off as the ravings of that lunatic, Abd.

It's your choice. As someone who wishes you the best, and not only
because of all the hard work you have done, but, as well for your
original stand to transform politics for the ... for what? How would
you put it? -- I trust that you will choose well. Your future is unlimited.

[Warren D Smith]

I recommend ignoring Lomax. He outputs over 1000 pages of posts on
voting methods per year, of which perhaps 5 pages worth have some use.

Anyhow, I stand by what I said -- I simply want to not allow Lomax
to give the impression that he has in any way refuted my posts here,
nor that I in any way agree with any such assessment, no matter how
many 100s of pages Lomax posts. There is no need for 100 pages on
this topic, it is a simple matter of a few numbers.

[Abd ul-Rahman Lomax]

At 10:47 AM 5/17/2013, Jameson Quinn wrote:
>OK. The conclusion seems to be that Richie was wrong. He should be
>told. Charitably, we could say that telling him would help him
>correct his mistake. Uncharitably, we could say that it would let us
>call him out if he repeats the lie.
>
>I think that one person should tell him, and that it should be done
>politely. Even if he is lying deliberately, we gain nothing by
>saying that to his face, or by swamping him with multiple letters.
>
>I would be willing to be the person. I'd also be happy if it were
>anyone on the CES board.
>
>Opinions?

I think we should first make adequate effort to find out the vote
totals. Warren reported certain efforts, they are by no means all
that could be done. I'd have to find time for it, I could make some
efforts, and it also occurs to me to ask Jan Kok, who lives in Fort
Collins, Colorado. The University is in Boulder, about an hour away.

But there may easily be better options than that.

While anyone could attempt to get the information, it might be better
if someone is authorized to act for CES, and make the inquiry in a a
professional manner. That includes making phone calls, and it might
take a lot of calls. I do know how to do this kind of work. But I
could also advise or coach someone, I've also done that.

Actual election results are of high interest to CES. We need examples
of approval voting. These elections were not good tests, for reasons
that Warren -- and others -- have pointed out. But we need to know the details.

Actual ballot votes, if available, would be spectacular, showing
voting patterns in the Representative at large election.

The simplest thing we need to know is the number of voters
participating in each election.

It's possible nobody knows that, which would point to very poor
voting system design.

In any case, who will volunteer to investigate for us? This means not
taking "No," or "I don't know" for an answer. It could mean being
persistent and patient. It is not necessarily a huge job, but it
might be, at places, difficult.

The fastest way to develop skill in this kind of activity is to
declare that one will do it, and then when one fails, stand back up
and keep going for it. I'll be happy to help.

When we have a volunteer, we can then seek CES authorization, which,
as long as the volunteer is reasonably suitable, won't be difficult.

Let's do this right.

The volunteer need not live in Colorado ... if it turns out we need
personal presence, rather than merely phone calls, we can look for
someone close.

[Clay Shentrup]

On Friday, May 17, 2013 8:38:13 AM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

Usual Terminology:
An approval "vote" is the set of approvals emitted by a voter.

No. When people say "votes" they mean "approvals". I've never encountered anyone who stated the opposite.

[Warren D Smith]

--if you do not like my preferred terminology, then fine; this does
not affect my arithmetic demonstration in the slightest. Three of
Richie's four numbers were wrong, and since
there was an internal contradiction within his 4 numbers and 2 more
numbers which he had to have known, this cannot be ascribed to Richie
merely being misinformed.

[Clay Shentrup]

On Friday, May 17, 2013 12:18:37 PM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

I recommend ignoring Lomax.   He outputs over 1000 pages of posts on voting methods per year, of which perhaps 5 pages worth have some use.

This has been my strategy for years.

[Clay Shentrup]

On Friday, May 17, 2013 8:38:13 AM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

C. http://cusg.colorado.edu/content/elections
describes elections & who to contact.  I tried to contact both as well as
Lora Roberts, director of communications for CUSG.  All 3 attempts
failed to get any response.

Doh! I forgot that she already responded to me. It was 6,686 votes.

[Clay Shentrup]

She said: "The total number of voters was 6,686 in this election, which is the third largest in the last ten years."

[Clay Shentrup]

A CUSG official emailed me:

"The total number of voters was 6,686 in this election, which is the third largest in the last ten years."

So Unite got 50.2%, and Inspire got 49.2%.

[Warren D Smith]

Fine. 6686=3354+3332 exactly.

Therefore, the total number of voters who voted for both candidates,
was not 246,
but exactly ZERO.
Be nicer if you'd pointed this out before I bothered those people at CU.

[Warren D Smith]

And no,

"So Unite got 50.2%, and Inspire got 49.2%"

was false.

They need to add to exactly 100%.

[Clay Shentrup]

On Friday, May 17, 2013 3:43:08 PM UTC-7, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

And no,
"So Unite got 50.2%, and Inspire got 49.2%"
was false.

They need to add to exactly 100%.

Just a silly typo. I obviously meant 49.8%.

[Clay Shentrup]

My response to Richie (his comments removed)

A CUSG official emailed me:

"The total number of voters was 6,686 in this election, which is the third largest in the last ten years."

So Unite got 50.2%, and Inspire got 49.8%. So every single voter voted for exactly one of the candidates in that race.

 

..

If you were referring to this page..

http://cusg.colorado.edu/content/vote-count-spring-2013

..it cites 6462 voters as of 5pm. But the polls closed at 8pm, so if you extrapolate based on the previous rate, you get about 77 votes per hour, for an estimated grand total of 6693. Which is incredibly close to the actual total of 6686. But you apparently didn't extrapolate.

So then what number did you use? If you had used 6462, you would have gotten 224 "overvotes". But you said:

"The margin was 22 votes, with 246 students apparently voting for both candidates."

Meaning you used a figure of 6440 voters. Which is 22 less than the 6462 figure cited from the (three-hours-early) 5pm total cited on the CUSG page. I have no idea how you even got this 6440 number.

Incidentally, 22 votes happens to be the margin of victory. What an odd coincidence. (??)

[Warren D. Smith (CRV cofounder, http://RangeVoting.org)]

To correct Richie, I suggest referring to this web page:

http://rangevoting.org/RichieOnApproval.html

[Abd ul-Rahman Lomax]

At 11:11 AM 5/17/2013, Warren D Smith wrote:
>This page says the turnout was 25%.
>
>http://www.colorado.edu/news/features/your-student-government-preliminary-election-results-announced
>
>This means the total #voters was in [6727, 7001]
>since the set of eligible voters was 27457 and since 25 is above 24.5
>and below 25.5.

Garbage in, garbage out. They said "25%." Now, how accurate was that
figure? Remember, these are not mathematicians.

There is a possibility that the Executive election wasn't Approval.
Nobody has pointed to conclusive evidence that it was Approval.
Nobody has pointed to a ballot with the actual instructions.

But it is also entirely possible that it was Approval and that there
were some multiple approvals. With that number of voters, I'd expect a few.

>This is entirely compatible with the hypotheses that zero voters approved
>both alternatives in either 2-choice election. It also is compatible
>with my extrapolation
>that the total #voters was "about 6774."

I went over the available evidence with a fine tooth comb. The idea
of taking that turnout figure and assuming that they did an actual
calculation, instead of just roughing it, isn't a bad idea, but it's
totally unreliable. Did they have the *actual turnout*? We don't
know. If they did, where did they get the figure?

It is quite possible that they simple stopped locking out overvotes
from all elections. But that doesn't mean that whoever reported on
this election for the Student Goverment knew that the votes may not
sum to the number of voters. They might have had only the same data we had.

So, with a sum of votes of 6686, we'd have, with no dual approvals,
24.35% turnout. It is entirely possible that a reporter would write
this as "25%."

The total number of voters generally is considered as the total
number of those submitting a valid ballot. Since a ballot might
contain no vote for a particular election, even if we know the total
number of voters (i.e, if the 25% figure were actually in the range
of 24.5% - 25.5%, that would be 6727 - 7001 voters.

However, the number of voters in the Executive election cannot exceed
6686. That's because a voter who does not submit a valid vote in an
election is not considered to have voted in it.

Warren, you've missed something. Normally, the "voters in an
election" refers to a specific election, i.e., to ballots cast in
that particular election, not to all who voted in all elections held
at the same time. The official article you poiint to would be using,
if they are not simply making a mistake -- which must always be
considered possible -- the number of people who voted in *any* of the
elections. They "turned out to vote." So there will be people who
didn't vote in the Executive election who did vote in other
elections. Indeed, if we look at the ballot question, it's obvious
that there were different numbers of people participating in the two
elections: both were two-option elections, Check This or Check That.
Yet the total of votes in the Executive election is 6686 and the
total is, for the ballot question, 6386. There would be people who
only voted in one of these elections, and there would be people who
voted in both (most people), and there would be people who voted in
neither (but voted in the Representative-at-large election.)

>To my knowledge Richie has never issued a retraction or correction of
>any error I ever pointed out to him over an span longer than 10 years.

You are not, here, exactly the paragon of acknowledgment of personal
error. Lead the way.

(I will acknowledge that you have admitted errors many times and
routinely issue corretions. You are also in a very different position
from Richie. You are, at least roughly, a scientist. He is not. He's
a *political activist*. Which is very similar to "politician."

>I also have encountered many other people who have pointed out
>errors to Richie. None of them ever reported RR acknowledging or
>correcting his lie.

Of course not. Have you ever acknowledged any of your own lies?

Now, how do you feel about being asked that question?

By the way, I don't recall seeing any lies from you. I've seen you
make mistakes, including gross ones, including ones that you have
never corrected. You were given adequate information to find your own
error, but you didn't. So if you judge Richie for what he did, you
might consider that it can reflect on you. You have training in the
sciences, he doesn't. So who is more responsible?

> In the present case, Richie's goal was to
>prevent CES from winning that money prize for the video, so he
>invented a story to get the job done.

No. That's preposterous. He had no power to do that. Do you imagine
that the judges would award a prize based on Approval Voting having
no flaws? Do you think that the judges would see comments on the
video page and reject the video because Someone On the Internet says
something about it is Wrong?

Richie raised a valid point about Approval. We know that he assigns
higher significance to it than it deserves. He raised a point about
some university elections that is going to general a "huh?" from
those not familiar with the issues. None of this would take an award
away from CES. If they research Approval Voting they will find that
it is respected as a voting method, and that FairVote, Richie's
organization, has a position paper out there opposed to it.

This is not an "election science" contest. The video isn't being peer
reviewed for the science in it or about it. The judges are in a
consideration period.

I was concerned, first of all, that when someone brought up Instant
Runoff Voting, on our comment page, we criticized it. And that, then,
invited response and, sure enough, Richie showed up. His comments
were quite mild, and congratulatory overall. If any judge figures out
who he is, that might *increase* our chances of getting a prize. I
very much doubt that Richie would be out to stop us from gaining the prize.

*However*, if he did have that goal, he did the worst damage he could
possibly do, by pushing our buttons in a certain way, perhaps he knew
that Clay would attack him. And *that* is what could damage our
chances; if they figure out who Clay is.

Warren, I think you missed the sequence. The "interchange" started
when Clay commented on the RCV video by FairVote Minnesota.
http://lookingatdemocracy.org/submissions/15145-who-s-your-favorite-president-how-ranked-choice-voting-works

Richie responded there. I was concerned that Clay was commenting *at
all*, and said so at the time. ("Center for Election Science" was in
his signature; he later removed that, it appears.) Meanwhile this
comment was cited by another in our video comments, and Aaron echoed
a mild criticism of Ranked Choice Voting. It was, simply, no suprise
that Richie would show up, within a day. He was polite,
congratulatory, defended RCV mildly, and pointed out an *actual
problem* with Approval Voting, and made his comments about Dartmouth
and CU. Which would lead to very little effect on others.

But Clay, then, on the next page, went after Richie *personally*,
linking to his own blog as if it proved something (I covered that, it didn't.)

Clay has never apologized publically for this. He did stop talking
about it (which is to be commended).

All those comments *will* be read by the judges, if they are
considering our video for a prize. How do they make us look?

We went to great trouble -- and expense -- to create an image of
"clean" and "scientific." The quality of our video is higher than
that of any other video there. But then, how do we look on the
comment page. The most I can hope is that the judges don't realize
that the comments were from CES board members, especially Clay's.
Clay's comment on the RCV video page were mild for Clay, but I saw
this as a sign that he was going into his "someone is wrong on the
internet" mode. Very inappropriate for a CES Board member.

You are still caught, yourself, in an amateur, crank, pattern of
behavior. It's cost you your board position with CES. It could cost
you much more. Warren, stop it.

With the CSUSG student election, you are strongly attached to being
right. It's a stupid little piece of business that only has any
significance if we get the real information. Instead, you are looking
for ways to say, "See, I could be right!" You made a whole series of
errors and you keep making them, such as assuming that a turnout
figure would relate to the *specific election.* Strictly speaking,
there were three CSUSG elections, not one, and each would have a
different figure, but because of a unified voting process, I assume,
there could be a record of the total number of voters who voted in
any one or more of those elections.

Or the writer simply assumed it from the voting figures, having no
clue about "approval," and didn't care about precision. Happens all
the time, Warren. In the Representative-at-large election, the writer
didn't give any information on matters of interest to anyone who
wants to see how the method is working.

http://cusg.colorado.edu/content/spring-2013-election-results-cusg

That was a 5-winner election, and had always been done by
plurality-at-large, I'm guessing. So they are quite accustomed to
seeing vote counts much higher than the number of voters.

The total of votes there was 25540. If I assume that there were no
dual approvals in the Executive election, and I assume the same
number of voters, I come up with 3.82 votes per voter. The voters are
not using their voting power, many of them. Under plurality-at-large,
they could have cast up to 5 votes each. Now, it would be interesting
to see the results from former elections. My guess is that the number
of voters per voter, here, would be larger, because some voters might
vote for more than 5.

Not a lot, though. It was a partisan election. Each party put up four
candidates, and there were two independents (who got few votes).

Here is my analysis. Most voters voted a party ticket, and stopped
there. That would be 4 votes per voter.

This election was a little like the Executive election. Whose idea
was it to use Approval here? Yeah, great idea, but these elections
were not the place to use it, necessarily. A two-candidate election
for a triple executive, i.e., two slates corresponding to two
parties? (That is, politically, one very weird idea. Previously it
was also two slates, a President plus two vice-presidents)

Someone tell these people about Asset!

Yes, Richie's comment was in error. But the information from CSUSG is
quite confusing, not the way we -- including Richie -- are accustomed
to seeing election data presented.

[Abd ul-Rahman Lomax]

At 02:18 PM 5/17/2013, you wrote:
>I recommend ignoring Lomax. He outputs over 1000 pages of posts on
>voting methods per year, of which perhaps 5 pages worth have some use.

To Warren. In any case, Mike Ossipoff apparently ignored what I wrote
about his problem with voting for the video, and, as a result, didn't
vote. Warren has ignored a great deal about his peevishness, and, as
a result, has lost a great deal of personal power.

His choice. I write a great deal, indeed, sometimes. I do not write
1000 pages per year on voting systems. There are times when I write
many pages per day, but most days I don't write anything. There has
been high activity of interest lately on this list, because of

Arizona HB 2518
The video
And some possibilities that have opened up for cooperation with FairVote.

> Anyhow, I stand by what I said -- I simply want to not allow Lomax
>to give the impression that he has in any way refuted my posts here,
>nor that I in any way agree with any such assessment, no matter how
>many 100s of pages Lomax posts. There is no need for 100 pages on
>this topic, it is a simple matter of a few numbers.

No, it's a matter of the behavior of voting systems advocates. The
numbers are just an excuse. I wrote an exhaustive report on the
controversy, and Warren simply ignored it, wrote information clearly
contradictory to what was stated in that report, with sources, and
continues to justify himself, all to prove that he was right about
Rob Richie "lying." *No evidence has been presented for that.*
Evidence was presented that Richie erred, and I detailed that conclusively.

My activity, some years back, with CES, disappeared for a time
because Warren had become privately abusive and basically ... crazy.
He was *certain* he was right about a topic where I'd become expert
(not related to voting systems), and where he knew very little, but
only some theory -- which did not apply. And then he acted to attempt
to disempower me with respect to this organization, behind the
scenes, because I was obviously crazy.... and he didn't want crazy
people around.

Warren is a loose cannon. Fortunately, he's no longer on the Board.

I don't care if Warren thinks I've "refuted" his posts or not. He's
made errors, and he's made interpretations that, if they were
associated with CES, would be politically damaging.

And my purpose is to establish that he does not speak for CES, in
spite of being critically involved in the founding of the
organization. CES will be, I declare, moving to cooperate with
FairVote in many ways. We will still stand for election science,
that's our core mission.

I am, however, not a board member or officer of CES. I'll claim that
my views are broadly representative of many CES members, but I claim
no authority for that. I can, and do, communicate with the CES board,
when I have a concern, and I'm heard. I'm not in charge.

[Abd ul-Rahman Lomax]

Warren made up his definition to try to make someone else wrong.

He is correct that the term could be used that way, but, quite
simply, what he was criticizing it was using the term as commonly
defined. I already explained this, in detail....

[Warren D Smith]

Lomax:

My activity, some years back, with CES, disappeared for a time because
Warren had become privately abusive and basically ... crazy. He was
*certain* he was right about a topic where I'd become expert (not
related to voting systems), and where he knew very little, but only
some theory -- which did not apply. And then he acted to attempt to
disempower me with respect to this organization, behind the scenes,
because I was obviously crazy.... and he didn't want crazy people
around.

--WDS:
The "topic" Lomax was an expert on, was "cold fusion." Lomax stated he
spent a substantial fraction of his personal wealth to build a cold
fusion reactor.
He planned to sell it and make big bucks.

I "basically crazily" tried to "disempower" Lomax because of that, yes.
He ran for a high position in the CES.
Lomax also wrote pages and pages and pages all about how cold fusion was great,
he was an expert in it, I wasn't, I needed a "wider notion of 'truth',
and on and on.
Sort of like now. Poor innocent fellow was so "abused."

[Abd ul-Rahman Lomax]

This is complete nonsense. Richie made a single incorrect assumption,
as to the total number of voters in the election. (That's a strong
inference from the voter percentages he reported, which correctly
calculate from this single error (it was the 5 PM total voter figure,
and we have all been mistaken about the signifcance of this,
including Warren, and including myself, I just realized another error
in interpretation from what was written earlier today. Again, we have
*all* been wrong.)

Then Richie made a small math error. That error was of no serious
significance, which could be why he didn't catch it. I.e., it was
*approximately correct,* given his original error. Basically *one mistake.*

This is mostly stupid. Warren is desperately trying to prove that
Richie is a liar, morally despicable, not merely misinformed. He is
attempting to present what Richie did (which is pretty obvious) in a
way that takes one significant error -- using an early total for the
total number of voters -- and one insignificant error, a small error
in an arithmetic problem of no importance, the result was still
*approximately* correct, given his first error -- in such as way as
to make Richie look really bad.

And it's not working. Warren is just fouling his own air.

We have been laboring under *another error*. We assumed that the
voting totals given on the CSUSG web site would apply to the
Executive election. Warren made this mistake earlier today, using a
"turnout figure" to attempt to calculate the total number of voters,
and I corrected it in a post on that, but I still had the idea that
the 5 PM totals would be the number of voters in the Executive
election at that time.

No. Those numbers are the total number of voters who voted in *any*
of the *three* elections, almost certainly.

http://cusg.colorado.edu/content/vote-count-spring-2013

There were fewer voters in the CSUSG Executive election than we have
been assuming, because some of the voter who had voted did not vote
in that election. They voted in the other two elections only. That's
nearly certain to be so. People skip specific elections. For whatever reason.

Essentially, bottom line, we don't know how many dual approvals there
were in the Executive election because we plain don't know the number
of voters. We only have the vaguest of ideas, with lots of room for
major variations.

Richie erred, that's clear, but his statement wasn't actually false,
because he'd qualified it with "appears." Or "seems," I forget. Yes,
it seemed that way to him. I see utterly no reason to doubt that.

I've seen this many times. The more certain we are that we are right
and others are wrong, or *worse*, are liars, the more likely we are
to become making lots of mistakes ourselves. Our certainty blinds us.

[Abd ul-Rahman Lomax]

It's been quite obvious lately. I write a report on a topic, I spend
half a day writing it, and you then pop in with a one-liner that is
... befogged.

There is no obligation o read my writing. But those who do read me
are following me in my learning process, and tend to get smart.

It's not that I'm smart, it's that I know how to learn.

Your choice.

Clay, did it ever dawn on you to ask why I was the only one directly
elected to the ESF steering committee several years ago? Warren was
the founder, and came in third place. You were the most active
volunteer, and came in second. Why?

I'm *happy* with CES and the way things are going, and I see
possibilities beyond what, I'm guessing, you have noticed. I also
remember who you were when we started this. Clay, I don't really want
to remind you, you have come a long way.

But you may not read this, just as Michael Ossipoff did not read what
I wrote to try to help him with the voting problem. I provided what
nobody else provided, specific links. I go the extra mile.

Since when did going the extra mile become extra?

[Abd ul-Rahman Lomax]

It's unclear, unless her answer was specifically about the Executive election.

6,686/27,457 is 24.3%

The student government page announced that turnout was 25%, and that
would be *total turnout* covering all three elections.

25% is 6,864 voters.

Warren already pointed out that the sum of votes must be 100% *or
more*. The reason is that if a voter does not vote for any of the
options, and unless write-ins are enabled (were they?), the voter has
not voted in the election. Were these paper ballots, they'd be
considered blanks.

6,686 is the sum of votes. If an offical tells me that this was the
number of voters, and because of all the errors that can be easily
made, I be wanting to gain definitive *confirmation* of *how she
knows* what she is saying.

"How do you know this figure, how was it reported?"

It is a possible figure, but unlikely unless double approvals were
locked out. We see overvotes in regular public elections, where they
are illegal votes, at higher rates than this, if they are not locked
out. So did the system lock out overvotes or not? If they did, then
this was *not* an approval election at all.

What we really want to know is substantially more information.

Now, the percentages given by Clay must be incorrect, that's visible
in a flash. Where is the error?

3354/6686 is 50.16%
3332/6686 is 49.84%

Math error on the second percentage Clay reported.

(Clay later corrected this as a typo, after Warren pointed it out.)

Unless the conversation and questions were very specific -- which
Clay has not reported -- I don't consider the answer of "no dual
approvals" definitive.

We need someone with investigative reporter training here.... It's
really necessary to realize, quickly, how many errors can possibly be
made. People make mistakes all the time, including officials. She
might be right, I wouldn't rule it out, but I don't see that the
corroborationg questions were asked.

[Abd ul-Rahman Lomax]

At 06:18 PM 5/17/2013, Clay Shentrup wrote:
>My response to Richie (his comments removed)

Your response where?

(no more orignal comments below)

>--
>You received this message because you are subscribed to the Google
>Groups "The Center for Election Science" group.
>To unsubscribe from this group and stop receiving emails from it,
>send an email to electionscien...@googlegroups.com.
>For more options, visit

><https://groups.google.com/groups/opt_out>https://groups.google.com/groups/opt_out.
>
>

[Clay Shentrup]

Your link to "results" is to the same page, rather than to the CUSG results page.

[Warren D Smith]

broken link fixed.

[Abd ul-Rahman Lomax]

><http://rangevoting.org/RichieOnApproval.html>http://rangevoting.org/RichieOnApproval.html

Mature audiences only. Good technical information, but it's framed
with what is likely a lie, and it is polemic, not exactly neutral
information. It *contains* many useful points.

Warren (apparently) wrote:

>Rob Richie, the head of FairVote.org, on 8 May 2013, posted the
>following criticism of
><http://rangevoting.org/Approval.html>approval voting on the web as
>a comment to a video about approval voting. Richie's comment was
>directed to the judges of a contest and aimed to stop the video from
>winning that contest's multi-$1000 money prize. We give Richie's
>entire criticism, verbatim, in full (with no attempt to correct his
>spelling or grammar errors), with green background. We have numbered
>his paragraphs for reference.

There is no link to the original page yet, so that readers can read
the original *in context.* The comment was not, in tone or content,
directed to the judges, who are not judging voting systems but
videos. The comment, as can be seen, praises the video, referring to
it as a "successful demonstration" of the problems of plurality
voting. Richie's motive was apparently three-fold: to defend his own
political agenda and associated opinions against criticism previously
posted on the page, and to point out problems with Approval that he
has long believed were realistic and telling, and to respond to and
acknowledge a shift in approach on the part of CES, one that could
open the way to cooperation.

The page does not comment on Richie's final point, which is where a
long-term FairVote agenda and the stand of most of us with CES
converge. So I'm repeating that here:

>7. Finally, let's keep in mind that if we want broader and fairer
>representation for more of us, we have to have more elections where
>more than one person wins -- and those winners can represent
>different opinions and voters. A big focus for some reformers is
>changing winner-take-all voting rules to fair voting plans like
>these ones FairVote has drawn for Congress at http://www.fairvoting.us

Richie had begun with this:

>1. Advocates of different methods of voting for single winner
>offices have a history of rather fractious behavior, but it's
>improving. Congratulations on that front to those seeking to work
>more cooperatively.

And we congratulate Richie back, and I personally apologize for the
behavior of Warren Smith and that of, as well, Clay Shentrup. Warren
does not represent CES at all, any more, Clay is still a member of
the board. He and Warren more or less created the Board, ignoring the
prior ESF Steering Committee election. He's a tireless worker, like
yourself, Rob, and what goes around does come around. The rest of us
with CES are working to restrain Clay, and he *is* responsive. CES is
not a tight, centrally controlled organization, top down organized,
or, more accurately, not fully so. And that's appropriate in an
organization formally dedicated to "science."

Clay, and Warren, especially, are often correct on fact, though, when
they are emotionally involved, out of the history, they become as
stupid as the rest of us, and will select only the facts or alleged
facts that support their positions, neglecting or deprecating the rest.

Warren, in his critique, refers to spelling and grammar errors.
Richie's comment was informal and ad-hoc. He obviously didn't sweat
it. I saw three very minor spelling errors with practically no impact
on readability, and some places where a comma might have been an
improvement. Richie wrote clearly, as he usually does.

The "lie" is about Richie's intention, i.e., to influence the judges.
While Richie is a public actor, his goal would be to mollify damage
from what he saw as unfair criticism of IRV, and to discuss this and
criticism of Approval with advocates. He raised issues that show how
he thinks, and he was remarkably fair, compared to what I've observed
in the past. He made what amounts to a single substantial error, an
assumption that the 5 PM vote count was the final result, when *this
wasn't even about this election specifically, it was about
participation in *three* elections, * not to mention that voting
continued for three hours after that.

Given that incorrect figure and another small error, Richie was
literally correct, there were "246 students apparently voting for
both candidates s in a two-candidate election. " Key word:
"apparently." Richie did not state this as established fact. He may
have known he was making some assumptions. The small error: Richie
somehow came up with 246 instead of the "correct" figure (based on
the web site 5 PM total) of 224 votes. As Warren somewhere noted, the
error is 22 votes, which is the margin of victory. Richie got some
number mixed in his arithmetic. I like to provide links so people can
check my work, though only a few will actually do that. I make
mistakes. Often. My posts to this list are *informal*, not formally
reviewed, edited and fact-checked. But RangeVoting.org web pages
should be -- or should become -- much better than that.

Currently, Warren's link to the results was an error, he copied his
local URL into his text, not the elections page. So,
5 PM total: 6462. (For all CUSG elections.)
http://cusg.colorado.edu/content/vote-count-spring-2013
Result total: 6686 (Sum of votes in Executive election.)
http://cusg.colorado.edu/content/spring-2013-election-results-cusg

There are points of interest here:

1. What was the true overvote rate? Warren assumes that the mail from
a CUSG official is correct, when there is a very good possibility
that this official simply looked at the sum of votes, it would be a
trivially easy assumption to make. However, from our understanding
and experience with voter behavior, zero dual approvals is quite
unlikely *unless those votes were locked out by the system*. Which is
quite possible. A careful investigative reporter would not merely
want an official statement by a person, informally in an email, but
would seek to know *how the person knew the total.*

("Oh, I just added up the individual vote totals." "Oh, we got a
report of the number of voters, so, yes, in response to your question
about dual approval, I can say definitively that there were none."
"Oh, the system locked out dual approvals, because this was a
two-candidate election." "Oh, the system did not lock out dual
approvals, but none were cast.")

Note that "none were cast" is possible but unlikely. From data Warren
gives about that French election, we could expect a dual approval
rate of 0.58%, which would imply roughly 37 votes. Zero is *very
unlikely* unless such overvotes were locked out. So we had a CES
official asking naive questions and accepting a first answer naively,
instead of being surprised. A true expert would be surprised, would
immediately have asked the further questions -- or would have asked
those questions in the first place. People know how they know what
they think they know, usually. It would have been no trouble for the
official to answer. A very real possibility is that this was simply
*not an Approval* election. The Representative-at-large election obviously was.

2. At some point these questions should be asked, but, for the
future, CES inquiries, inquiries in the name of CES, should be
*authorized*. That way we don't pester people with multiple
inquiries. The ad-hoc informal procedures we have used in the past
can lead to damage as we move into a *successful* future. We aren't
in Kansas any more. The "someone is wrong in the internet" obsessions
that were common for us are incompatible with becoming a sober
*scientific* organization. Clay is not a voting systems scientist or
researcher. He's an activist, like Richie, who is also not a
scientist. Warren is theoretically a scientist, and often behaves
like one, but then he gets caught in what is *not science*.

3. Rob Richie's comment was *progress* and should not be attacked.
Following up on this comment and Clay's response, I was led to a blog
by Clay, to Tweets between Richie and Clay, and to a blog apparently
by Richie, referring to and quoting a FairVote position paper on
Approval voting, that lays out, with substantial care, a FairVote
official position on Approval voting. It's nice-looking and factually
and substantatively incoherent, which is why I spend a great deal of
time writing a deconstruction of it, in a series of posts here. We
can and should develop an official web page responding to the
arguments there, but I would also see as appropriate an approach to
FairVote, requesting them to stop promoting obsolete and incoherent
criticism of Approval, because FairVote has been expert at making
fair-seeming arguments that are highly misleading.

Richie, however, didn't do that here, not seriously. He *raised* the
issues, in fact, and didn't go far down the road of attempting to
prove his points. With his history, he thought that the overvote rate
in the UCSG election was significant, that it meant something. In
fact, regardless of the actual rate (whether it is zero or, even,
say, as high as the 3.8% Richie might have calculated from his
errors) it is largely meaningless as to what Richie actually has in
mind. Later-No-Harm failure. On the important LNH issue, Richie
actually stated the problem correctly:

>3. Let's be concrete. Among the few examples of contested elections
>with approval voting are student elections for president the past
>couple years in Dartmouth. In both of them, the winner earned the
>approval of less than than 40% of voters. Perhaps that means voters
>just didn't know much about the candidates. But perhaps it means
>many didn't want to dilute their vote for their favorite by voting
>for their second favorite also.

"Few" was misleading, and Richie should know that. He did some level
of study of Bucklin elections, which are "Instant Runoff Approval"
and theoretically vulnerable to LNH issues. But so dominated by the
hyped-up Later-No-Harm problem has he been, so focused on current and
recent elections, that I don't think this was deliberate.

What is remarkable here is that, for the first time I've seen, he
ascribes low additional approval rates to voter ignorance, *which is
the probable cause of most of it.* He then *asks the question* as to
whether or not voters "didn't want to dilute their vote."

That is the *real question,* and, we can show, Approval actually
works by testing this concern against preference strength. A voter
with higher preference strength will be less likely to add an
additional approval, than one with low preference strength, thus
causing Approval to measure preference strength, which is what is
needed to optimize overall satisfaction. Approval thus dithers toward
Range results, where voter more accurately estimate preference
strength. Approval, then, *uses* the fear of LNH to measure real
preference strength. And this points to the usage of Approval in
runoff voting and with Bucklin systems, where voters make strategic
decisions that are more nuanced.

So this *discussion*, this *conversation* should be encouraged.
Calling others "liars" is not a way to facilitate discussion. Above,
however, I state that Warren wrote what was "likely a lie." Am I not
eating my own dog food, one of my favorite Shentrupisms? What I did
there, in fact, was to use the kind of logic that Warren (and Clay)
have used. Errors that are repeated after someone has "been shown the
truth by my conclusive arguments and my blog," or the like, are
"lies." No, they could mean, and often mean, that the person has not
understood the arguments, or that the arguments are defective. Some
people do lie, but not usually when it can easily be caught. In fact,
my training suggests that we all *actually lie.* But there is a
trick: we know that what we are saying is not the truth, i.e., we
have enough information, but we manage to keep ourselves confused on
the point. So it's different than a true *wilful* lie. Essentially,
we believe our own lies, leading to a contradiction: if we believe
it, is it a lie?

Ultimately, we are responsible for what we say, that's my stand, and
someone can die because of a false statement. ("I saw him shoot
Fred.") Does it matter if we believe it if it wasn't him? It does
matter if we are prosecuting for perjury, but the guy that was
executed, does it matter to him that I could have known that it only
looked like him and I didn't have proof? And that my certainty was a
product of my prejudices about what this guy looked like?)

But crying "lies" never facilitates communication, cooperation, and
coordination. Richie, to my knowledge, has never called us "liars."
He's said lots of things that were misleading. That's the past. I see
him as reaching toward the future, and it will be a lost opportunity
to reform voting systems if we don't reach toward that as well, and
get stuck in the past. Indeed, getting stuck in the past is exactly
what we can see happened to Richie and FairVote. We all do this, but
we all, as well, can and will move beyond it, one way or another.

[Abd ul-Rahman Lomax]

Apparently Warren wants to go into this. Okay, here we go. There are
some useful lessons here.

At 10:33 PM 5/17/2013, Warren D Smith wrote:
>Lomax:
>My activity, some years back, with CES, disappeared for a time because
>Warren had become privately abusive and basically ... crazy. He was
>*certain* he was right about a topic where I'd become expert (not
>related to voting systems), and where he knew very little, but only
>some theory -- which did not apply. And then he acted to attempt to
>disempower me with respect to this organization, behind the scenes,
>because I was obviously crazy.... and he didn't want crazy people
>around.
>
>--WDS:
>The "topic" Lomax was an expert on, was "cold fusion." Lomax stated he
>spent a substantial fraction of his personal wealth to build a cold
>fusion reactor.

No. Very, very inaccurate, this isn't what I told him. I could go
back and find the emails, but this exchange was watched by Jan Kok,
who tried to moderate Warren's response, with no luck. Warren
literally became unable to understand what I was telling him, so
violent was his response. It obviously pushed some deep button.

>He planned to sell it and make big bucks.

No. Not at all, and not in any way. This is great, it shows exactly
how Warren thinks, when he's been triggered by something. I will
explain the actual situation below, but start with my "claim" --
which I could *prove* (with the emails) but won't unless it becomes
necessary -- that Warren was radically mistaken about what I'd told
him. He didn't understand it at all, and made up a story, which is
exactly what he remembers. His own story, not what actually happened.

And his story caused damage to his cause.

>I "basically crazily" tried to "disempower" Lomax because of that, yes.

Yup. Glad he acknowledges this. There is some hope for Warren,
because a true liar would deny having done any such thing. He's
merely crazy, and crazy people can, in fact, get sane. It's been said
that for those who can be honest, there is hope. For those who cannot
be honest, no hope.

Warren made the assumption that because I had "crazy ideas" -- now
widely accepted in peer-reviewed scientific journals, about which
Warren obviously knows nothing -- I was therefore not to be trusted
with any responsibility with regard to voting system activism. It's a
non-sequitur, but I can at least understand the concept. Were I
trying to attach my stand on cold fusion to voting systems, it would
be a matter of concern, and there used to be a similar concern about
another activist, who was openly racist. He dropped that, I don't
want to name him, and the past is the past. We all can move on. But can Warren?

>He ran for a high position in the CES.

Um, no, again. As I recall, at the time the cold fusion thing came
up, I was already on a 3-person steering committee, intended to
coordinate the formation of the Election Science Foundation, with
Warren and another, we had been elected in the first Asset election
in history, AFAIK, where the initial results had me in first place,
with enough votes to easily elect a second, and I'd elected Warren.
In second place was Clay, without a personal quota to win, but enough
votes to create another winner by naming a second member, and he
chose to do that. In third place was Warren.

So I never ran for "high position in CES," which didn't exist yet.
The only position I "ran for" was the steering committee, and I was
elected, with enough votes left over to name Warren. However, it
looks like Warren (and maybe someone else) agreed to privately bypass
that committee. We were never asked to approve anything. A completely
independent and less-public activity founded CES and created an
initial board. Naturally, Clay and Warren were on the new board, and
I wasn't informed about the process at all except by another. If any
of us know the history of FairVote, can they see a similarity? Except
there never had been, as far as we know, with FairVote, any formal
representative body created out of the original conference for
proportional representation, so it wasn't as blatant.

None of this is intended to deprecate CES, which is doing a great
job, and which is only a little ragged about the edges sometimes. I
knew about the sidestepped process, and simply chose not to make a
stink about it, valuing organizational unity over personal control,
and you will see that from me over and over. It's part of my stand,
and it's also part of my training as to how to accomplish
transformation in the real world, with people who do this routinely.
As one little project, gay marriage in Massachusetts was the result
of focused activity by someone trained in this way. And many other
projects. I trained with people who have, for example, Obama's ear.
People highly active and successful in politics and science and
business and the arts, and on and on.

>Lomax also wrote pages and pages and pages all about how cold fusion
>was great,
>he was an expert in it, I wasn't, I needed a "wider notion of 'truth',
>and on and on.

Indeed. This is the precise situation: Warren Smith has training in
quantum mechanics, and knows that applications of quantum mechanics,
using what seemed to be reasonable approximations, predicted that
"cold fusion" is impossible. But "cold fusion" is not a defined,
specific reaction, and all that math can only predict the probability
of a *specific reaction.* What was actually claimed in 1989 was an
"unknown nuclear reaction," not a specific one, and the researchers,
experts in their field -- electrochemistry, not nuclear physics --
had made errors in nuclear measurements. The direct evidence for a
nuclear reaction did not appear until 1991 and was not adequately
confirmed until years later.

I attempted to explain this to Warren, a thoroughly wasted effort. I
work with people who have a far deeper understanding of quantum
mechanics than he, who are using the relevant theories and
mathematics of quantum electrodynamics to attempt to understand the
mystery of "cold fusion," and they are having *some* success, but it
is *still* a mystery, many details remain to be elucidated. My
general stand is simple: it's a mystery and nobody knows how it
works, we only know the fuel and the ash, not exactly how the fuel is
converted to the ash.

Both the original researchers and I knew and understood the
theoretical predictions, already (which is all Warren knows, he
clearly has no familiarity with the vast body of experimental results
that now exist, that's what I know something about). They expected,
in fact, to find nothing, they were doing basic science, where we
*test* the predictions of theory. They were not looking for limitless
energy and naively accepting shoddy experimental evidence as proof.
They erred in the measurement of neutrons, which was outside their
expertise. Their expertise was in electrochemistry, and their
measurement technique was calorimetry, and they were world-class
experts at that, and that work was not only never shown to be
artifact, it was confirmed by expert review, and the basic findings
were *widely* confirmed. The whole cold fusion affair has been called
-- by a dedicated skeptic -- the "scientific fiasco of the century,"
and he didn't know the half of it.

From his *math*, Warren believed that I had fallen victim to
"fraudsters." He had no patience for actual evidence, none. In other
words, *he was operating outside of science and the scientific
method,* and my training in that method was directly from Richard P.
Feynman at Caltech, I spent two years sitting in those physics
lectures that later became the standard physics textbook for years,
and I heard his famous stories directly from him in a fireside chat
at Page House, about 1963.

I am *not* an expert on quantum mechanics and the math. I have
studied and know the history and much of the literature (there are
thousands of papers, over 1000 in peer-reviewed journals), and my
writing is respected and recognized by the real experts, who include
a Nobel Prize winner (other Nobelists who supported cold fusion have
died). I'm being *successful* in this field, which is a long story.
In the end, my work is toward generating scientific consensus, as a
writer. Sound familiar?

Technically, what I designed and sold one of, was a "reactor," only
because, from many reports and by its design, it was possible it
would set up a reaction. A very low-level reaction, we didn't even
bother to look for a temperature rise, and there are dozens of ways
to get this wrong. The levels of neutrons expected might have been on
the order of ten triple-tracks per detector chip (1.5 cm^2),
accumulated over about three weeks, roughly ten times background. If
we were lucky. There would be many more proton knock-on tracks, again
if we were lucky. I will look, when I do this again, look for a
temperature rise, I'll record cell temperature compared to ambient
and to calibrations, but that's primitive cheap calorimetry and I
really don't expect to see anything significant, not with the size of
the possibly active material, essentially the surface of about an
inch of 250 micron gold wire, plated with a little palladium as part
of the run, in about 12 grams of heavy water. A very small-scale
experiment. Designed to look for neutrons when the reaction is
essentially without any major neutron emissions.

>Sort of like now. Poor innocent fellow was so "abused."

Nope. Warren abused himself, with his firm belief in his own
Rightness about Everything. I've just pointed to what happened. The
"abuse" part Warren made up, like a lot of other stuff. Nothing bad
happened to me, and my life careens from one amazing and excellent
experience to another. I'm 69 this month, and it keeps getting
better. I'm hardly beginning to express this.

>-
>Warren D. Smith
>http://RangeVoting.org <-- add your endorsement (by clicking
>"endorse" as 1st step)

And he's, then, writing officially here, as the owner of
RangeVoting.org. Associating his very personal opinions with his
declared cause. I'm sure it was a very difficult step for the CES
board to remove Warren, and he certainly complained about it. He
doesn't understand why they would do that. He thinks they were wrong.
But that action was *necessary*. I'd hoped it would not come to that,
but Warren, it's easy to understand, made it necessary. We are
seeing, right here, why they did it.

CES should not link to RangeVoting.org without a disclaimer, in fact.
He's accumulated, with the cooperation of myself and others, a great
deal of highly useful information. That should be mirrored ASAP,
because people like Warren frequently go down, and often web sites
disappear, for various causes.

Now, what did actually happen with me and cold fusion?

Well, I had been editing Wikipedia and had come across in very early
2009, an abusive blacklisting of a web site, lenr-canr.org. I knew
about cold fusion from 1989, and I'd assumed, with many others, it's
a common trope, that the original findings had never been replicated.
Boy, was I wrong! But that wasn't important at first. At first I was
merely concerned with Wikipedia process, and the blacklisting was a
serious process violation, a position later confirmed by the
Arbitration Committee. I did then start acting to push the cold
fusion article, carefully, toward following policy, which was opposed
by an anti-pseudoscience faction (that was itself routinely violating
policy in the name of "truth." A truth which they made up.) Wikipedia
is *not* about truth, it is about verifiability, and supposedly,
science articles would most strongly follow peer-reviewed reviews of
a topic, strong secondary sources, independently published, not the
popular press and tertiary sources. The Wikipedia problem is that
they have excellent policy with no reliable enforcement mechanism.)

Because I'd gotten a prominent administrator trout-slapped, his
faction came after me; this faction included many adminstrators, and
what had been predicted (by a neutral and quite prominent
ex-administrator) did eventually come to pass, I was topic banned,
and banned from the entire site for three months. I knew and
accepted, from the beginning, that this could happen. It happens to
whistle-blowers, all the time.

So, I'd been putting a lot of time into Wikipedia. What to do now?
There had been some experimental work by the U.S. Navy, the SPAWAR
group, showing very low levels of neutrons from a chemical
environment, the kind of thing Warren would think is impossible. This
work had been published in a major peer-reviewed journal, but had
never been replicated. And I looked at the protocol and saw that, if
materials were bought in quantity, a kit to attempt replication could
be sold for about $100, well within the range of, say, high school
science project budgets. So a high school student could do
cutting-edge research, affordably. I found that idea inspiring. So I
had, at that point, a bit of cash. I spent about half of it, about
$6,000, buying materials and equipment as necessary. The equipment
would largely be for my own use, but for the resale of materials I
did need an accurate scale, I needed a drill press and other things,
and I needed other equipment (that high school students would have
available in high school labs) to do my own testing. I bought some
very cool stuff, in fact, the kind of toys I used to have, but that
had slipped away over the years, and now better stuff is available. I
bought a 50 MHz dual-channel 1 GHz storage oscilloscope, and I bought
a nifty digital microscope.

Most of the money invested went for expensive materials: platinum
wire, gold wire, palladium chloride, and heavy water. These are
essentially commodities, I can always resell them.

There was no desire to get rich, no expectation of that, and it's
unlikely. What I hoped to do was recoup costs, my mark-up was a
retail markup, about 50%. About half that $6000 came back immediately
as a $1000 gift and a $2000 interest-free loan from a scientist
inspired by the idea.

I've not extensively promoted this work yet and became, again,
distracted by other work, most notably getting the training I talk
about, which is about as absorbing as can be imagined. Indeed, for
most people, what it takes seems *impossible*. Can't be done! I
thought that way myself, but I was encouraged to *do it anyway.* And
that's part of the training, to do what seems impossible. We were
routinely *expected* to do this. And I did it. I completed.

So. What's been accomplished? Mostly, with cold fusion, I've
established myself as a respected voice in the cold fusion research
community. Given that most of these people are senior scientists,
many have been involved since 1989, with established careers in
science long before then, that was something.

Because of this support, I can say I'll be at the next formal
conference, ICCF-18, at the University of Missouri, with a press
pass. I will share a room with a physicist who is eager to spend time
with me. I'll not be presenting a paper at this conference, I'll be
reporting on it, for a publication that I'm *creating*. I will have
funding support from others, so it's possible I won't have to spend a
penny. There is another person in this field who started a
publication, and he was making $90,000 per year according to official
reports from his nonprofit *after expenses, that's been his salary.*
He got stuck on a particular theory, and stories he made up about
people, he developed a kind of conspiracy theory, and managed to
alienate nearly every scientist in the field. I've attempted to reach
out to him and he rejected it. He's had, in fact, some of the exact
same training I had, but ... concluded he didn't need any more and
dropped out. I could have said the same thing, but I didn't drop out,
I kept going -- and I'm still going -- and I can see exactly how he
created his own downfall. And it is a downfall, his funding has been
collapsing, and I'm being positioned -- by others -- to effectively
take that place, because of being widely trusted. I can talk with
these people. Sometimes I can even understand what they are saying,
which is saying something for the theoretical physicists. Usually it
takes time for that stuff to penetrate my skull, but I can understand
*what they are getting at* and I can understand *their approach.* I'm
not competent, with them, to criticize their math. "Langevin
equation? If you say so!"

The main thing I have done since 2009 has been to develop a
comprehensive understanding of the cascade that led to the idea that
cold fusion had been definitively rejected in 1989-1990, an
understanding of the scientific errors and, as well, the reactive
errors of the cold fusion community. I am a *skeptic,* not a
"believer," so I can sympathize with skeptics, but I also know the
evidence, so I can declare that there is a mystery, an unexplained
anomaly. I'm not a pseudoskeptic, see the Wikipedia article on
pseudoskepticism.

I've sold one cold fusion kit, and the experiment was run by a high
school student, a brilliant kid who will go far. That kit showed up,
with the kid, in a recent movie about cold fusion called "The
Believers," a truly awful title. I've sold some LR-115 nuclear track
detector film to two other customers, and I'll eventually get reports
on results. It's a totally cool material, and cheap and easy to use
and to analyze. I'm the only source for small quantities in the
United States. As this gathers momentum and people catch on to what
it can do, this alone could pay for my project. The latest sale, a
couple of weeks ago, has nothing to do with cold fusion.

The experimental result from that student project? Inconclusive.
Something happened in the etching of the detectors, and that's
unknown. I did see, on his detectors, one "triple track," which is
diagnostic of a single neutron. No evidence, so far, that this was
above background. The experiment was designed, in a unique way, to be
able to distinguish between detector background and experimental
signals, but most of the detectors were *totally ruined.* So I need
to run this experiment myself. And I haven't given it the time yet. I
will, and I'll add the bells and whistles of my own specific approach.

This is *science*. It is not designed to "prove" something. Indeed,
the whole neutron report is largely irrelevant to the *main
reaction.* It demonstrates nothing about the main reaction, even if I
were to confirm neutrons, other than that *something* unexpected is
happening. What? We don't know. Much of the evidence in the cold
fusion field has developed is *circumstantial evidence,* not direct
evidence. There is an exception, and I hinted at it above:

In the originally discovered effect, it's been found, and quite
adequately confirmed, by multple independent research groups, heat is
being produced correlated with helium, measured (as a correlation
across many experiments) at approximately the right ratio for
deuterium fusion to helium. That is *direct evidence* of a nuclear
reaction, not merely a heat anomaly. There are almost no neutrons or
high-energy gamma rays. It's a *mystery*. Someone is going to win the
Nobel Prize, if they figure this out.

See the most recent and most comprehensive review of cold fusion ever
published in a peer-reviewed journal, "Status of cold fusion (2010)"
http://lenr-canr.org/acrobat/StormsEstatusofcoa.pdf

Notice that I'm credited just before the sources. That's another
accomplishment for me in this field, I ended up being mentioned in a
journal that Einstein was published in. I consider this *way cool,*
for an amateur. Of course, Einstein also started as an amateur,
though he was much yournger.

And I did contribute to that review, with editing suggestions, but
more than that, the review was written at my suggestion. Maybe Dr.
Storms was ready to do it anyway, but that's exactly how it did come
down. There have been about 16 reviews in peer-reviewed journals and
academic publications in the last eight years or so, *all positive
about cold fusion,* but the *direct evidence* was being neglected for
some reason. I'd raised this issue with Dr. Storms and he responded
with a paper, which was submitted. The editor came back and asked him
to write a review of the whole field, which he did.

I have been having more fun than I could have imagined, even two
years ago. A piece of that has been seeing the CES video. I'm proud
of the work that our community has accomplished, and we have only
begun. Arizona implementing a two-round election system with an
Approval Voting primary, in nonpartisan elections? That, as an
imminent possibility, is years ahead of what I'd thought might be
possible in the public arena, unless massive -- and fast --
transformation had happened in NGOs. That hasn't happened yet. Watch. It will.