|

 
|
| | Name : | Rebecca Mercuri | Organization : | Notable Software, Inc. | Post Date : | 9/30/2005 |
| Comment : | The VVSG repeats many mistakes of earlier standards in both content and structure. Foremost among its problems is that the VVSG, like its predecessors, sits somewhere between being a design standard and a performance standard, and by failing to determine which direction it intends, provides only mediocrity for either purpose. Although the VVSG purports to “define functional requirements and performance characteristics that can be assessed by a series of defined tests,” in actuality, it relies heavily on pre-existing balloting metaphors, and thus is implicitly predisposed to the assessment of only a limited set of designs. For such designs, it attempts to cover a multitude of bases, offering guidance for disparate sets of equipment that must satisfy mutually incompatible constraints. For example, the usability requirements for ballot casting are necessarily different, in many regards, for paper-based, electronic, and electro-mechanical systems, but the VVSG chooses to convolve the human factors aspects of all of these products into a confusing mish-mash of criteria. As a performance (or functional) standard, the VVSG is overly prescriptive in terms of acceptable manifestations, hence it harbors the potential of discouraging or even thwarting the development and deployment of viable designs that have not heretofore been considered. Thus, the VVSG perpetuates vagaries over acceptability for use of unaddressed configurations. This omission has knowingly been exploited by certain lobbying groups through attempts to defeat the adoption of competitive innovations (like voter verified paper ballots intended to increase auditability and transparency, and overlay templates to improve accessibility) or alternatively, has served vendor interests with allowances for the uncertified introduction of new components with dubious security (such as telecommunications products).
| |
|
|