[Image courtesy of wax-wane]
If I go there will be trouble / And if I stay it will be double / So come on and let me know
This indecision's bugging me (Esta undecision me molesta) / If you don't want me, set me free (Si no me quieres, librame)
- The Clash, "Should I Stay or Should I Go"
Day 2 of the EAC/NIST Future of Voting Systems Symposium was a deep, deep dive into the policies, procedures and process behind standard-setting at the federal level. The morning was devoted to a discussion about how federal standards are developed and how market players (especially vendors and consumers) conform to them. It was truly fascinating to hear how different standards work in practice, especially since the speakers were so enthusiastic and detailed about the subject. [My highlight of the morning was the discovery that low-flush toilets are tested using Japanese bean paste.]
But it was in the afternoon, when the talk turned to voting system standards in particular, that things got interesting. During a series of panels, it became clear that there are still numerous unresolved issues in the federal certification and testing process, including:
- How broadly to define a "voting system" for the purposes of testing, i.e., whether to incorporate new technologies like e-pollbooks; and
- Whether to expand the types of testing required to include a wider family of "-ilities" - auditability, accessibility, usability, etc. - in order to ensure that such features are "baked in" to voting systems and merely added on at the back end.
Near the end of the day, however, an issue emerged which has been percolating for some time in the field: namely, whether certification is still working for states. At numerous points during the afternoon, election officials like Michigan's Chris Thomas expressed concern about the slow pace of standard-setting and testing and suggested that jurisdictions may look elsewhere - for example, to interstate testing compacts - to achieve the same kind of regulatory protection that the the current regime is supposed to accomplish.
It's difficult to overstate how important this is; remember, the federal standards are "Voluntary", meaning that the only thing that makes them required is state law. [For example, New York's Robert Warren talked about how the state took the federal standards and changed all of the "shoulds" to "shalls".] If states feel like the current system isn't working then they may begin to wonder why they are requiring themselves to use it.
Indeed, some states are learning that they can move on without it: Washington State's Patty Murphy talked about how her state ended up provisionally certifying new equipment for multiple years because of the slow pace of standard development and adoption. On Day 1, Los Angeles' Dean Logan and Austin, TX's Dana DeBeauvoir talked about their efforts to develop new voting systems as a response to a stalled market due (in part) to the lack of movement at the federal level. It certainly doesn't help that that the Election Assistance Commission lacks members and is thus severely limited in what it can accomplish; conferences like this week's symposium are important but won't take the place of a fully-functioning federal agency.
Consequently, the message from states and localities yesterday was simple: they need an answer about whether and how federal certification will work - and until they get that answer they are leery of any effort to expand the scope of federal testing. If they don't get their answer soon, you can expect to see many more questions about the future of the federal regime - and if they don't get the response they're looking for, we could see states and localities deciding that they "should go."