| || ||TGDC Resolution #06-11||Resolution #06-11: On August 25, 2005, the EAC Standards Board adopted Resolution 2005-003, which addressed Volume 1, Section 2 of the VVSG. On page 2-22, Section 126.96.36.199.2.2.6, the Standards Board called for striking a VVSG requirement that was very similar to this resolution. Until we have pursued the line of inquiry described below, the Standards Board should stick with its past vote on this subject, and advise other EAC boards accordingly.
We need to avoid the one-size-fits-all approach that this resolution seems to endorse and seek to discover what kind of assistive devices are available and in use - particularly those devices that do not require a data bridge or software from existing voting systems. I would like to hear from NIST and disabilities experts about what devices persons with disabilities currently rely on to better understand and interpret text on a page, including optics, lighting, personal automated read-back aids and cell phone applications. I would like to hear what features, such as ballot design, would be useful to render these peripherals and other aids more feasible and effective.
To achieve the biggest bang for the buck in VVSG 2.0, we may want to encourage and enable the use of the most successful tested methods of independent verification, leaving VVSG 3.0 and 4.0 to set ever more ambitious goals, bearing in mind that the more independent these devices are from the voting system itself, the better they serve the purpose of verification.
I am concerned this resolution will lead us to forego opportunities for persons with disabilities that would otherwise be achievable had we aimed at a more reasonable target. With reference to Resolution #06-11, section 3.b., I am concerned that an awkwardly worded resolution could lead to voting systems that will be out of reach of the budgets and the capacity of many of our local elections jurisdictions. The Standards Board has a duty to establish practical requirements that vendors can supply in a time frame compatible with voting system replacement requirements, and it cannot achieve this goal by establishing specifications that would place the new Standards in indefinite economic limbo, or that would contain too many moving parts to allow for implementation.
| ||20||5.3||As stated in this portion of the report, "As the innovation class is not actually an architecture, but rather a process by which presently unknown, new architectures could be considered, it adds nothing to the alternatives previously discussed." The executive summary (see page 1 above) states that, after reviewing the range of approaches to auditability, the Working Group found "no alternative that does not have as a likely consequence either an effective requirement for paper records or the possibility of undetectable errors in the recording of votes." This key finding is well supported in the body of the report. It should put to rest the notion that a viable alternative to software independence, achieved through use of voter verifiable paper records, exists at this time. Without the benefit of this sort of rigorous analysis,the Standards Board in the past has placed undue reliance on the possibility that a technological alternative to software independence would emerge in the "Innovation Class." On this basis, Standards Board Resolution 2007-06 recommended that the EAC should not make software independence a requirement in the next iteration of the VVSG. Four years have passed and no such alternative has emerged or appears to be on the horizon. The possibility of undetectable errors in the recording of votes is unacceptable where public confidence in election outcomes is at firstname.lastname@example.org||6/14/2011
| ||6||1||The executive summary of the Auditability Working Group report states that, after reviewing the range of approaches to auditability, the group found "no alternative that does not have as a likely consequence either an effective requirement for paper records or the possibility of undetectable errors in the recording of votes." This key finding is well supported in the body of the report. It should put to rest the notion that a viable alternative to software independence, achieved through use of voter verifiable paper records, exists at this time. Without the benefit of this sort of rigorous analysis,the Standards Board in the past has placed undue reliance on the possibility that a technological alternative to software independence would emerge in the "Innovation Class." On this basis, Standards Board Resolution 2007-06 recommended that the EAC should not make software independence a requirement in the next iteration of the VVSG. Four years have passed and no such alternative has emerged or appears to be on the horizon. The possibility of undetectable errors in the recording of votes is unacceptable where public confidence in election outcomes is at email@example.com||6/14/2011
| || || ||Resolution #06-11
As a new member of the board, I was intrigued with the Auditability Report and the notion of Software Independence. For the disabled voter, the requirement for private and independent voting using a paper or electronic ballot is critical. However, 3.b. of the resolution, specifically states a requirement that is certainly not plausible with current equipment such as the AutoMark. The disabled voter not only handles the ballot to insert it into the marking equipment, but after verifying the markings, releases the ballot in a security sleeve to insert in a tabulator. Unless the equipment for the disabled operates entirely electroncially, touch screen marker and tabulator combined, the voter with a disability will need to handle the ballot. So, 3.b. does not reflect the current equipment in use across the country at this time. ||firstname.lastname@example.org||6/14/2011
| || || ||Resolution #04-11
Software independence and auditability of election systems are worthy goals to accuracy and voter confidence. The three kinds of verification of ballot choices (Cast Verification, Read Verification, and Count Verification) are sensible. The independent verification functions outlined are sensible. However, neither system can be considered without recognition of the Election Administrator’s role in assuring security and accuracy. Neither system can be stand alone outside of the context of Election Administrator’s active involvement.
There should not be a one size fits all approach for accessibility for independent voter verification of a paper ballot. Further, accessibility considerations should take in to account the Election Administrator’s role in assisting and complimenting individual voters’ needs. The definition of a determinative vote record fails to consider assistance of a third party with minimal risk of ballot disclosure, and fails to consider the wishes and desires of a voter with disabilities and the role of Election Administrator’s assistance. A “holistic” approach should consider trained third party assistance when evaluating accessibility. ||John.Gale@nebraska.gov ||6/13/2011
| ||1|| ||As a former EAC staffer that worked very closely on the VVSG I wanted to attempt to provide the Standards Board some historical background on the reason for the creation of the “Alternatives to Software Independence” white paper adopted in the corresponding resolution. This comment is not intended to discuss the merits of SI or the alternatives but rather provide background.
Software Independence (SI) was the most commented on portion of the Next Iteration of the VVSG by far. Many comments were submitted supporting SI and its goal of preventing any undetectable fault or error. However, many comments were submitted raising concerns with the current technological consequence of SI,that being that a paper record remains the only way to meet the SI requirement. As the NIST powerpoint on this subject pointed out, SI does not formally require a paper record. However, the only current technology that can meet the requirement is some sort of paper record. That was the focus of much of the criticism of SI.
In December of 2009 the EAC tasked the TGDC with developing an alternative solution to SI that does not effectively require a paper record but does, as best as technically possible, achieve the goals of SI. It is important to note that the EAC has not made any policy decision on the inclusion or exclusion of SI in the Next Iteration of the VVSG. The EAC is simply asking the TGDC to provide it with alternative requirements to SI so that an informed and educated policy decision can be made. To date SI remains an option for inclusion in the Next Iteration of the VVSG.
The white paper approved by the TGDC broadly outlines some possible alternatives and the various strengths and weaknesses of each. It does not offer any actual requirements that could be added to the VVSG. I plan on offering my thoughts on the plausibility of the alternatives presented in the white paper later in the virtual forum.
I think it is also important to note that “Independence Verification” requirements already partially exist in the 2005 VVSG. Volume 1, Appendix C of the 2005 VVSG is an informative section created by the TGDC in order to guide the possible development of Independent Verification systems. This section led the TGDC to later discuss Independent Verification systems and ultimately to the discussion and development of the idea of Software Independence. Here is a link to Volume 1 of the VVSG, simply scroll to Appendix C to view the section I am referring to (http://www.eac.gov/assets/1/AssetManager/2005%20voluntary%20voting%20system%20standards%20volume%201.pdf).
I hope this information helps inform this virtual forum. Again, this note is not intended to support or reject SI but simply provide some additional historical background to the development of the “Alternatives to Software Independence” white paper and corresponding TGDC resolution.
|Rows: 1 - 6 of 6|| Page: 1 of 1|