Need ideas, logistics, costs for publicly run hcpb election

[GuvWurld]

Imagine you are going to talk to one of the Supervisors of your County and you wanted to ask her not to buy Diebold machines and to vocally oppose their certification. Suppose she says ok, but if we don't buy Diebold, what should we buy? When you tell her the case for publicly run elections using hand counted paper ballots, she is surprisingly receptive. She asks you for a proposal explaining how it would work and what it would cost.

I don't expect anyone to actually write a full proposal, but what are the things to think about to move this scenario forward?

[onthebench]

1. Think north of the border

Tell her to call the Ontario Board of Elections. In Canada, their provincial elections are cast using hand counted ballots. In fact each election has a new registration as well. In Canada a provincial election can be a random event (see no confidence voting) and they have as short as 28 days and as long as 56 days to actaully have the election. The blind can even vote on paper using a simple tape recording and a template that fits over the ballot.

Talk to Ontario Board of Elections.

[GuvWurld]

2. Thanks onthebench and welcome to DU!

Thanks for the lead. That sounds like it could be a fruitful source of relevant info.

I'm betting some of the usuals on this board have a whole rap on this question and can run it down.

And speaking of no confidence voting, are you familiar with the Voter Confidence Resolution?

[onthebench]

3. Thanks for the welcome

Their website is http://www.electionsontario.on.ca/en/home_en.shtml?nocache=true

The problem is that the rap is falling on deaf ears. I suggest that a new approach needs to be made. The whole procedure for voting needs to be revamped. The voting machines are really a small part of the problem. We have too much money invloved in elections. Chicago is notorious for election fraud issues and they have been using paper ballots the whole time. I think it will come down to having a real constitutional convention in order to fix the problems.

I just have come here to vent. I was in the elections industry in the technical end but I hob nobbed with Secretaries of State and even once testified in Congress. Now goody two shoes Bob Ney is a target of corruption and he was the co-author of HAVA. Sheesh.

[GuvWurld]

4. You are speaking my language

I am all for a new approach, no amount of change is too much, I will see you at the convention! Seriously, I think on that scale too. See Blueprint For Peaceful Revolution.

[Neil B Forzod]

5. excellent points

I enjoy reading your posts!

To nitpick a bit though, I see the Canadian example used a lot in discussions of hand-counted paper ballots and I don't think it's an especially good analogy. Canadian elections (provincial or federal) typically consist of a single vote-for-one race with 4-8 candidates and no special rules or districting issues or reporting requirements. Canada has no requirements for party-declaration by the voter, so there are no primaries and everyone in a given riding gets the exact same ballot. Canada also has only 10% of the voters that the US does, so there's a lot less votes to count. They do vote on paper and count it by hand, but it's a comparatively easy task.

US elections are typically a lot more complicated, as we know. At the very least there's usually a lot more races to count. That's not so bad by itself, but then you couple it with races where you vote for more than one candidate and add some associated jurisdiction-specific rules for under or over-voting. Sprinkle in some crazy reporting requirements and even crazier voting procedures. In states with variations of straight-party voting it's almost impossible to manually apply the straight-party rules and count the ballots by hand. States like NY have even crazier ballots with candidates endorsed by multiple parties, which introduces a whole new set of counting issues. States like PA combine a form of straight-party voting with cross-endorsement of candidates to get a system that's almost incomprehensible to normal folks. As long as the election system is as complicated as it is now, hand-counted paper ballots just won't work in a whole bunch of states (although it could work in some states that are a little more sane and a little less populated).

I agree with you that voting machines represent only a small part of the overall problem. Not an especially popular viewpoint in ERD, I suspect. But I tend to think there are plenty of bigger issues that overshadow the voting machines themselves. I do like some aspects of the Canadian system, especially the ability to force an election on non-confidence issues and (more especially) the stricter campaigning rules and the fact that the whole election campaign is only 6 weeks long. Adoption of some of those aspects of the Canadian system might help things a lot more than adoption of their (decidedly different) hand-counted paper ballot.

Neil

[Febble]

6. As a Brit

one thing that has shocked me as I have got to know the multitude of ways in which your democracy is broken, is the sheer laxity on the part of election officials regarding to the conduct of elections.

The attitude that recounts are a bore and need to be avoided; the attitude that voters are out to cheat and need to be discouraged; that it's not worth sealing a ballot box if there is going to be a recount; that it's worth making your random handcount "match" to avoid handcounting the rest.

In the UK ballots really are treated as valuable objects - we hand count them, but at every stage between the moment the ballot is dropped in the ballot box to when it is counted or recounted, the process is scrutinised by bipartisan volunteers who take their duties seriously. A single lost ballot is considered a serious matter. Whereas "close enough for government work" seems to be the culture in at least parts of America (and notably Ohio).

So no, I don't think machines are the root of the problem. They are a symptom, if you like, of a culture that regards a rough approximation as good enough. It's the same culture that denies ex-felons the right to vote (not the case in the UK). It means the principle of one person one vote has been eroded (if it was ever fully established).

Sorry, more venting. The reason I feel even entitled to a view is that a) America is the most powerful nation on earth, and so the leader you elect matters to everyone and b) your nation claims to be a democracy, and seeks to spread "democracy" to other nations. And yet your own democracy is broken.

[eomer]

7. It is more than mere laxity. It is criminal conspiracy.

The Ohio recount illegalities cannot have been caused by mere laxity in each of the counties. If they were then you would see variation from one county to the next in how that laxity expressed itself. The fact that almost all counties used exactly the same tactics to avoid a full hand count is proof that this was a coordinated criminal conspiracy.


edit: grammar

[Febble]

8. I would probably agree

but criminality is easier where there is a culture of laxity. Judy Alter's San Diego election didn't seem criminal but it certainly seemed lax. But I would agree with you about Ohio from what I've read.

I suppose what I was trying to say is that to get to the root of the problem there needs to be a whole new culture in which votes are actually taken seriously by those who are responsible for processing them.

[eomer]

9. Agree: "but criminality is easier where there is a culture of laxity"

Yes, I agree. We need to do two things. One is to preferably bring to justice but at least throw out the criminals who have infiltrated the process. This would include both government officials and vendors.

The second thing is what you say about a whole new culture.


edit: wording

[Febble]

10. Absolutely

I really wish your Democrats would run with this. Isn't it why they are called Democrats? It's got nothing to do with "sore losers" and everything to do with civil rights. And democracy. It should be a vote winner in every sense.

[OnTheOtherHand]

11. are you referring to the precinct selection?

A somewhat arbitrary selection from the recount observer reports:

http://www.iwantmyvote.com/recount/ohio_reports/counties/morgan.php

"I was an observer at the recount for Morgan County on Wednesday. (I was representing Libertarians.)

"The B of E (Board of Elections) ladies said they chose the precinct to be hand counted, because it was just over three percent. It was not randomly chosen.

"...(I)t is my opinion, and that of the other observers I asked, that the B of E in Morgan County had conducted a fair election, and did their best, as they understood the rules, to accomodate (sic) the recount process and the observation."

Here's an excerpt that is almost odd:

http://www.iwantmyvote.com/recount/ohio_reports/counties/warren.php

"In general, the recount was conducted in a cordial and cooperative atmosphere, despite obvious resentment by some about the recount, resistance by many involved about the prospect of a full hand recount, and opinions by some that this was a waste of time and taxpayers' money.

"The witnesses and coordinators of the Green and Democratic Parties agree that the events of Tuesday were a valued expression of democracy. They went part of the way toward restoring our faith in the transparency of Warren County's election process, which was clouded by the lock-down of the vote count on November 2nd."

Based on the bits I've read (I've looked at maybe 20 counties' reports), I am closer to Febble's view than to yours. I think kiwi_expat recently read many or all of the observer reports -- dunno what she thinks.

BTW I'm not at all asserting that the observer reports rule out a criminal conspiracy, and I can't imagine how they could. I just don't feel the force of the argument that only criminal conspiracy could explain multiple BoEs' failure to use a Kish table, or whatever you consider the logical implementation of random selection. Or I may be missing your argument entirely.

(EDIT: Well, I seem to have misjudged Febble's view, but I will await clarification on that. More importantly, I forgot to make an obvious point: a permissive reason why so few counties did truly random precinct selection is that the secretary of state didn't order them to. That is neither coincidence nor "conspiracy." Whether there was conspiracy on top of that, I don't know. But I would feel a bit scuzzy to see dozens of people accused of felonious conspiracy without at least expressing honest doubts.)

[eomer]

12. Yes, that is what I'm referring to but I would put it more generally.

The tactic that was seen in almost every county was to make sure that the 3% machine count matched the 3% hand count.

This goal was accomplished through the use of a few different techniques in different counties but the overall goal was the same everywhere (with only one or two exceptions if I recall correctly).

The specific techniques included:

• non-random selection of precincts
  
• multiple reruns of either the hand count or the machine count until they could be forced to match
  
• advance selection of precincts
  
• advance handling of ballots such as pre-counting and pre-sorting
  
• physical tampering with ballots (stickers)
  
• communication of hand count results to those performing machine counts for no innocent reason that is apparent
  
• communication of preferred precincts and cheat sheet totals from vendors to election workers
  

My argument depends on there being not just "multiple BOEs" that followed the approach because that could be explained by mere coincidence. It is the fact that almost every county followed this same approach that, to me, is proof that it was an intentional and planned approach rather than the result of laxity.

It's been a while since I have looked at the recount observer reports but in the past I have read every one of them multiple times. When I can find the time I will try to make a list of which counties took extraordinary (that is, illegal) measures to force the 3% counts to match and which did not. If it is, say, all but one then I believe that is pretty compelling proof that it was a planned and coordinated approach.

[kiwi_expat]

14. The Ohio counties' recount procedures varied wildly.

I don't think the lack of random precinct selection was dictated by Blackwell, but he obviously did not insist on random selection. The number of different ways the precincts were selected is truly mind boggling. Some counties selected recount precincts from those precincts with less than 3% of the voters. Some used the precincts with the lowest number of voters. Some selected recount precincts from only those precincts that had MORE than 3% of the voters. And so on. My personal favourite is the county (Lake) that let the OBSERVERS select the precincts.

But, precinct selection aside, the goal in all of the counties was to avoid a full manual recount. And I'm sure that Blackwell planted that seed. The common explanation given was that a full recount would be too expensive and would not begin to be covered by the $10 per precinct (paid by the Greens-Libertarians) dictated by Ohio law.

No doubt Blackwell wanted to avoid a serious discrepancy with the official vote count. The count had already been certified by the time the recounts were begun. Blackwell probably knew that some vote fraud had occurred, whether or not he had orchestrated it himself. He had certainly orchestrated various forms of voter suppression - but of course, voter suppression could not have shown up in the recounts.

[OnTheOtherHand]

19. OK, see my response to Febble

I think "extraordinary" is not synonymous with "illegal" in my mind. Doug Jones suggests (as I recall) that many basically honest election officials take shortcuts -- which may well be illegal -- to avoid what they regard as pointless busy work.

I'm not quite sure how I would define "extraordinary." But I might be more impressed by 5 examples of "extraordinary" (to indicate a conspiracy -- for lesser purposes, of course one might do) than by 50 examples of "illegal."

I swear I am not trying to bust your semantic chops here. The underlying issue is whether the behavior of election officials seems bizarre under a benign interpretation. Certainly that should take account of patterns. Some of the reports kinda cry out for follow-up -- what's with those Clermont ballots, anyway?

[Febble]

13. clarification coming up

I find it hard to believe that any honest SOS for elections could have been so paranoid about voter fraud as Blackwell seems to have been, and yet so laid-back about the possibility of vote-count fraud. Actually, impossible to believe, given that that set of attitudes was bound to result in a net benefit to the campaign of which he, scandalously, was chairman.

So that is why I am inclined to interpret the messing about with 80 gram paper, provisional ballots, challenge rules etc on the one hand and the failure to ensure any kind of adherence to the rules that are essential to a valid recount on the other, as prima facie evidence of intent to swing things his party's way. Which is the sense in which I agree with eomer.

But I think he presided over a BoE culture in which officials seem to think that their first duty was to economise on time spent on taxpayers' money, rather than to ensure that all the votes were properly counted. And that probably suited his purposes rather well.

Same goes for Franklin County and the machine provision; the story may well have been that machines were allocated on the basis of 2000 turnout in order to save money. But the fact is that the result was to ensure that precincts with low turnout in 2000, which were overwhelming Democratic, had the longest queues. It maybe that everyone was too stupid to see what the consequences of the policy would be, but it may also have been that it was a neat way of capping the Kerry vote in a way that could be attributed to good housekeeping on the part of the BoE.

(edited to make more sense....)

[OnTheOtherHand]

18. OK -- the last thing on my mind was to exonerate Blackwell

I think it's open to interpretation just how far Blackwell would go to deliver the election to Bush, but I can't imagine anyone arguing straight-faced that he was just 'calling it like he saw it.'

On the recounts, I don't think that Blackwell had much influence over the BoE culture in this particular respect (although he certainly had the authority to mandate random precinct selection, and chose not to use it). It seems to me that if I am a basically honest Ohio county BoE official presiding over a recount, (1) if there are big discrepancies, I want to know about them, and (2) if there are trivial discrepancies that are likely due to 'random glitches' (e.g. op-scan ballots that aren't quite properly marked), I surely don't want them to trigger a full manual recount that will break my budget, demoralize my staff, and (apparently) serve no useful purpose whatsoever.

This mindset creates opportunities for fraud. Needless to say, a basically dishonest (or perhaps merely craven) mindset creates further opportunities for fraud.

We do can at least attempt to assess whether BoE actions were consistent with a desire to distinguish between trivial error and massive discrepancies. For instance, from the Washington County report:

http://www.iwantmyvote.com/recount/ohio_reports/counties/washington.php

"The first 3 ballots we saw in Aurelius precinct had the oval for write in candidate filled and without a candidate name written AND the oval for Bush filled in. The write in oval on each one had a white oval sticker covering the mark so that they were counted for Bush rather than as overvotes. The 4th ballot had an "x" on the Bush oval which had been marked over to make it show up.

I objected to the altering of ballots and we were told that this was the policy of the board to correct the ballots so that they would count correctly in the machine.(...)

(...)I repeated my objection to fixing the ballots, but agreed to go forward with the machine count. Since they refused to stop fixing the ballots, the hand count and machine counts were going to match no matter what precinct was counted in the opinion of the two observers that were participating.(...)

I want to note that all the re-marking and band-aiding of ballots that I observed DID reflect the will of the voter.

So far, I haven't been able to agree with people in other threads who see evidence of 'ballot tampering' during the recount as proof of fraudulent intent. (The Clermont account is not inconsistent with at least some fraudulent tampering, although innocent explanations also seem credible.)

I think the Franklin County example is textbook: yes, the voting machine allocation scheme perfectly well could have been a dumb mistake, a deliberate ploy, or both (deliberate by the Reps, dumb of the Dems not to catch on). Backing up a step, the circumstance that makes this disaster possible is the rationing of expensive voting machines as a scarce resource, and that too can be perceived as dumb, deliberate, or both. It's a little hard to sort out (isn't it?) the extent to which the problems in the British health care system, or in U.S. Superfund cleanup, or in any number of other government programs are intended versus unintended.

Obviously we want an election system that doesn't leave us wondering whether officials are sloppy or crooked.

[Febble]

20. Well, I suppose my point

(and it's not a million miles away from GuvWorld's OP) is that to a Brit, the whole system looks as though its designed more to save bother than to ensure one-person-one-vote.

So to try and address GuvWorld's point directly:

what needs thinking about is the cost-benefit ratio of the whole damn system, bearing in mind that the aim of the exercise is to elect a government, not to make elections less troublesome.

Paper ballots are dirt cheap, which is one of their huge advantages - no rationing is necessary.

In the UK they are counted by hand usually by bank tellers who are experienced in counting valuable pieces of paper (but I think you told me that in the US money is also counted by machine....) So all you need to do is to provide paper and pencil and a box. And with the money you save on all the other stuff, you hire some temporary peoplepower to count like crazy. It's really not that hard, except for the complexity of your ballots.

How important is it that all races are on the same paper? In the UK if there is more than one race, you have a separate ballot for each. It does mean you can't do interesting stats on who voted for what combination, but it does mean that you simply multiply the counting effort by the number of races. The less high profile races are usually counted after the Big One.

Anyway, just some thoughts from abroad....

[OnTheOtherHand]

21. whoops, yeah, I sorta bumbled into a thread hijack!

Sorry, Guv, just lost track of my surroundings.

The history is complicated by various half-assed reform efforts, but yeah, certainly the institutions aren't precision-engineered to ensure verifiable voting!

I think the advantage of having all races on the same paper isn't doing "interesting stats on who voted for what combination" (which is certainly not the purpose of a ballot! although it makes certain forensics possible -- as, of course, they are not in New York with our lovely lever machines). It minimizes the paper to track, which seems desirable considering how much trouble we have tracking our ballots already. (The paper-tracking is an issue during voting as well as after. Under current arrangements, as you know, a single polling place could include a number of different ballots, and handing someone the one right ballot may be easier than handing her the right stack of ballots.) It may also help reduce undervotes, but that could go either way -- I don't know if there is research on that. Dunno, I'll try to learn more.

Has anyone floated a cost estimate yet, either here or on BeFree's thread?

[OnTheOtherHand]

22. Ellen Theisen has a cost estimate, but I'm not sure how to use it

http://www.ballotintegrity.org/EM2004Implemention.pdf

Proposal was to have the 2004 federal races (president, House, and Senate where applicable) done on handcounted paper ballots. Cost estimate for the entire country was about $53 million, including $38 million for 200,000 polling places, each with an average 4 people working 4 hours at $12/hour.

To put that figure into some perspective, the Franklin County (Ohio) BoE noted that in order to match its 2000 machine/registered voter ratio in 2004, it would have had to spend an additional $3.16 million -- or about $8 million to attain a ratio of 200:1. So the economics certainly seem pretty favorable (even allowing for amortization of the capital investment).

Don't feel I have yet come close to finding the whole range of cost estimates.

[Febble]

23. A piece by the BBC

on Sunderland, the constituency that likes to be first with a result in election night:

http://news.bbc.co.uk/1/hi/uk_politics/vote_2005/frontpage/4515173.stm

I am not 100% certain that the vote counters are actually paid. This piece gives the going rates for poll clerks and others, but not the counters, who it describes as being "chosen". Certainly the scrutineers are voluntary (my mother used to do it for the Liberal Democrats).

BTW, the word "teller" is used in two senses in this piece - as in bank tellers, but also it is used to describe party volunteers who tick people off on a list of those who have said they will vote for that party. I used to be a "runner" taking the lists back to Party HQ, so we could round up those who hadn't voted yet. We don't register by party in the UK. That idea always sounds completely bizarre to a Brit.

(edited for clarity and error)

[Neil B Forzod]

25. those estimates seem a bit off

Ballot printing costs typically run closer to $0.25 to $0.50 per ballot, not the $0.10 that's estimated. Although I suppose you might be able to get away with cheaper printing for a special hand-counted ballot as described in her paper -- high precision (required by OS ballots), heavier stock, and short runs of multiple ballot styles probably aren't required for her example.

But even at her projected numbers, that's still $53 million to hand-count just three jurisdiction-wide races in a single election. Multiply the numbers to accommodate the other races on the ballot, and then multiply that times the number of elections that take place in a typical election year. It actually adds up very quickly, which is why election officials are so keen to buy machines to do the work.

Neil

[OnTheOtherHand]

26. yeah, that's what I had in mind about not being sure how to use it

Two to three federal races are just the tip of the vote-counting iceberg.

Unfortunately, I was too sleepy to point out the corollary problem: one can't very well (or usefully) balance the cost of a hand count against the cost of machines if one ends up buying the machines anyway! Of course, Thiesen wasn't trying to argue that hand counts would be cheaper than machine counts.

Personally, I have been leaning toward op-scan with mandatory partial recounts (and some other procedural enhancements), at least for most jurisdictions. But I don't know enough about the cost estimates to feel entitled to much of an opinion, so this thread is useful.

[kiwi_expat]

27. Why not DREs with voter-verified receipts and mandatory partial recounts?

Why would touch-screen equipment with voter-verified receipts be inherently inferior to op-scan?

Isn't "mandatory partial recounts" the key?

[Neil B Forzod]

35. I think that works too

Any voter-verified, hand-recountable record of the vote works as long as a statistically significant audit/recount of those records is performed (i.e. as long as enough of those records are audited/recounted to demonstrate with some specified high level of statistical certainty that the results reported by the voting machines accurately reflect the selections made by the voters on those machines).

So a touch-screen with a voter-verified paper record of each voter's selections works for me, as long as that paper record is audited as described above. So does a paper-based optical-scan system. A lever machine doesn't, on the other hand, since there's no record (paper or otherwise) of individual ballots cast, just a total on an odometer.

Neil

[OnTheOtherHand]

36. actually, I think it is a tough question

I agree that mandatory partial recounts are crucial either way. I don't object to the "idea" of DREs, but they certainly pose their own issues.

DREs may improve accessibility (or at any rate the sense of a level playing field -- I think Neil was elaborating on this point elsewhere), and potentially can (even should) reduce undervotes, compared with op-scan. DREs make it trivial to change mistaken votes, although the trick is to make sure it isn't trivial to mistakenly (or fraudulently) change intended votes. In principle they can reduce costs, although there is at least controversy about whether that is or ever will be true in practice.

From the standpoint of verification, if we are really worried about individual voters verifying their ballots, well, I think the research indicates that it is hard enough to get voters to look at the summary screen and do something if it is wrong, much less to check a paper receipt as well. That really bugs me. Yes, my ATM gives me a paper receipt, but if I were supposed to study forty different things on it to see if they were correct, with people waiting behind me in line, I just don't think it would happen. Tell me that I really should study the receipt because it is legally authoritative, and I feel like keeping my money under the mattress. (That is somewhat irrational, but hey, people are irrational.) Moreover, my impression is that a lot of the early implementations aren't very robust or voter-friendly. A DRE-plus-paper-trail approach also seems to lend itself to distributed denial of service attacks, where folks wander into crucial polling places and insist that the voting machine has switched their votes.

And of course I haven't enumerated nightmare stories about DREs, some of which are hyped, but which still raise real doubts.

[Ellipsis]

55. On ballot cost

Curious... between $.25 and $.50 If you have the cost break out percentages or access to those stats.

Give consideration to just Presidential, Senate and Congressional races being hand counted.

Also give consideration to Printing on Site at the precincts using toner based printers with the appropriate duty cycle.

A toner based solution with paper raw costs comes in around 3 cents per page...

[onthebench]

29. See my cost estimate for one county in illinois

here http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=403008&mesg_id=403411

one caveat - congress can only make federal voting laws for how their members are elected. Each state has the right to create procedures for their own elections. Who pays the state costs? Fed will not because they can not tie the money to specific procedures because that law would never pass. That is why HAVA includes voting using punch cards as allowable.

[kster]

33. I guess we are chumps,because

with electronic voting we are allowing our country to get taken over, With out a shot being fired. How smart is that? Count the damn ballots by hand, no matter what it takes.

[Neil B Forzod]

34. never going to happen (nt)

[onthebench]

38. It is great to feel that way

I am just trying to give you all the heads up with what the competition (meaning DRE vendors) will use. That is why I advocate changing the entire process. The counting part is really a small part of the process. In the coming days I hope to get deeper than just the cheerleading talking point level. I have worked on elections from start to finish. I have read most of the state laws of the US for elections. Some states have a procedure that is literally three pages and some that are 1000 pages. NJ's voting machine page was just one page. There is something wrong when there is that much difference among states. We have to start with what an election is really supposed to be. We have to make it manditory that everyone over 18 is involved. We have to train the 16-17 yr olds before they are 18 with more than making them memorize what offices are up for grabs. We have to train them to look at the issues and look for information about the candidates and make informed decisions.

This topic, "Election Reform" should be more than just the case for paper ballots. I would even say that a voice vote is the best method. Use a confessional type booth. If you do not want your voice to be recognized then use an audio computer voice device to disguise your voice. Have audio observers listen and each record the votes for all to see. These are the ideas swirling around in my head.

Hope I have time to get them all out before they take over.

Peace

[kster]

39. Create petition

drives to get the question of how the votes will get counted, hand vs computer, on the ballot especially at the local and state levels. (Probably be wise to initiate legal action preventing the use of computers in the election this ballot question appears.)

[OnTheOtherHand]

40. right, we are talking about state and local decisions (see OP)

-- although I happened to grab some numbers about federal elections as one starting point. (Somewhere nearby I just posted some numbers that look more like yours.)

As we know, some U.S. jurisdisctions use hand-counted paper now, and it works for them (at least subjectively). Of course, there is little evidence of a groundswell in that direction, despite the efforts of some activists. I am trying to get a better handle on the nitty-gritty, something more nuanced than a talking point that "PBHC only works in small jurisdictions" or "PBHC will save democracy." I appreciate your efforts to fill in some of the details.

[GuvWurld]

15. Please don't hijack this thread

onthebench answered the OP but all these other posts are off-topic, as far as I'm concerned. There is nothing wrong with that content, other than most of it is tired and re-hashed, but let me provide more context to consider as a starting point for this exercise. The Voter Confidence Resolution, mentioned up thread, contains the following election reform platform. This is not something for debate here and now either. It has already been adopted in Arcata, CA and serves as the "internal agenda" for the Voter Confidence Committee, the local group that will lead this charge through the aforementioned County Supervisor. These things are not carved in stone but do make a skeleton of a new election system:

1) voting processes owned and operated entirely in the public domain, and
2) clean money laws to keep all corporate funds out of campaign financing, and
3) a voter verified paper ballot for every vote cast and additional uniform standards determined by a non-partisan nationally recognized commission, and
4) declaring election day a national holiday, and
5) counting all votes publicly and locally in the presence of citizen witnesses and credentialed members of the media, and
6) equal time provisions to be restored by the media along with a measurable increase in local, public control of the airwaves, and
7) presidential debates containing a minimum of three candidates, run by a non-partisan commission comprised of representatives of publicly owned media outlets, and
8) preferential voting and proportional representation to replace the winner-take-all system for federal elections;

Obviously we can't implement a national holiday here on the local level. Likewise, some of these other points may evolve through further local debate that would necessarily be part of this process.

So, to restate the OP, and with these guidelines in mind, what are the nitty gritty things I'd have to get down on paper and get costs for in order to show the Supervisor I am serious, this is feasible, and she should engage in developing this further?

Thanks, and no offense intended to anyone posting above.

[Febble]

16. As well as Canada

look at the UK, and Australia.

Our ballots are simpler - referendums are rare, and there is usually only one race at a time, though not always. We sometimes have local elections and EU elections on the same day, in which case we use separate ballots for each race, printed on different coloured paper.

[eomer]

17. No problem.

No offense taken, even despite the "tired and re-hashed" poke.

I was thinking the same thing, that this should really be its own thread, and will start one if I have anything more on this tangent (unless someone else does so before me).

Peace.

[emlev]

24. Have you seen this thread in GDPolitics about San Francisco?

I haven't read it all yet (nor this thread of yours, for that matter, Guv, it's that kind of day, well--month, well---year...)

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=2284191&mesg_id=2284191

[emlev]

31. Turns out that thread contained misinformation

given by a city government staffer (who was well-intentioned) and has been deleted.

[kster]

28. kick.nt

[kster]

30. To study the very inexpensive,

transparent, and public, same day hand counting of ballots in countries like Canada, France, Israel, India, Switzerland, Britain, and many other countries. See

http://www.ecotalk.org/VotingMachines-LegalVoting.htm

http://www.blackboxvoting.org/

[Ellipsis]

32. It's the local races ...

which causes the complexity of the ballot.

Make the ballot a booklet... sorta.

National race ... One page, state race one page, county race one page, local race one page. If it's all paper the complexity boils down to the printing and fulfillment and compilation of getting correct printing to the correct locations... and when applicable unique identifiers (numbering system) of each ballot. With current technology, printing could be handled at each voting location with printers on site... or... submit RFB's (Request for Bids) to traditional local printers..... If Unique numbering of ballots are necessary... solutions are available to work with PDF's (Adobe's portable document format).

[OnTheOtherHand]

37. unique numbering is not too hard; counting seems hard

(When I saw PDF I thought you were gonna say PDF417, a two-dimensional bar code, which is one way to identify and to track individual ballots. But never mind that; I just think they are cool.)

But if we are looking at 30 or 40 or 50 races to be counted, there is gonna be a lot of paper to hand-count (if that is what we are committed to), no matter how we divide the races among ballots. Of course if we decide to mandate a hand-count of federal races but to allow discretion on other races, the problem is more tractable.

[Ellipsis]

41. Agreed, It's not.

http://www.intellipdf.com

As far as counting from what I've read from your posts we differ... to me... the more people who count the votes the better.

As to how... Let each county to figure out for themselves.

We need to get away from the "instant gratification" of immediate returns. What's the hurry?

[OnTheOtherHand]

42. it's not that I'm opposed to having people involved

Although I do worry that if you have too many people involved in a particular precinct, you move from a healthy situation where everyone can keep tabs on each other, to a chaotic situation where someone might get away with something bad. But let's not kid ourselves that the status quo is secure, eh? Generally speaking, the more people involved, the better -- government is supposed to be of and by the people, and that means us, I guess.

I also agree that counting quickly isn't a paramount concern as long as the ballots are secure.

Where to get the people? If they are volunteers, I think that many BoEs would already like to get more volunteers than they do. If they are paid, then BoEs who aren't actually convinced that hand-counting is an ethical imperative will have to see if they can live with the cost. (Or not, depending on what you have in mind. As we've said elsewhere, the federal government can impose rules for federal races; it can't very well do so for other elections.) -> So, sure, we can let counties figure out for themselves, but if we are doing this for all races, then what they figure out may be that they prefer op-scan.

Sorry if I am talking past you; just bear with me!

[Ellipsis]

43. You are not un"bear"able.

I have said this many times before... The children are the answer.

I would invite high schools to incorporate within curriculum active involvement in community and elections.

...that is, unless ya think we can't trust our children.

I know your a hard sell I don't necessarily disagree with with your postures... someone has to have it. I do wonder with your obsession (I paraphrase) or constant need for the glass to be half empty.

Opscan technology when it comes to accuracy (without human intervention) is WITHOUT A DOUBT A NON ISSUE ...even with write ins.

Open to potential exploitation? You betcha... as with any binary solution. There are are others here with more qualifications who can speak to the specifics of that then some poor simple farmer like myself.

[OnTheOtherHand]

44. well, stepping back...

Here we are on a thread responding to GuvWurld's request for practical logistical advice to counties who are considering paper ballots, hand counted. I admit that I didn't remember, or at any rate didn't apply, your previous comments about high school students to this thread.

Umm, without retreating into armchair navel-gazing (it's a bit chilly to be pulling up my shirt), I think I am not a "glass half empty" person, but rather a "must do good in minute particulars" person. When the county BoE chair says to you, "Actually, hand counts sound pretty good, but I'm not sure how we would make them work," you want to be able to say a bit more than "Oh, I betcha can figure it out -- and how about getting high school students involved?" I'm not saying it won't work, I'm just saying the job isn't done!

Also I am a fox rather than a hedgehog; I don't generally believe in One Big Idea that answers the important questions, but rather lots of little ideas and realities that tend to bump into each other. I think you are probably a fox too -- I doubt that hedgehogs (of this sort) survive very long as farmers. (EDIT: transposed words, haven't had coffee yet)

"glass half empty" is generally taken to connote pessimism; I seem to be a lot more optimistic about American democracy (and other things) than many folks here. But the object isn't to be optimistic or pessimistic, it is to build things that work for people.

OK, enough of that.

I'm a bit leery about sending high school students to precincts unsupervised, yes -- or rather, I should say, I'm a bit leery about having the count depend on that. If we need supervision, how many supervisors do we need? What proportion of high school students ends up doing this... whatever exactly it is? It needs to be sorted out somewhat.

If you are saying at the end of the post that with appropriate safeguards, opscan is quite good, then I agree. That's a whole different thread. This is a hand-count thread.

[Ellipsis]

46. just a little fishing expedition.

;-)

What you just said about yourself was important to establish for a multitude of reasons. Thank you sincerely.


WHO counts isn't the point... there would be a large pool of honest citizens to draw from... more important is establishing the proper ratio of counters to potential ballots cast with consideration to double and triple checking votes. I have yet to read Febble's post so perhaps she has or can provide incite to this... I can say that if a DRE can only process 120 votes in a given 12 hour day (fogerox)... This makes hand counting much more attractive...

You... you fox you. *pinch*

[OnTheOtherHand]

48. yes, "how many" precedes "who"

(Feel free to fish, by the way -- 'slong as it isn't ice fishing with a saw, all is well.)

A running theme has been that 'those other countries have fewer races,' so I guess we really should find one of the _U.S._ locations with hand counts and get some info there....

Yes, a big problem with DREs is that they tend to lead to rationing, and thus vote suppression. It's a comparative plus for either op-scan or hand-counted paper ballots. (I think someone here recently made the point that precinct-based op-scan can also create a big problem if people have to wait in long lines to scan their ballots after completing them.) Not that I pretend to expertise on the tech stuff.

[Ellipsis]

52. I think if one defines the hierarchy the complexity subsides.

oooh big words...

I mean if we establish boundaries and responsibilities at each level (treating this ballot now as a document, assigning a chapter each) ie. National, state, county ...etc we exponentially decrease the complexity of the ballot... if... each level is responsible for it's own page or pages within the ballot...

Copeesch (sp)

[OnTheOtherHand]

53. run it by Neil, and/or others who do more with election admin

If people are voting on multi-page ballots -- and not all the same multi-page ballots at a polling place (at least that is true now) -- then that seems pretty complex no matter who is responsible for what. But I do see your point that decentralizing production of parts of the ballots might be a net win, at least for hand-count ballots that don't have to meet stringent production standards. Dunno.

[Ellipsis]

57. Navels can be quite inspiring. n/t

[Febble]

45. Well, in the UK

we have bipartisan volunteers to scrutinise the count, which is done mostly by bank tellers, who I think do it voluntarily, but I am trying to find out. The scrutineers are definitely voluntary. Usually, what happens physically, is that the count takes place in a large hall (school gym, town hall) and the counting tables are behind a rope barrier. Scrutineers, bank tellers, and, I think, candidates, can go inside the barrier; members of the public stay behind. But it's only a rope, and I don't think there always is one. It's just to contain the chaos. We are literally counting pieces of paper here, and we don't want people barging into the trestle tables, and knocking over the piles.

But the point is that there are always lots of people milling about within a short distance of the counting tables, so it is a pretty transparent business, and added to that, ALL the counting places have television cameras and a reporting crew.

How you would graft that kind of system on to the American structure I have no idea, but it really isn't that hard. I'm increasingly thinking that it's your complex ballots that are the problem. We rarely have more than one or two races, and referendums are extremely rare (only on constitutional questions).

[Ellipsis]

47. Hello Febble...

Do you now the ratio of counters compared to ballots cast? I understand one need take into account complexity of the candidates and referendum etc...


:hi:

[Neil B Forzod]

49. it's the complex ballots

The complex ballots are a big part of the problem. In some jurisdictions you can get well upwards of 50 races and issues on a single ballot, and that adds a lot of work to any manual count.

If it were just the additional races, it wouldn't be so bad -- just more effort to count. Unfortunately the problems caused by the length of the ballot are compounded by party registration issues: voters are generally only allowed to vote on some races, depending on their declared party registration. So you often end up with different ballots for different voters at the same polling location, depending on what they declare as their party, and that complicates both counting and reporting. (Incidentally, this is a big part of what makes the election management so expensive and complicated: election supervisors have to do all the ballot setup and layout for a bunch of different ballot styles, figure out how many of each type of ballot they need at each polling location, then factor in things like different language considerations to ensure that voters can get a Spanish or Chinese or other multilingual ballot upon request. In some places, like WA, there can quite literally be thousands of different ballot styles -- before considering language -- to prepare for each election... the actual design, procurement and management of the ballots alone is a logistical nightmare).

On top of all of that, different states have different rules for counting ballots. Random example: some states, like North Carolina (which has been in the news here lately), use "straight-party" voting. So at the top of the ballot you pick a party -- let's say "Democrat" :) -- and the result is that democrats are implicitly voted in subsequent races whether you mark them or not. Easy enough (but also easy to accidentally overlook or miscount by hand). But easy enough. Except that voters can override the implicit selection by explicitly selecting a candidate -- say a Green party candidate -- in a second race controlled by that original straight-party vote. Still easy enough. Except maybe that second race is a vote-for-three race, and you've now got a single explicit vote for a Green candidate and an implicit vote for the Democrat candidate. So the race has one candidate marked, but two votes, and is still undervoted. How is that race counted? Unfortunately, it depends entirely on where you live. In some states straight-party rules dictate that the explicit Green vote completely overrides the implicit Democrat vote. In other states, the explicit and implicit votes are combined and both are counted. In some states, it may be an error to explicitly vote in a race that's been implicitly voted by a straight-party race. Better example: make the race a vote for one instead. Who gets the vote? In some states it's the Green candidate, because the explicit vote takes precedence. In other states it's actually an overvote, and neither gets the vote (which would probably come as a surprise to the voter, who probably didn't know any better). In others it might be the Democrat, because explicit votes are counted except when they'd cause an overvote. Unless they're write-in votes (writein votes throw a wrench into the whole thing -- because they're sometimes prioritized higher, or lower than "regular" votes but only as it pertains to the application of implicit straight-party votes that would otherwise result in an overvote.) There are other variations on the theme as well, but you get the idea.

I could easily go on... sometimes you get a mix of races on a ballot where some are controlled by the straight-party race and some aren't. Or there are multiple straight-party races on the ballot, each controlling a subset of other races on the ballot. Or states like NY, which allow for cross-endorsement of candidates (i.e. Joe Smith is listed on the ballot twice, once endorsed as Democrat and once as Green) -- and states that allow cross-endorsement also typically use straight-party voting. So re-read the previous paragraph, but consider one or more cross-endorsed candidates on the ballot and figure out which one gets the vote (bearing in mind that each endorsement of a candidate typically needs to be dealt with separately until the very end). I haven't even touched on preference races, and the problems of split-ticket ballots. In Washington state they have a mandatory preference race rule, which means you have to vote in the preference race or your ballot won't count... and once you've voted in the preference race you have to be careful not to vote at all in the wrong subsequent races (races not applicable to your preference selection) because doing so will create a split ticket situation and you've just invalidated your entire ballot. I also haven't touched on how primary elections differ fundamentally from general elections. Or the multitude of local processing rules that vary from state to state. You surely get the idea though.

By the time you factor everything in, you can't (in most states) reasonably expect a bunch of volunteers to count the ballots by hand: it's far too complicated, far too error-prone, and it's far too labor-intensive. The "paper ballots now! hand counts now!" crowd simply doesn't appreciate that it's basically impossible to hand-count paper ballots in half the states in the nation.

The sheer overwhelming complexity of the American electoral system is what differentiates it from places like the UK and Canada, where it's both feasible and economical to count the ballots by hand. (It's also the entire reason why there's a multi-billion dollar election equipment industry in America, where Canada and the UK mostly don't have one at all.) Anyone who seriously makes the claim "we should just do it like Canada!" or "let's do it like the Brits!" is really just speaking out of pure ignorance (at least until the electoral system is radically reformed, at any rate).

Amusing side note: I suspect most of the people yelling "let's be like Canada!" would probably be surprised to find out Canadians don't even get to vote for their own prime minister (except for the ones who live in his home riding, anyhow). I assume it's the same in the UK (although mildly ashamed to suddenly realize mid-sentence that I don't actually know for sure :D ).

Neil

[Febble]

50. Yes, you are right about the UK

it is virtually identical to Canada, give or take a hereditary peer or two.

I did a primer here if you are interested:

http://www.dailykos.com/storyonly/2005/3/20/6453/39287

But you shoudn't be shocked about the PM thing. At least it is easier for us to get rid of our PMs mid term if they mess up. Tony Blair is proving hard to shift, but give us a few more months.

[Ellipsis]

51. truly naive am I

Good information all....
absorbing

thank you.

[BeFree]

54. Of course it can be too complex

But what makes anyone think a distant programmer can be successful at figuring it all out and writing a program that counts it all just right?

And then how do you run checks and balances on the program to insure its accuracy?

The simple solution is hand counts. If it forces us to simplify the ballots - so be it. Simple is good. Our elections need to be good - so lets make them simple.

Your compilation of the idiocy of the current election processes is an indictment of the situation, and one which demands a 'radicalization', eh?

Well then lets do it. Let's radicalize the system into an ever more simpler, accountable, and efficient process. And what process could fit that bill more-so than paper ballots counted by hand?

[Ellipsis]

56. A couple of things...

First, This is just an exercise right... in an ideal world where people understand we want hand counts because we are afraid of manipulation?

Whether that manipulation be lack placement of DRE's causing long lines

...or a DRE which for some reason becomes "confused" and Defaults to an unwanted candidate cleverly assigned as default

....whatever ...

At every potential opportunity for fraud...we should be practical make adjustments taking into consideration the 5% of the population that will do ANYTHING to win.

It's not like we are deciding the future of the free world or anything being the only Superpower left on the planet...Sorry China ya ain't there... yet.

:sarcasm:

All I would like to see is a system where I know my vote will be counted PERIOD.

Three different pieces of paper. President, Senate, and Congress Color code them if you like.... Red. White. Blue.

Use Technology for the rest.

Segue:

I do think instead of looking at this from the top down we look at it from the bottom up as far as vote reporting on hand counts.(I know huh?) Leave the people at the local level to figure out whats best for them.... Just tell them to count the votes and let us know when your finished. No time constraints.

Count the votes THREE times compare the numbers, denote any inconsistencies, average the count and move on. I would think since would significantly lower inaccuracy.

On close races Go back to the inconsistencies and recount.

Given everything you said to the complexity of the different ballots it would be wise to implement STANDARD DESIGN SPECS. as a RULE for companies who wish to design equipment used in future elections.

Also, existing ballots should not be designed by election admins.... but professionals, typographers and designers.

...would'nt want to have another butterfly ballot fiasco like Florida.