Monday, February 01, 2010
Ireland ponders how to dispose of its voting machines
Eight years after they were acquired for €52m, the government wants to return 7,500 barely used electronic voting machines to their manufacturer.
John Gormley, the environment minister, announced last March that he had set up an inter-departmental taskforce to deal with the disposal of the machines, after deciding they would never be used to count votes in an Irish election.
Times Online - Voting machines to be cast out - January 24, 2010
Previously:
February 10, 2005 understanding the true costs of voting machines
February 07, 2005 Ireland does things right
February 03, 2005 Canadian e-voting officials, behold your future: Ireland
Labels: electronic voting, ireland
six years
Labels: meta
article about the Internet voting dialogue
I'm quoted:
“The municipalities are perhaps naive
about the amount of risk they’re assuming,”
warned internet voting security expert Richard
Akerman of the PaperVoteCanada.ca
blog, though. “Very closely contested elections
like Al Franken’s recent race for the U.S. Senate
were only settled because people could
actually see the ballots,” he said. Had it been
conducted over the internet, “the expense of
defending the integrity of that system in the
courts would have been huge,” he claimed.
My concerns include:
* for the risk of coercion, they are acknowledging but accepting this - but have we had a serious debate about whether this is a risk that should be accepted?
* for the risk of a recount, they are simply accepting that all you can do is go and look at the digital data (the "data points" as it was described at the event) - there is nothing to actually recount - while this approach has been accepted, I can easily see an aggressive challenge that required a complete end-to-end forensic audit, which would require a level of technical expertise and time that would be, as I said in the quote, hugely expensive AND raise huge trust issues once people realised both how complex and how opaque these systems are
Labels: elections canada, internet voting, ivotecan
Sunday, January 31, 2010
tweet archive
I have flipped the order so it is more readable - it's oldest first.
First tweet is at 8:54 AM Jan 26th 2010 and last one was at 4:56 PM Jan 26th 2010.
There are a total of 276 tweets.
BEGIN TWEETS
am set up on tethering and will be liveblogging under hashtag #ivotecan - there is a media section here but I only see one person so far
Elections Canada communications has very graciously allowed me to sit at the media table and get power for my netbook. #ivotecan
event is being opened #ivotecan - Elections Canada speaker up next
2/3 of Canadians likely to vote online according to recent survey - Elections Canada #ivotecan
lessons Canada can learn from other jurisdictions within Canada and outside Canada #ivotecan
Elections Canada pilot project will test secure voting via Internet for selected groups eg disabled, Canadians in other countries #ivotecan
Elections Canada emphasizing convenience of Internet voting - but "must maintain level of integrity that Canadians expect" #ivotecan
"Internet voting as an online service" #ivotecan - Elections Canada
Group is working on consistent cross-level standards (provincial, national etc.) #ivotecan
members of parliament and other experts reported to be in audience #ivotecan
Prof. Alvarez up next #ivotecan
Prof Alvarez and audience #ivotecan http://twitpic.com/zuo00
Alvarez will talk about American experience, upsides and downsides #ivotecan
Rationale for Internet Voting: evolution in US from handcounted to optiscan to paperless (nonnetworked and networked systems) #ivotecan
electronic technologies also used throughout the elections process in the United States #ivotecan
defining Internet voting: transmission of ballot over network - references his book One Click One Vote #ivotecan - public elections context
EDITORIAL NOTE: I misheard Alvarez, the book is actually Point, Click and Vote: The Future of Internet Voting. He has also written other books on the topic. END EDITORIAL NOTE
both home computer as well as kiosk Internet voting #ivotecan
Why innovate election tech? - turnout, accessibility, security (!), accuracy (!), efficiency, international access, cost #ivotecan
"How can these technologies improve the efficiency and reduce the cost of election administration?" #ivotecan
American experience - elections have vastly decentralised administration - run at the county level - not national #ivotecan
American experience - "complexity of ballots, regulations and procedures" #ivotecan - may be "dozens and dozens" of items
American experience - multiplicity of ballots, in different languages, covering huge number of items to vote upon #ivotecan
American experience - 2000 Presidential election - controversies have continued about use of electronic voting tech #ivotecan
American experience - California Internet Voting Task Force (2000) - has shaped a lot of US thinking #ivotecan
American experience - Internet voting - Alaska Republican party (Jan 2000) - Arizona Democratic party (March 2000) #ivotecan
Internet voting in 2000 Presidential election - 6 million Americans overseas (military, gov etc.) - special voting rights #ivotecan
international voting - mail transit time to and from e.g. Iraq is a big concern - Internet voting reduces transit time #ivotecan
2000 experiment was a proof of concept - focus on feasibility - electronic version of mail voting system #ivotecan - limited # participants
US international Internet voting used PKI credentials for authentication #ivotecan
not a lot of data - 91 registered, 84 voted using international Internet voting system for US in 2000 #ivotecan
"no security breaches found" for 2000 international Internet vote for US #ivotecan
followup: SERVE - Secure Electronic Voting Registration and Voting Experiment - planned to involve as many as 100,000 #ivotecan
SERVE wasn't implemented because in early 2004 study by computer security experts caused it to be cancelled #ivotecan
in early 2004 Michigan Democratic Party allowed online voting - 28.57% online votes of 162,000 votes total #ivotecan
"Controversies regarding electronic voting machines in 2004 and 2006 elections" #ivotecan
"Election admins and stakeholders reluctant to take on risks associated with voting pilots experiments or transitions to new tech" #ivotecan
ODBP - Okaloosa Distance Balloting Project, implemented in 2008. Kiosk voting for UOCAVA citizens at 3 international locations #ivotecan
there were a few problems with Okaloosa tech but tiny number (<100) voters #ivotecan
use of kiosks means you can ensure the kiosk is secure, rather than using insecure personal computers #ivotecan
(for tests) "Without better scientific design, most of the important outcome variables are difficult to assess" including security #ivotecan
"insufficient data collected" based on US Internet voting experiments to date #ivotecan
Security: What are the real vulnerabilities? How can you mitigate vulnerabilities? Need real experiments #ivotecan
next up: panel on Canadian experiences with Internet voting #ivotecan
Nicole Goodman of Carleton moderating and introducing the panel, which will discuss Canadian municipal Internet voting #ivotecan
first up: Markham's Online Voting Experience by Kimberly Kitteringham and Andrew Brouwer (Town Clerk & Deputy Town Clerk) #ivotecan
Markham Internet voting: 2006 election and plans for 2010 #ivotecan
80% of Markham residents have high-speed Internet access #ivotecan
Why online voting: electronic service delivery, multichannel service delivery, changing lifestyles, "new electorate", convenience #ivotecan
municipal turnout hovered around 30% - Internet voting a channel to encourage participation in voting process #ivotecan
online voting a way to enhance participation by people with disabilities #ivotecan - equal access to the electoral process
2003 positive Internet voting experience positive, recommended online voting for 2006 #ivotecan
Principles identified: security, accuracy, privacy, authentication/verification #ivotecan
Independent Risk Analsys by Henry Kim of York University; Gartner Group security review of IT platform #ivotecan
Dr. Kim found "similar reasonable risks" with two-step voting to in-person voting, and better characteristics than mail-in voting #ivotecan
Partnered with Election Systems & Software (ES&S) for provision of online voting; security of platform verified by Gartner Group #ivotecan
Comprehensive communications plan about Internet voting / voter awareness provided by Delvinia Interactive #ivotecan
2006 online voting only available during early voting period #ivotecan
reporting positive numbers >75% satisfaction from Delvinia survey #ivotecan found it convenient, voted from home
approx 6000 voted online in 2003, approx 10,000 voted online in 2006 #ivotecan
Change in online voting: earlier campaigning, be clear about ID requirements, change in nature of scrutineer function #ivotecan
scrutineers obviously cannot see voters receive and cast their ballot, unlike in-person voting #ivotecan
2010 Markham issuing RFP for online and tabulator vote systems - 3rd party review of online voting security - access plan #ivotecan
Markham "online voting viewed by staff as continued opportunity for service excellence and civic engagement" #ivotecan
Halifax Regional Municipality (HRM) Internet voting experience next up #ivotecan
Cathy Mellet, Acting Clerk/Manager, HRM #ivotecan
HRM covers large physical area, estimated to have population over 410k by 2012 #ivotecan
4 year "e-voting journey" starting in 2004 - Jan 2007 council approved Internet/phone advance voting with "2 levels of ID verify" #ivotecan
discussing mitigating risks while taking advantage of opportunities #ivotecan
RFP in 2007, selected Intelivote for HRM #ivotecan - had to change Municipal Elections Act and HRM by-law to permit
2008 event demographics 279,000 electors; advance voting: 10% of eligible, 28% of votes cast, 88% used Internet. #ivotecan
"engagement matters to voters" HRM #ivotecan
Principles Balance: accessibility vs scrutiny, engagement vs. integrity, convenience vs security... #ivotecan
objectives: ensure integrity, ensure compliance with regulations... #ivotecan
Partnership with Elections Nova Scotia & vendor #ivotecan
HRM election system & data transfer to vendor #ivotecan - also needed support/help centre and contingency plan
something about firewalls but presentation is going way too fast for me to keep up #ivotecan
voter identification "2 shared secrets" - mailed out password + voter birthdate #ivotecan
Sept 2009 special election - "complete internet voting from advance voting to election day" - "realtime voters list", kiosk #ivotecan
"substantially increased turnout" for special election (30% vs. 10% in previous special elections) HRM #ivotecan
e-voting works, well received, cost effective, greener #ivotecan
Jon McKinstry, Sales Manager, Dominion Voting Systems - presenting City of Peterborough story #ivotecan
Peterborough population 75,600. Internet voting 4400 registered, 3500 cast a vote, total 7% of votes were cast over Internet #ivotecan
if you registered for online but didn't vote over Internet, you could still come and vote in person #ivotecan
reasons: leader in delivery of voting systems, embrace tech, increase voter participation, adapt to changing lifestyles #Ivotecan
spike in demographics for Internet voting actually people 40-50, didn't actually have a peak in younger voters #ivotecan
needed realtime strikeout of voters list so that you couldn't vote online and then vote again in person #ivotecan
wanted a system that would consolidate votes from optical scan and internet voting #ivotecan
Principles: ... going too fast for me to keep up #ivotecan
independent security audit of Dominion Voting by Digital Boundary Group (London, Ontario) #ivotecan
again a shared secret system with the secret being the year of birth being the "secret" along with a preselected q/a #ivotecan
PIN number through regular postal mail or encrypted email #ivotecan
audit: password strength, denial of service, injection, ensure intrusion detection in place, system security vulnerability scans #ivotecan
audit reported "Dominion system was a very secure solution" #ivotecan
vote: elector ID + PIN number, separate website, answer preselected question set at reg time, ?enter birthdate? (not mentioned) #ivotecan
Peterborough - ease of use - could cast ballot for 5 days, 24 hours a day #ivotecan
election help desk as well as 1-800 call centre provided by vendor #ivotecan ("about 100 calls came in")
computers also provided at city hall, library, other sites #ivotecan
enhanced features: accessible ballot with zoom, audio, JAWS compatibility #ivotecan
Lessons learned: important for officials to have "complete understanding" of process and technology #ivotecan
Lessons: important to have dedicated marketing, increase number of laptops, run longer (from advance to election day) #ivotecan
approx 15 minutes for questions #ivotecan
am sitting next to @punditsguide
Q to panel from @punditsguide : privacy - 1 destruction of e-ballots? (e-ballot could be
linked back to individual) #ivotecan
Q to panel from @punditsguide : 2 what about voters being coerced at home #ivotecan
Markham: unsupervised voting - one person in a household could do all the voting - part of the risk assessment ... #ivotecan
Markham: unsupervised voting "a risk we were willing to accept" - used education about one person, one vote, secrecy of vote #ivotecan
?Markham? - how are online ballots handled - retained for same duration as paper ballot #ivotecan
?Markham? - paraphrase: no way to connect an individual voter to how they voted in the system #ivotecan
HRM - created substantial penalitys ($10k, 2y in jail) for voter fraud, collusion, or influencing #ivotecan
HRM - asked for certificate of destruction for online ballots from vendor #ivotecan
HRM - "two separate systems" that ensure no connection between voter and votes cast #ivotecan
Q City of Toronto: How do you handle recounts? #ivotecan
Halifax - recount = paraphrase "reopen the encrypted file and look at the data points" #ivotecan
Q City of Toronto: do you capture a (screen) image of the vote as cast? A from HRM: no we just record a data point #ivotecan
A on recount from Markham: "an electronic recount of an electronic vote" #Ivotecan
something about "data as recorded when polls closed and put on memory stick for auditor" ? #ivotecan
Jeremy Clark from Waterloo - privacy question - what kind of data is kept about timing of votes - ... #ivotecan
Jeremy Clark... if you keep timing info you can look at vote time and vote recorded and correlate to figure out who cast what vote #ivotecan
answer from panel: timing is kept, it is a risk but ... someone internal would have to do this attack #ivotecan
Q from Elections Ontario: is a preaudit done - is it possible to test the system before event - and is there postevent test #ivotecan
A from HRM - "audit ballots" cast before, during and after election #ivotecan - realtime tests of the system
A from Peterborough - security tests in advance, intrusion tests etc. #ivotecan
A from Markham: similar process to Halifax #ivotecan
Q: load testing? A from HRM: yes, Oracle platform not even stressed, a non-event. Markham: similar to Halifax #ivotecan
Q did you survey people who didn't use the system? do you know why people registered to vote online but didn't? #ivotecan
A from Markham: survey appeared online right after you voted online #ivotecan
EDITORIAL NOTE: At this point I hit an unexpected Tweet cap for a new account (128 tweets). For the rest of the morning I had to move to liveblogging on FriendFeed. I will try to integrate that reporting here later, but for now you can see it by paging through http://friendfeed.com/electronic-voting-in-canada (which also includes some of these tweets)
END EDITORIAL NOTE
tweeted so much, so fast, from this new account that I got temporary twitter lockout. morning reporting at: http://bit.ly/84ynMb #ivotecan
@kirkschmidt there was a Q "risk of internal staff", the response from HRM was "this is a risk we've always had to deal with" #ivotecan
@pmarchi No one has a good (technical) answer to the coercion issue. HRM made coercion "more illegal" with $10k fine, 2y prison. #ivotecan
Just wanted to mention @punditsguide has been doing a great job of tweeting this very fast-moving event. #ivotecan
@jasonkitcat Yeah and in fact several speakers have said convenience mostly helps save existing voters time, no big turnout boost. #ivotecan
I have blogged a brief summary this this morning's very fast, info-packed set of presentations: http://bit.ly/aqPSjY #ivotecan
Tech considerations session presenters: marketer, vendor, open-source guy, tech guy (Peter Wolf of IDEA, Masters in Computer Eng) #ivotecan
Tech considerations panel: Peter Wolf stuck in snowstorm in Frankfurt or something. #ivotecan Projector also not working (tech irony).
Wolf's notes: trust, transparency, but no external evidence of system's correct operation. Hence systems depend on public trust. #ivotecan
Wolf asserts you must then extend greater trust to the entire electoral system as well as have auditors #ivotecan
Wolf: Internet voting - client computer - "nobody can know if this computer can be trusted" #ivotecan
Wolf: observers would like to get insight into operation of systems, and computer security experts may be fundamentally opposed #ivotecan
It's too bad Wolf isn't here, because his notes raise many excellent points. #ivotecan
Wolf: trade secrets may block trust in system, ability to observe operation, due to black boxes e.g. operating systems, code #ivotecan
Wolf: Opening the Black Box. Norway - public access to source codes. Council of Europe - certification guidelines / standards #ivotecan
My editorial comment: it doesn't matter if your source code is open, you can't prove that's the code that is running. #ivotecan
Wolf: commercial vendors were willing to divulge codes if made a condition of Internet voting contracts #ivotecan
Wolf: lack of common standards for certification - issue recognized by Council of Europe #ivotecan
Wolf: sequoia source code released in USA (editor's note: just google that term to find out the results of analysis of the code) #ivotecan
Adam Froman: Delvinia Interactive - marketer/comms for Markham Internet voting #ivotecan
Adam Froman admits up front he doesn't know or care about the technology. He's going to talk about the voter experience. #ivotecan
Delvinia got CANARIE grant to study the use of broadband tech for municipal services - brought $200k to the table for Markham #ivotecan
@zippyFX it's not hard to write a trojan that sends a response back claiming to be the correct software
Delvinia positioning Internet voting as an option, not a replacement for traditional paper vote #ivotecan
Delvinia studied voter attitudes. And also worked on the voter outreach. Including education about registration changes #ivotecan
Delvinia - 2003 - interactive guides - but there's a general need for voter education, regardless of whether they're voting online #ivotecan
Delvinia - web site satisfaction survey - postpolling, online surveys #ivotecan
[ED COMMENT:] In case people don't know Canadian system: scrutineers from all parties watch the open counting of the paper ballots. Many eyes. #ivotecan
Delvinia - with advanced poll, sometimes politicians would show up at people's doors and discover they had already voted #ivotecan
Delvinia: voter registration process was main barrier to Internet voting #ivotecan
@zippyFX the trojan hides in the query stream and lies. Gives the correct CRC, size, response. See e.g. rootkits.
over 90% of people who voted online in Markham said they would be interested in voting in Federal election #ivotecan
Delvinia guy makes "tech is a part of people's lives" argument #ivotecan My counterargument: educate them about the risks of Internet vote.
Delvinia has a point that the new political engagement is a "digital dialogue" with citizens. Engagement beyond vote #ivotecan
Editorial comment: don't mix social media engagement with the need to secure one-time voting experience #ivotecan
Dean Smith of Intelivote also says he will not talk about the tech side of things at all #ivotecan Small Nova Scotia company.
getting sales pitch for Intelivote now #ivotecan
Intelivote assists in writing electronic voting legislation for countries (!) #ivotecan
Intelivote - integrated polling stations, telephone and Internet voting #ivotecan
Intelivote - pitch is "more choice" #ivotecan
talking about components of election system: help center, auditors, Intelivote control, electors, candidates, officials #ivotecan
components of election system diagram shows "Intelivote system" in centre of everything, which kinda freaks me out #ivotecan
Intelivote considers it a benefit that you can vote from anywhere in the world #ivotecan
Intelivote - anecdotal report about first time visually disabled voters were able to cast vote on their own thanks to technology #ivotecan
Intelivote - 2009 by-election "almost 70% voted electronically" is I think what he said #ivotecan
33 municipal elections in Ontario used Internet and/or phone voting #ivotecan "Canada as a leader" rhetoric coming from Intelivote
Speaking of rhetorical questions: Intelivote - "Why are Canadians so open to eVoting?" #ivotecan
Intelivote pitch: choice, flexibility, immediate, auditable results, voter intent clear - no spoiled ballots, enviro friendly #ivotecan
Intelivote pitch (continued): don't have to staff polling stations #ivotecan
Jason Gallagher: open source vs. propriety in 10 minutes or less #ivotecan
err vs. proprietary that is #ivotecan
defines source code #ivotecan
Gallagher explains in proprietary code, you never get to see the source code #ivotecan
looks like @punditsguide has hit a status update limit as well. have directed to http://friendfeed.com/electronic-voting-in-canada
Gallagher explaining open source software - allows peer review of software, no vendor lockin, gives rights to software users #ivotecan
Gallagher: free to modify open source, don't have to rely on vendor #ivotecan
Gallagher: why open source for voting - transparency, not a black box, accountability, auditability, security #ivotecan
Gallagher: how can shared source code be secure? paraphrase "many eyes make bugs shallow" - don't rely on secrets #ivotecan
Gallagher: there will always be hackers, but if your system is open, you also allow people to help you to improve #ivotecan
Gallagher: proprietary advantages - ready made ./ off the shelf, someone to blame if it goes wrong #ivotecan
Q from ? Alex Sussex ? Univ of Ottawa: everyone can witness paper ballot tally. "you can't actually see software occuring" #ivotecan
Q (continued): what role do candidates play in the observability of the tally? #ivotecan
Q (continued): you don't know what's going on inside the system... what role do candidates play to convince the voters #ivotecan
A from Intelivote: candidates want to be involved... the module shows people being struck off the voters list as they vote #ivotecan
A from Intelivote: no equivalent role for scrutineers in electronic world - no recount #ivotecan
A from Delvina: you're asking the wrong question. Should be "What would you need to see equivalent to paper voting?" #ivotecan
Editorial comment: there is no equivalent to observing the internals of the system analogous to scrutineer role #ivotecan
A from computer security researcher who asked original question: "there are new ways that allow voters to engage in the auditing" #ivotecan
Intelivote: system observing itself is "placebo effect" - one electronic process is observing another electronic process #ivotecan
Intelivote does allow peer review of its code #ivotecan
Intelivote uses randomization to avoid matching timestamps to determine who voted for whom #ivotecan
Q: how do panel see Internet voting rolling out across Canada #ivotecan
A from Intelivote: says Canada (and by extension Intelivote) has reputation and experience #ivotecan
Delvinia guy says you can use open source if you have the resources to build the solution #ivotecan
Editorial summary: Intelivote guy argues "reputation and experience", Delvinia guy argues "it's inevitable anyway" #ivotecan
Q from Elections Quebec: is there established, audited open source software available #ivotecan
A: one example in Australia, project has since been cancelled. Professor found error in source code. was fixed. #ivotecan
A from Tarvi: not about open source - about auditability and transparency. Estonia does not publish its source code. #ivotecan
A from Tarvi: Estonia ready "at any second" to sign NDA and provide code for auditing purposes #ivotecan
A from Tarvi about client side code: could be very easy to create malicious client side app - don't give out client side code #ivotecan
A from the audience: more open source - Scantegrity open source system, open voting consortium, ?OSEB? - DRE software #ivotecan
break and then roundtable discussion #ivotecan
observations from Alex Treschel - should do trials, with Canada-specific-research and analysis of the results #ivotecan
Alex Treschel - make sure you are not generalising from very small data sets or experiments #ivotecan
Alex Trechsel - cautions against generalising even from e.g. Halifax to other Canadian municipalities #ivotecan
Tom Hawthorn - when is it right to move? should we lead new tech (in elections) or follow well established technologies? #ivotecan
Tom Hawthorn - experience in UK was that perhaps they hadn't thought things completely through #ivotecan
Tom Hawthorn but if you wait too long, you may miss an opportunity #ivotecan
Tom Hawthorn - need to understand who is driving the process, who is holding the budget - better if electoral admins drive #ivotecan
Tom Hawthorn - place development of voting systems / software in an international context rather than individual countries #ivotecan
Tom Hawthorn - should develop common understanding and set of benchmarks #ivotecan
Tarvi Martens - electoral system is about trust. holds the same for evoting as for paper. #ivotecan
Tarvi Martens - example of failure in Netherlands. example of failure in Lithuania due to suggesting banking credentials #ivotecan
Tarvi Martens - example of failure in ?Finland? - if you screw up deployment, you will be set back a decade or more #ivotecan
Tarvi Martens - if the deployment of your system, including the user part, does not build trust, you will fail #ivotecan
Tarvi Martens - asserts user identity is critical to system (not surprising since he is expert on computer credentials) #ivotecan
Tarvi Martens - password based systems or weak credentials are easy to attack #ivotecan
Tarvi Martens - if people succeed in compromising your system, you will have a huge setback in trust #ivotecan
Jon Pammett: a wide variety of "policy laboratories" in Canada for Internet and other voting systems experimentation #ivotecan
Jon Pammett: not an expert in tech, wondering if Internet voting will increase turnout, but it seems based on today it won't #ivotecan
Jon Pammett: Internet voting doesn't appear to address voter engagement, which is the true driver of turnout #ivotecan
Jon Pammett: concerned about (my words) consequences of Internet voting road not taken #ivotecan
[ED COMMENT:] argument from panel that mixes "tech use" with youth. In my opinion, this is a false mix. Young people are not tech experts. #ivotecan
Editorial comment: I think there needs to be better research into what actually drives voting, rather than speculating #ivotecan
Q from @punditsguide: Canada examples are municipalities which are low turnout, not highly contested elections #ivotecan
Q @punditsguide: how will this work in a much more competitive election where votes are closer #ivotecan
Q (U Calgary): assess evoting based on increased efficiency? (code for saving money) - but if used in advance voting... #ivotecan
Q (U Calgary, contd) will increase cost of elections without noticeable effect on voter turnout? #ivotecan
Q (U Calgary, 2nd question): where research has been done on impact by age, no positive impact in bringing youth vote #ivotecan
Q (U Calgary, 2nd q): seems that Internet vote is mostly middle-aged turnout. #ivotecan
Q (U Calgary): seems like greater cost and no greater turnout - then what is justification for Internet voting? #ivotecan
A (Jon): age profile data is from municipalities - young people not engaged in municipal politics #ivotecan
@jasonkitcat seems to be a dialogue between desire for turnout and issues about trust #ivotecan
A (Jon): in competitive elections - possibly true people would be more likely to attack systems #ivotecan
A (Alex): in competitive elections higher risk - try it out in less competitive contexts too (and remember Swiss cap evote at 10%) #ivotecan
A (Alex): (not exact quote) "doesn't cost that much, comparitively" for "making people happier in democracy" #ivotecan
A (Alex): also remember youth never had high turnout, but it is dramatically low in e.g. Canada #ivotecan Internet voting not a panacea
A (Tarvi): to use Internet voting in Federal election for the first time is a bad idea - start small #ivotecan
A (Tarvi): Estonia formed a group of IT security experts, every step was security, security, security #ivotecan
A (Tarvi): Estonia knew exactly the potential failure points, the risks #ivotecan
A (Tarvi): if you haven't done your security due diligence, hackers can expose issues and destroy trust in your system as in NL #ivotecan
A (Tarvi): if you reuse your system, then over the long term the costs are lower #ivotecan
A (Tarvi): Internet voting not to increase turnout, it's to PRESERVE the turnout #ivotecan
A from Markham: cost for Internet voting were "quite small", "reasonable" #ivotecan
A from Markham: did see increased turnout #ivotecan not enough data to attribute directly to Internet voting
A from Markham: hackers "a cynical argument" against Internet voting, look at opportunities instead #ivotecan
A from HRM: if you can decrease the number of poll locations you decrease cost and "risk" (training / staff risk) #ivotecan
Comment (Nicole Goodman?): We don't know how any particular Internet voting model will work in any jurisdiction, need trials #ivotecan
Comment: yes there will be a large upfront cost, and there should be since it needs to be done right #ivotecan
Comment: cheaper over the long term #ivotecan
Comment: we can't fix turnout with Internet voting but there is no one solution, young people are not homogeneous group #ivotecan
Editorial comment: cheaper over time is hard considering you need 24/7 physical & net security for data centre 365 days/yr #ivotecan
Q: what are the main arguments against Internet voting? #ivotecan (other than security)
Q (Elections Canada): can academics map when a region is "mature" enough to go on an Internet voting route #ivotecan
A (Tom): Germany ruled use of Internet voting unconstitutional as it was inherently un-understandable by avg citizen #ivotecan
A (Tom): no one knows what the cost model is going to be in the future. may see some new kinds of costs #ivotecan
A (Tom): new costs = auditors, consultants, security experts - could be very expensive #ivotecan
A (Tom): most people in elections systems are not experts in electronic systems / security design - maybe they need to be #ivotecan
A (Tarvi): in Estonia Internet voting was challenged about uniformity of voting #ivotecan
A (Tarvi): ruling was that multiple times to vote over-rides privacy concerns (not sure I understand his answer) #ivotecan
A (Alex): groups in Geneva were strongly opposed to Internet voting (computer security experts) #ivotecan
A (Alex): in Geneva they engaged in a dialogue with the computer security experts #ivotecan
http://www.e-voting.cc/ - Internet voting conference, models #ivotecan
A: an argument against Internet voting - voting in person is a communal experience #ivotecan
Editorial comment: first mention today of compulsory voting as a direction for turnout and
engagement #ivotecan
audience comment: 8 million voters in Ontario, 800000 will be voting "electronically" - "it's happening" #ivotecan
I think it's the Intelivote guy: cost savings of electronic voting #ivotecan
aaaand we're done #ivotecan
@jasonkitcat I didn't get a strong sense of a driver other than "seems like a good thing to try"
@punditsguide good to meet you as well
END TWEETS
Labels: elections canada, internet voting, ivotecan