4 ms·

This is similar to what Nate Silver put out earlier (as mentioned in the article). The difference in voting to each candidate by county and voting machine type

by pfooti 10y ago

This is similar to what Nate Silver put out earlier (as mentioned in the article). The difference in voting to each candidate by county and voting machine type is completely explained by the population distribution in those counties. In Wisconsin, the counties that used paper ballots tended to be less white and better educated. That's why Clinton got more votes there, probably. It is unlikely that there is real widespread tampering. That said, there is still good reason to audit all national elections. They're quite important, after all. The mad dash to say who won really quickly does a disservice to the process, IMO.

grzm 10y ago
there is still good reason to audit all national elections. They're quite important, after all. The mad dash to say who won really quickly does a disservice to the process, IMO. Agreed. Maybe it's late, but I'm leaning towards thinking auditing should be closer to the norm than otherwise. Kinda like testing? Get the system so smooth and clean that everything is obvious and trustworthy. Edit: Okay, it is getting late. I just starting thinking of what an election Chaos Monkey would look like.
- theparanoid 10y ago
  State elections do go to automatic recount in 20 states - http://www.ncsl.org/research/elections-and-campaigns/automatic-recount-thresholds.aspx http://www.ncsl.org/research/elections-and-campaigns/automat...
- tehwalrus 10y ago
  Donald Trump is an election chaos monkey.
- Klathmon 10y ago
  Nobody else mentioned it, but an election "chaos monkey" would be interesting. From the few second I thought about it, you could have some "approved troublemakers" that are VERY well controlled and licensed to go and do things like attempt to tamper with machines, multi-vote, ballot stuff, or something similar. If it's done in a way that they know exactly how many extras they put in (how i'm not sure, but it'd need to be bullet proof) they could basically "audit" election systems. Currently, there is no legal way to "test" the system as far as I know.
  - grzm 10y ago
    Interesting idea, isn't it? I suspect that many of the failure modes you'd want to test would require a full re-do of the election, at least at the level of the precinct. I don't know how tolerant the population would be. Though I encourage you to follow through and see what you come up with :) I'd be interested to read about it.
    - Klathmon 10y ago
      I'm not so sure it would need to be that drastic. If I get a "stamp of approval" to tamper with a voting machine, I could get to the point of being able to tamper, but not do anything. Or I could (if it was safe to do so) have it add 30,000 votes to a false candidate. It would be similar to how "white hat" hackers need to be able to show a POC but not actually do any harm. Hell even having "false elections" could be a useful first step. It wouldn't be perfect, but it could at least start figuring out where the weakness are.
      - grzm 10y ago
        Hell even having "false elections" could be a useful first step. That, I think, would be potentially useful.
        nkurz 10y ago
        Here's a 2012 paper from a team at the University of Michigan detailing a "mock election" gone wrong: Title. Attacking the Washington, D.C. Internet Voting System Abstract. In 2010, Washington, D.C. developed an Internet voting pilot project that was intended to allow overseas absentee voters to cast their ballots using a website. Prior to deploying the system in the general election, the District held a unique public trial: a mock election during which anyone was invited to test the system or attempt to compromise its security. This paper describes our experience participating in this trial. Within 48 hours of the system going live, we had gained near-complete control of the election server. We successfully changed every vote and revealed almost every secret ballot. Election officials did not detect our intrusion for nearly two business days—and might have remained unaware for far longer had we not deliberately left a prominent clue. This case study—the first (to our knowledge) to analyze the security of a government Internet voting system from the perspective of an attacker in a realistic pre-election deployment—attempts to illuminate the practical challenges of securing online voting as practiced today by a growing number of jurisdictions. https://jhalderm.com/pub/papers/dcvoting-fc12.pdf https://jhalderm.com/pub/papers/dcvoting-fc12.pdf
zw123456 10y ago
I agree completely, this was an extremely important election with a lot of future implications and it, and really all elections, should be meticulously audited. I think there is a lot of pressure to come up with a quick result when the emphasis should be more on accuracy.
nkurz 10y ago
In Wisconsin, the counties that used paper ballots tended to be less white and better educated. The linked article says that all counties in Wisconsin use paper ballots, and thus the claim that there is a difference between counties that use paper ballots and other counties "cannot be true". Are you disputing this, or did you perhaps mean to say Pennsylvania instead of Wisconsin?
- makomk 10y ago
  This isn't entirely true. All counties in Wisconsin may use paper ballots as the main form of voting, but there are two different methods of accessible voting offered: direct-recording electronic machines, and machines that print paper ballots. See http://elections.wi.gov/sites/default/files/page/179/voting_equipment_by_municipality_09_2016_pdf_19011.pdf http://elections.wi.gov/sites/default/files/page/179/voting_... (Of course, since only a small percentage of voters use these machines...)
  - tomohawk 10y ago
    The article has been updated with a note that addresses this. It's at the very end.
flexie 10y ago
Absolutely, there is no harm done in double checking. In a way it's weird that in 2016 there is no automatic auditing of all elections for congress and president, given how much is at stake.
learnstats2 10y ago
If Nate Silver's analysis is correct, it suggests strongly that electronic ballots are not for white, educated people. That's a problem in itself, isn't it? What are the reasons for this demographic difference in how the election is conducted?
- SubiculumCode 10y ago
  It also probably introduces suppression effects from multicollinearity when entering the same multiple regression, like Nate Silver did.

by pfooti 10y ago

grzm 10y ago

there is still good reason to audit all national elections. They're quite important, after all. The mad dash to say who won really quickly does a disservice to the process, IMO. Agreed. Maybe it's late, but I'm leaning towards thinking auditing should be closer to the norm than otherwise. Kinda like testing? Get the system so smooth and clean that everything is obvious and trustworthy. Edit: Okay, it is getting late. I just starting thinking of what an election Chaos Monkey would look like.

theparanoid 10y ago
State elections do go to automatic recount in 20 states - http://www.ncsl.org/research/elections-and-campaigns/automatic-recount-thresholds.aspx http://www.ncsl.org/research/elections-and-campaigns/automat...
tehwalrus 10y ago
Donald Trump is an election chaos monkey.
Klathmon 10y ago
Nobody else mentioned it, but an election "chaos monkey" would be interesting. From the few second I thought about it, you could have some "approved troublemakers" that are VERY well controlled and licensed to go and do things like attempt to tamper with machines, multi-vote, ballot stuff, or something similar. If it's done in a way that they know exactly how many extras they put in (how i'm not sure, but it'd need to be bullet proof) they could basically "audit" election systems. Currently, there is no legal way to "test" the system as far as I know.
- grzm 10y ago
  Interesting idea, isn't it? I suspect that many of the failure modes you'd want to test would require a full re-do of the election, at least at the level of the precinct. I don't know how tolerant the population would be. Though I encourage you to follow through and see what you come up with :) I'd be interested to read about it.
  - Klathmon 10y ago
    I'm not so sure it would need to be that drastic. If I get a "stamp of approval" to tamper with a voting machine, I could get to the point of being able to tamper, but not do anything. Or I could (if it was safe to do so) have it add 30,000 votes to a false candidate. It would be similar to how "white hat" hackers need to be able to show a POC but not actually do any harm. Hell even having "false elections" could be a useful first step. It wouldn't be perfect, but it could at least start figuring out where the weakness are.
    - grzm 10y ago
      Hell even having "false elections" could be a useful first step. That, I think, would be potentially useful.
      - nkurz 10y ago
        Here's a 2012 paper from a team at the University of Michigan detailing a "mock election" gone wrong: Title. Attacking the Washington, D.C. Internet Voting System Abstract. In 2010, Washington, D.C. developed an Internet voting pilot project that was intended to allow overseas absentee voters to cast their ballots using a website. Prior to deploying the system in the general election, the District held a unique public trial: a mock election during which anyone was invited to test the system or attempt to compromise its security. This paper describes our experience participating in this trial. Within 48 hours of the system going live, we had gained near-complete control of the election server. We successfully changed every vote and revealed almost every secret ballot. Election officials did not detect our intrusion for nearly two business days—and might have remained unaware for far longer had we not deliberately left a prominent clue. This case study—the first (to our knowledge) to analyze the security of a government Internet voting system from the perspective of an attacker in a realistic pre-election deployment—attempts to illuminate the practical challenges of securing online voting as practiced today by a growing number of jurisdictions. https://jhalderm.com/pub/papers/dcvoting-fc12.pdf https://jhalderm.com/pub/papers/dcvoting-fc12.pdf

zw123456 10y ago

I agree completely, this was an extremely important election with a lot of future implications and it, and really all elections, should be meticulously audited. I think there is a lot of pressure to come up with a quick result when the emphasis should be more on accuracy.

nkurz 10y ago

In Wisconsin, the counties that used paper ballots tended to be less white and better educated. The linked article says that all counties in Wisconsin use paper ballots, and thus the claim that there is a difference between counties that use paper ballots and other counties "cannot be true". Are you disputing this, or did you perhaps mean to say Pennsylvania instead of Wisconsin?

makomk 10y ago
This isn't entirely true. All counties in Wisconsin may use paper ballots as the main form of voting, but there are two different methods of accessible voting offered: direct-recording electronic machines, and machines that print paper ballots. See http://elections.wi.gov/sites/default/files/page/179/voting_equipment_by_municipality_09_2016_pdf_19011.pdf http://elections.wi.gov/sites/default/files/page/179/voting_... (Of course, since only a small percentage of voters use these machines...)
- tomohawk 10y ago
  The article has been updated with a note that addresses this. It's at the very end.

flexie 10y ago

Absolutely, there is no harm done in double checking. In a way it's weird that in 2016 there is no automatic auditing of all elections for congress and president, given how much is at stake.

learnstats2 10y ago

If Nate Silver's analysis is correct, it suggests strongly that electronic ballots are not for white, educated people. That's a problem in itself, isn't it? What are the reasons for this demographic difference in how the election is conducted?

SubiculumCode 10y ago
It also probably introduces suppression effects from multicollinearity when entering the same multiple regression, like Nate Silver did.