1 \documentclass{article} |
1 \documentclass{article} |
2 \usepackage{../style} |
2 \usepackage{../style} |
3 \usepackage{../langs} |
3 \usepackage{../langs} |
4 |
4 |
5 \begin{document} |
5 \begin{document} |
6 \fnote{\copyright{} Christian Urban, 2014, 2015} |
6 \fnote{\copyright{} Christian Urban, 2014, 2015, 2016} |
7 |
7 |
8 %https://www.usenix.org/sites/default/files/conference/protected-files/jets15_slides_epstein.pdf |
|
9 % |
|
10 %Jeremy Epstein, SRI International |
|
11 % |
|
12 %In April 2015, the US Commonwealth of Virginia decertified the Advanced Voting |
|
13 %Solutions (AVS) WinVote voting machine, after concluding that it was insecure. |
|
14 %This talk presents the results of Virginia's analysis of the WinVote, and |
|
15 %explores how we got to the point where a voting machine using an unpatched |
|
16 %version of Windows XP from 2004, using hardwired WEP keys and administrator |
|
17 %passwords, could be used for over a decade in most of Virginia. |
|
18 |
8 |
19 %% second angle of the problem |
9 %% second angle of the problem |
20 %Jonathan Zittrain is interested in algorithmic accountability, |
10 %Jonathan Zittrain is interested in algorithmic accountability, |
21 %from Facebook’s ability to tell that two people are in a |
11 %from Facebook’s ability to tell that two people are in a |
22 %relationship before they announce it, to their ability to |
12 %relationship before they announce it, to their ability to |
101 punch cards, now DREs and optical scan voting machines. |
91 punch cards, now DREs and optical scan voting machines. |
102 But there is a lot of evidence that DREs and optical |
92 But there is a lot of evidence that DREs and optical |
103 scan voting machines are not as secure as they should |
93 scan voting machines are not as secure as they should |
104 be. Some states experimented with Internet voting, but |
94 be. Some states experimented with Internet voting, but |
105 all experiments have been security failures. One |
95 all experiments have been security failures. One |
106 exceptional election happened just after hurrican Sandy |
96 exceptional election happened just after hurricane Sandy |
107 in 2012 when some states allowed emergency electronic |
97 in 2012 when some states allowed emergency electronic |
108 voting. Voters downloaded paper ballots and emailed them |
98 voting. Voters downloaded paper ballots and emailed them |
109 back to election officials. |
99 back to election officials. |
110 |
100 |
111 \item Estonia used since 2007 the Internet for national |
101 \item Estonia used since 2007 the Internet for national |
153 \end{center} |
143 \end{center} |
154 |
144 |
155 \noindent The abstract says: |
145 \noindent The abstract says: |
156 |
146 |
157 \begin{quote}\it |
147 \begin{quote}\it |
158 In April 2015, the US Commonwealth of Virginia decertified the |
148 ``In April 2015, the US Commonwealth of Virginia decertified the |
159 Advanced Voting Solutions (AVS) WinVote voting machine, after |
149 Advanced Voting Solutions (AVS) WinVote voting machine, after |
160 concluding that it was insecure. This talk presents the |
150 concluding that it was insecure. This talk presents the |
161 results of Virginia's analysis of the WinVote, and explores |
151 results of Virginia's analysis of the WinVote, and explores |
162 how we got to the point where a voting machine using an |
152 how we got to the point where a voting machine using an |
163 unpatched version of Windows XP from 2004, using hardwired WEP |
153 unpatched version of Windows XP from 2004, using hardwired WEP |
164 keys and administrator passwords, could be used for over a |
154 keys and administrator passwords, could be used for over a |
165 decade in most of Virginia. |
155 decade in most of Virginia.'' |
166 \end{quote} |
156 \end{quote} |
167 |
157 |
168 The reason that e-voting is such a hard problem is that we |
158 The reason that e-voting is such a hard problem is that we |
169 have requirements about the voting process that conflict with |
159 have requirements about the voting process that conflict with |
170 each other. The five main requirements for voting in general |
160 each other. The five main requirements for voting in general |
370 incompetence and the inferiority of their products by |
360 incompetence and the inferiority of their products by |
371 requiring that election counties must not give the machines up |
361 requiring that election counties must not give the machines up |
372 for independent review. They also kept their source code |
362 for independent review. They also kept their source code |
373 secret. This meant Halderman and his group could not obtain a |
363 secret. This meant Halderman and his group could not obtain a |
374 machine through the official channels, but whoever could hope |
364 machine through the official channels, but whoever could hope |
375 that revented them from obtaining a machine? Ok, they got one. |
365 that prevented them from obtaining a machine? Ok, they got one. |
376 They then had to reverse engineer the source code in order to |
366 They then had to reverse engineer the source code in order to |
377 design an attack. What all this showed is that a shady |
367 design an attack. What all this showed is that a shady |
378 security design is no match for a determined hacker. |
368 security design is no match for a determined hacker. |
379 |
369 |
380 Apart from the obvious failings (for example no paper trail), |
370 Apart from the obvious failings (for example no paper trail), |
540 somewhere the money went). Even if there might be more |
530 somewhere the money went). Even if there might be more |
541 gigantic sums at stake in online banking than with voting, it |
531 gigantic sums at stake in online banking than with voting, it |
542 can be made reasonably secure and fraud-safe. That does not |
532 can be made reasonably secure and fraud-safe. That does not |
543 mean there are no problems with online banking. But with |
533 mean there are no problems with online banking. But with |
544 enough thought, they can usually be overcome with technology |
534 enough thought, they can usually be overcome with technology |
545 we have currently avialable. This is different with e-voting: |
535 we have currently available. This is different with e-voting: |
546 even the best have not come up with something workable yet. |
536 even the best have not come up with something workable yet. |
547 |
537 |
548 |
538 |
549 This conclusion does not imply that some special cases of |
539 This conclusion does not imply that some special cases of |
550 Internet voting cannot be made to work securely. Just in a |
540 Internet voting cannot be made to work securely. Just in a |