How to optimize the vote count on election day

September 8th was the first election I worked at, not even as an observer, but as a whole PWG (member of the election commission with casting vote). The elections at my precinct were held using KOIBs — such wonderful machines that make throwing difficult (you won’t put a ballot in it with a packet, and observers won’t give one by one) and give a result printed on a piece of paper at the click of a button. Excellent, in theory, thing. But, as everyone who monitored the election results knows, it took almost the whole night after the election to calculate the results from polling stations equipped with KOIBs — that is, as in the good old days with manual recounts. That is why it happened and what I can do about it and I want to tell.


Churo The miracle of selective technologies
The principle of operation of the KOIBs is very simple: it reads the data from the ballots and uploads it to a special flash drive ( article about software in KOIBs ). Before the polling station opens at 8 am on election day, it is tested whether it considers the ballots entered correctly, after closing ballots from portable ballot boxes from a home ballot are additionally downloaded, control ratios are checked: the number of ballots received by the PEC, spoiled, entered, the number of voters who voted, canceled ballots . If the numbers do not converge by more than 3 ballots, this means a manual recount of the results. Then, through the connected printer, KOIB prints the results in this form:



On the evening of September 8, those who watched the election results live had a lot of nervousness: hour after hour passed, but there was no information. It would seem: everything is automatic - so what prevents you from quickly calculating the data? Many seriously expected that by 22 o’clock everything would be known. Naive people.

In fact, no automation in the current system of vote counting and did not smell


The situation with KOIBs is very typical of state innovations and is most easily described by the example of an accountant with Excel, which recounts all the numbers on the accounts. To begin with, the presence of KOIB does not cancel the heaps of manual work: the number of voters who have voted in the books is recounted, the remaining ballots are recounted and then quenched. Say, in my precinct it was 1,900 voters in books + about 900 unused ballots remaining after the completion of voting. Error for one voter or one ballot - we start all over again. However, I got into a fairly experienced team, so we managed all the papers in an hour and a half, and already at half-past nine (half an hour before that time, when everything had supposedly already been calculated), they opened the COIB.

Then another hour or two filling in the paper, packing the canceled and counted ballots, but this is not so interesting. The coolest thing begins further: the flash drive that was in KOIB is taken from there and personally taken by the chairman of the PEC to the territorial election commission (TEC), accompanied by a policeman whose responsibility is to bring this flash drive safe and sound. Then a couple more hours of the queue at the TEC in the office, where the data is clogged up already at the GAS Elections, the acceptance of all the papers by the higher commission - in general, you can finish by two in the morning.

Where the weak point of the system, from this description, I think, is understandable. We get three stages, three weak points of the system:

  1. manual filling of voter books
  2. manual delivery of flash drives in TEC
  3. manual data entry into the GAS Elections.

In addition to unreasonable time and labor costs, this system also has serious security holes, the most important of which is manual data entry from KOIB in TIK. Many observers travel with PEC chairpersons to personally control data entry - but there are not enough observers, observers can be removed, and indeed it is a dubious pleasure to go to the TECs after a 15-hour working day.

What i suggest

Replace voters' paper books with electronic ones. Give each member of the commission on a netbook with a connected tablet, where he will score passport data, and voters - sign with a stylus. Publish the filling of books online, with the ability to verify participation in the vote by name and passport number.


This is what the voters book looks like now. On average, there are about 400 voters in each

  • speeding up the procedure, minimizing errors: minus half an hour to calculate the number of votes cast manually
  • transparency: each citizen will be able to check according to passport data whether his voice or the voice of his close relatives is taken into account - to avoid the situation when grandmother was sitting at home and her vote was counted at the polling station.
  • safety: only the fact of voting is taken into account, the secrecy of the vote remains untouched.

Equip COIBs with a terminal for sending data to the TEC. As an option - in order not to redo everything from scratch - to supplement the system with special terminals, into which data from KOIB will be interrupted by hands - an ordinary tablet is suitable. After entering the data, they are confirmed by the members of the commission with the casting vote (now we sign on the printout), the members of the commission with the right of deliberative vote (PSG) and observers can leave their comments. Signature technology can be any:

  1. a stylus on the screen,
  2. entering a special password,
  3. flash drive with digital signature (the most confused version and probably the one that will probably choose the CEC, hehe)

After the protocol is signed by all members of the commission, the data is instantly sent to the TEC and published on the electoral commission website.


Voting card in the TEC. Isn't it better to have this plate online?

Benefits:

  • Simplification of the system - no driving with flash drives, manual input
  • Acceleration of counting - the data will be available online, for recounting and processing by everyone as soon as they are entered - that is, by about 10 p.m. when manually recounting the voting lists or an hour earlier if you have electronic books
  • Increased security - the risk of loss or damage of the flash drive disappears, the risk of data fraud after delivery of the flash drive to the PEC disappears. Observers can be sure that in the territorial, city and central election commission, exactly the data obtained are calculated on the site.

Further, of course, it will still be necessary to carry the canceled and voted ballots to the TEC, sign for all the formalities. During this time, the KOIBs will remain on (as is happening now - they will be turned off only when they accepted all documents at the TEC and gave permission to turn them off: in case something does not converge and a recount is required).


The superpower of our state is to create lines anywhere. In the photo: PEC chairmen and secretaries stomp in the TEC corridor, waiting in line to enter the received data

However, all changes in the results (cancellation, sending for recount) will be recorded online, as well as by any third-party recount services connected via API. To increase transparency, you can also configure sending notifications of any changes to observers - but this is already a reinsurance method, as the chances of canceling the results after the protocol has already been published are frankly miserable. Now all the fears are connected with the fact that only a few people see the results of a particular section before entering data into the CEO.

Advantages of the described scheme:

  1. The physical voting process is maintained using paper ballots. This is important both in terms of transparency and security - maintaining the secrecy of the vote.
  2. Both procedures - replacing voters 'paper books with electronic and online sending of data to higher election commissions - can be carried out separately from each other (say, save paper voters' books, but automate the process of sending data to the TEC)
  3. Simplicity of embodiment. Ideally, you need to equip the COIBs themselves with data sending terminals. But to reduce the cost of the process, you can simply give tablets to the chairmen of the commission, in which they will hammer data with their hands - it won’t take much longer, but you won’t need to change the COIB.

The counting of votes and the implementation of all formal procedures until late at night is not only a problem for those who monitor and participate in elections, but also hard work for observers and commission members. The current system of vote counting is a consequence of the exclusively inertness of our state. Change it is not difficult at all - it would be a desire.