Hi! I'm in Alaska. The recount here is going well. There is so far no "dirt" as referred to in the last post.
We had some conflicting reports about this previously, but apparently, according to the explanation given in responses to questions as the recount was getting underway, the vendor does none of our actual election programming. We have a state employee who was hired through a competitive process, not a political appointee, who designs the ballots and programs our Diebold Accu-vote memory cards. He has been present at the recount, troubleshooting and available to answer observers' questions.
Diebold supplies the machines but, we are told, has no access at any point to our server or our memory cards. The elections people are adamant about this being true and seem proud of their independence. Alaska also prints its own ballots.
We have a bi-partisan statewide bi-partisan review board which signs off on each memory card after running it through tests they devise, and then regional bi-partisan boards review the cards again in the machines that will actually take them. There are seals and vaults and an effort to make sure that nobody else has access to the card or machines.
The elections division chief by law resolves questions about ballots that the machine can't count. She does this in view of the people operating the machine as well as observers, using guidelines that seem fairly cut and dried. The voters for example are instructed to fill out the ovals. A few of them make circles around the candidate's name instead, which are not detectable by the machines. Cards with circles but no filled in ovals are not counted, by law, which is alarming when you first see such a card rejected, because the voter's intent seems clear. (Cards with a circle AND a filled-in oval ARE counted, as are cards with an X in the oval but not completely filled in.)
I was told that in a very close contest, a ballot for example which has been rejected but has the name circle could be contested, and come before a judge, who would be able to rule on voter intent without being limited to using just the state guidelines. In an election which is not terribly close, this sort of thing theoretically should come out in the wash if you assume that some voters from each party are going to ignore the instructions, which are clear.
The recount involves reprogrammed cards being run through different machines. Hand counted precincts are being counted by at least three people at a table, and they can't all be from the same party. If the vote is off by even one, they are asked to count again.
Some people still have some suspicions and are worried that we may not be looking in the right places. At any rate, we have turned up nothing anonomalous as of yet. It is hard to watch everything and to know what to watch, as we are novice observers, most of us. There are 20 machines going, and a separate room for hand counting.
The 10% of hand counted precincts were randomly selected (by us, from a hat), one from each of the 40 districts and one from each region, making 44 (the elections people came up with this scheme and I like it; it means that no major regions are somehow missed. They also agreed to exclude the 35% of previously hand counted districts from the hat, which means in effect that we got 15% of all machine counted districts hand counted, very good statistically.)
Normally their procedure is to do recounts entirely by machine; the elections division accommodated our request for hand counts however. Their goal, as stated to me, is to be open and try to allay suspicions that way. This also has something to do with our decision to be very polite with them and avoid making them feel defensive. I would recommend this kind of dialogue to elections people (and to the people challenging them), across the country. (Of course, if you really ARE dealing with corrupt elections divisions or with ego-invested people who know deep at heart, for example, that the touchscreen computers they purchased without thinking all this through really were not the best idea, it's a whole different ballgame.)
As I've mentioned before, this PROCESS was put in place by our last lieutenant governor, Fran Ulmer,who ran the department as she does everything - thoughtfully and meticulously. We were one of two states to get an A+ from Common Cause for the way we ran our election in 2000. She was defeated in the 2002 governor's race by Frank Murkowski, and her successor as lieutenant governor, Loren Leman, sure hasn't done as well in the public relations department (if you want to know more about this, go to Juneauempire.com, click on archives, click on Oct. 24 on the calendar, click on "Opinion" and look at their editorial.)
However, his appointed head of the elections division, Laura Glasier, is doing fine, in my book. For one thing, she responded to complaints about our new touchscreen computers (for handicapped people) by mothballing them, at least for now. In spite of buying them, she didn't invest her ego in them, like some elections officials in other states apparently did. I've talked to her fairly extensively now and she strikes me as having integrity.
I speak as someone who is not convinced by any means that the 2004 presidential election was won fair and square, let alone the 2000 election. I talk to people - I talked to people in Florida. Sometimes you find out things...sometimes you don't. Juneau is not Tallahassee, that's the long and the short of it!!
I was concerned that Alaska would be a tempting target because of its Senate race being so close, and wondered where the weaknesses in our system were. I never suspected our elections officials or candidates themselves of foul play. But like many of you, I don't trust the vendor, its connections to one candidate, or its proprietary ways. However, as mentioned, their role in Alaska seems to be limited to repairing machines and sending software updates. As I understand it, the underlying software tells the machines how to detect marks. But our local programmer tells the machines how to interpret those marks - as a Knowles vote or a Murkowski vote, for example.
Other states can learn from this. To be as independent as possible, not to accept Diebold or other vendors programming their cards - that would be one step toward allaying citizen suspicions. And these optical scan systems are clearly better, at least when used intelligently - there's no reason to adopt touchscreen computers. They're easier to mess with, even when supplied with paper trails. There are other solutions for handicapped people. Here's one, for example:
http://biz.yahoo.com/bw/041101/15164_1.html .
So, to sum up: this is an educational process and worth going through even if we find nothing major wrong. It hatched a group called Alaskans for Fair Elections, and it may be our enviable job in Alaska to be watchdogs trying to avoid any kind of gradual corruption of an apparently decent process, as opposed to struggling to adopt a decent process. One of our objectives is to identify any problem spots in the process and to talk to them about it. I'd like for them for example to adopt a regulation requiring a random hand recount of a statistically significant number of precincts.
But it will be surprising if we find pools of 561 absentee ballots that weren't counted, as in Kings County. Absentee ballots are examined in an open process with observers present, and voters whose ballots are excluded for any reason are supposed to be notified.
Anyway, tell me if you can think of a way anyone could hack this system as described! So far, so good?????????
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
