* *** COMMENTS_2017_ON_FEEDBACK_ON_DOCUMENT_COMMENTS TXT - 5 Aug 2017 02:46:00 - JKNAUTH Thank you very much for this morning's feedback on my 2017 election document comments. Both the email from Amanda and the phone call with Mary were very helpful, as well as two followup emails I got from them to answer some questions I had asked. Below are my comments on their feedback, hopefully in time to make a difference before things are frozen for this cycle. Although I might prefer different wording in some areas, I understand your desire to simplify even if that might let some low-probability things slip thru the cracks vs. making things more complicated to catch more problems, but maybe leading to even more things going wrong. I know you are just trying to find a happy medium. I'm satisfied with the way this review cycle has gone so far, but would have preferred getting your feedback to my comments sooner so I could respond immediately, hopefully before it was too late to fix remaining problems. As you have seen, my normal turnaround is under 24 hours. Amanda's 8/4/17 Email ======== ====== ===== PG 9: Amanda's followup email said Nicole now agrees with the Manual wording -- all three judges do the ballot bin unloading and ballot box packing, not just the CJ alone. That's exactly what I wanted to hear. Resolved. PG 13: If I could not get a good volunteer emergency tabulator transporter, my fallback suggestion for my (former?) precinct would have been for me to finish my normal CJ closing work and then transport EVERYTHING, not just the tabulator, to the Raleigh Operations Center myself, skipping the long journey from Rolesville to Knightdale. Don't all the Knightdale things end up at the Operations Center anyway? The disadvantage of that approach would seem to be that the non-communicating tabulator would not get delivered until all my poll-closing work was done, so the early vote counts for the press or whatever might be made available somewhat less quickly, but probably more quickly than if a Coordinator was having to handle all this in multiple places. I do appreciate why some sort of change was needed. Anyway, we'll do whatever you say needs to be done. Fortunately I have never had a tabulator that didn't communicate -- eventually. We always just let it keep trying. The longest it has ever taken was 20 or 30 tries, and that was even despite some hookup problems at our end. PG 37: If the expiration dates and document currency cannot be adequately provided in the state's "Acceptable ID" form, or in the Manual, then I guess the Basics and CJ classes will need to do it. It's something any potential curbside official or other PO might be faced with. But maybe since it is for ID processing, it is considered not frequent enough to complicate documentation. I think that at a bare minimum, it should be said in the Manual that an NC driver license or NC ID card must be exactly current. With all the 2015-2016 Voter ID checking complications (no longer in effect), many people may have had their minds warped to still include the Voter ID loopholes which allowed some expired licenses to be considered valid in some situations. So "current" on the form may mean more than expected to such people. For example, expired licenses for ages 70+ may show up at curbside, even for attempted HAVA IDing -- not a valid ID. Just emphasize to call the Help Line if anything about a submitted license or other form is in question. I know that is considered always to be understood by POs, but maybe it needs to be reiterated here where so much can be flaky (and a call undesirable if you have to walk back a very LONG way to get to the phone). PG 59: Mary's followup email said there are legal reasons why the abandoned ballot text cannot cover the "We can't tell for sure." case. I haven't tried to plow thru that legalese yet, but I know you have to document along the lines of what the lawyers dictate, even if seemingly incomplete. In real life if this situation occurs, we'll just have to do as spelled out in my July comments: Treat "We can't tell for sure." the same as "HAS been disturbed" despite the incomplete wording. 8/4/17 Phone Call with Mary ====== ===== ==== ==== ==== Registration Table Quick Guide (8/1/17) ------------ ----- ----- ----- * Mary will check if pollbook tabs will still be supplied as something to be inserted separately by the POs (and used by me for flagging ATVs), or if the pollbooks will be pre-tabbed by the printer (and I will then need to bring my own tabs for ATV flagging). Ballot Table Quick Guide (8/1/17) ------ ----- ----- ----- * I mentioned that having the BT people check the reason box to see if an ATV flip is needed requires the HT people to be sure a reason box has been checked. Mary said this would be stressed in HT class. All Officials Quick Guide (7/28/17) --- --------- ----- ----- * Mary said the detailed sign placement information may now actually be on the signs themselves somehow. I said as CJ I still would need a checklist I could use at the end of setup to walk around to see if the signs had been placed correctly. Sometimes I have found they weren't. If the placement information is only on the signs (how?), I might miss things. She said she would see if there was still a checklist, a la the old AO QG. Voter Assistance Quick Guide (7/31/17) ----- ---------- ----- ----- * See Manual page 59 comment above. Chief Judge Pocket Guide (8/1/17) ----- ----- ------ ----- Page 2: Still need to fix Blue Reconciliation Bag vs. Chief Judge Bag error. Page 11: I still believe the "If the Emergency Bin was used during the day:" line is misplaced as detailed in my 7/26/17 comments. It was correct in 2016. Page 11: I still would like "NOT the Tabulator" to be added after "Insert the red key into the AutoMARK" as detailed in my 7/26/17 comments. Help Table Quick Guide (8/1/17) ---- ----- ----- ----- 1) I mentioned that in classes and by past coordinators we had been told to call the coordinator to do our Monday setup call-out phone test ("to get our phone number put in their cellphones"). In response to my asking why this was never put in the QG although I always comment on it, Mary said you now wanted to minimize calls to the coordinators -- we always call at the end of setup anyway. She said she would have the trainers no longer ask for this setup call-out test to the coordinator; instead we would call someone else. I guess a coordinator could still ask for the early call and we would honor that request as well as doing the end-of-setup call to them. We had just been doing what the trainers/coordinators had asked us to do and it seemed bad for the HT QG not to specify it. 2) As mentioned in the BT QG comments it is now more important than ever for the HT people to ensure a reason code is marked on an ATV before the voter is sent by them to the BT. I suggested having a FC #1 top left column note for this. Instead, to avoid making FC #1 more complex, Mary said this would be emphasized in HT class. 3) I again noted the inconsistency (sometimes present, sometimes not) of having the item "Review Section C and ensure it is complete and signed" in a block. I think it should also be in row C and Row F (without the "and signed" since you nixed my idea of filling out C completely for a name change) as it is in row E (first action block in each row). Mary said she would check again. [Mary, is this good enough row/block specification for future correspondence?] 4) I pointed out a bug in this draft of FC #1 that had not been there in the prior draft. In row D, action block 2, "Compare the address on label in Section C to ... Section A. should be "Compare the address on label in Section A to ... Section C. 5) Checking of Last-Minute Absentee Voter List (e.g., in FC #2, block 3) Mary gave me her rationale for why checking the Last-Minute Absentee Voter List is not needed in the normal FC #2 flow. I hope I have recreated her logic below, but want to check if I missed something or read between the lines too much. I will also stress some points of my own. [Buckle down for this. Sorry, but I have spent a good part of the last month working on math and programming stuff. I'm now booted into the mode for lots of Cases, If/Thens, etc.] Assumption: All voters on the Last-Minute Absentee Voter List (called "the last-minute list" below) should eventually be marked with pink As in the pollbooks. Preferably this should be done Monday night. Less preferably do it Tuesday morning before the polls open. Even less preferably, if not by "polls open" then do it right after an early call from the BOE, but definitely not after 10 AM on Tuesday. Unfortunately many people will have voted by mid-morning. Suggestion: If making a lot of these calls Tuesday morning is a problem for the BOE (I don't know if it is) and that causes a delay in dispersing the lists, have some way for the CJs who have already gotten the information from the website to update a file (a la the Monday Setup Times file) on the website so the BOE would not have to call them Tuesday. Then the BOE could focus on those precincts that really needed the calls. Of course this might leave open a way for some bad guy to enter bad data into such a file if this is a non-secured file. I'm curious if you have a good feel for how many CJs use a cell phone Monday night to get their Last-Minute Absentee Voter List data (the current CJ PG implication) and then handwrite the lists vs. how many use their home PCs and printers vs. how many require the Tuesday morning call and handwrite the lists while keeping the BOE on the phone. That would seem to be important information. I would certainly push the PC/printer technique for speed, accuracy, and minimizing tying up BOE resources on Tuesday. ---------------------------- Here we go! ---------------------------- Let's call someone a FRAUD voter who is on the last-minute list (so has already voted) and then tries to sneak into a precinct to vote a second time. If he somehow ends up at the HT without an ATV (FC #2), you now would now check only the laptop (no longer also the last-minute list) and see he has NOT been marked "ALREADY VOTED", even though he has voted. A) If the FRAUD voter IS NOT registered in your precinct per the laptop and insists on voting anyway, he must vote provisionally. Then the post-election checks will find he is a FRAUD: tried to vote twice. The Sheriff will call. The DA will thank the FRAUD for documenting his crime so carefully. Jail. Etc. Since that out-of-precinct voter would not be on this precinct's last-minute list, checking it wouldn't have helped catch the crime any earlier. That one was easy. B) If the FRAUD voter IS registered in your precinct per the laptop and shows up at the HT without an ATV, it means likely the RT people goofed and missed finding the entry in the pollbook. (Or the printer goofed -- I saw that happen for two voters a couple of elections ago.) Since the laptop said the voter was registered in this precinct, the HT then sends the voter and CJ back to the RT. Note that checking the last-minute list would have indicated an upcoming problem, since the voter has in fact voted already, but that discovery is now deferred, maybe now long deferred. Back at the RT, if the voter's ATV still cannot be found, call the Help Line. They will detect the FRAUD since they have the integrated voting status, including last-minute absentees. It's now their problem to decide how to proceed. If the ATV is found at the RT, 1) If the now-found ATV has a pink A, go back to the HT. Now the HT should check the last-minute list. a) If the voter is not on the last-minute list, the pink-A marker goofed and this voter in fact is NOT a FRAUD. The CJ should be notified, the ATV unpinked, and the voter allowed to proceed normally as if the pink A had not been marked in error. Of course the RT needs to then check if there were other problems for that voter. Also the CJ and RT need to review all the pink A markings to see if there were other errors. Probably there is an ATV that should have a pink A, but doesn't because its pink A was put on the wrong ATV; maybe there are other problems, too. Note if the HT had not checked the last-minute list in 1) above to catch the erroneous pink A and had just called the Help Line, presumably the Help people would have seen the error (voter has NOT voted) and told the HT to "Look at your last-minute list and then have your CJ fix the buggy pink As! Call back if there is still a problem." b) If the voter is on the last-minute list, the pink A was correct and the FRAUD voter has been detected (or there was actually a BOE database problem). Call the Help Line. It's now their problem to decide how to proceed. 2) If the now-found ATV does not have a pink A for the FRAUD voter, either the pink A marker has goofed or the marking was not done on time. Note that FRAUDs are probably willing to get up at dawn and try to vote early, beating the BOE call or any late pollbook marking by POs. That's why I much prefer that CJs mark the pollbooks Monday night using the website file. So the RT will allow this FRAUD voter to vote, but the FRAUD should be detected and handled later per the process in the assumption below. This hopefully rare case was caused by a combination of bad/late pink Aing and the HT not early checking the last-minute absentee list (which maybe doesn't exist yet) way up at the first paragraph. Of course in the case of bad/late pink Aing, the FRAUD voter could have gone directly to the RT, gotten a clean ATV, and voted. There would be no lack of HT last-minute absentee list checking at fault since the HT was never involved with such a voter. Needless to say, timely and accurate pink A marking is very important to curtail FRAUD voting. Assumption: The BOE can always eventually detect a FRAUD voter and cancel the absentee ballot, even without any HT checking of the list. So, at worst, the FRAUD voter actually votes only once, then goes to jail, if this was not some BOE database error. Since the last-minute list checking has been removed from FC #2 and the Manual (last draft I saw) does not cover the above use of the list, the HT class could say something, but I believe HT people will likely end up completely ignoring the last-minute absentee list if it isn't clearly and simply stressed when it should be used. There may be be more Help Line calls for the non-normal appearing situations, or there may be more FRAUD voters who slip thru, but who will be caught by the post-election checks and handled appropriately. If the above flow is accurate, the tradeoff seems to be between detecting some FRAUDs earlier by always checking the list at the HT in FC #2 vs. having HT people possibly ignore the list completely and detecting the FRAUDs later in the BOE post-election checks and/or causing more Help Line calls. I can see pros/cons for either early checking (FC #2) or not. I didn't think the FC #2 checking was very time consuming, but Mary said she had seen some precincts where it was. My problem was always more with reminding HT people to do the check since it kept being included and then dropped from the flowchart without explanation. ======================================================================== For reference, here are the links to my comments on last month's drafts: I never got a Laptop Quick Guide to review. Comments on the Manual and the RT, BT, HT, VA, and AO QGs http://jgkhome.name/WakeBOE/Comments_on_7-20-17_Package_of_Document_Drafts.txt Comments on the Provisional Envelope and related Manual pages http://jgkhome.name/WakeBOE/Comments_on_Provisional_Envelope_for_11-2017.txt Comments on the CJ PG http://jgkhome.name/WakeBOE/Comments_on_2017_Chief_Judge_Pocket_Guide.txt