IEEE.jpg

[Image courtesy of IEEE]

I recently received an email from John Wack regarding the new IEEE draft standard for election data – and it’s worth sharing key parts of it with you:

Hi Doug:

I’m writing [about] the IEEE 1622.2 election results reporting draft standard. I’m the chair of the sponsoring committee in IEEE and editor of the 1622.2 draft standard, and we’ve had significant input/buy-in into the standard from several of the manufacturers[], a number of election officials including the Ohio SoS (who published November results in the 1622.2 format), and some industry groups such as the Associated Press. I’ve enjoyed working closely with Kim Brace [of Election Data Services, Inc.] especially, who was very helpful in making this not only a format for election results reporting but also a format for election management system import/export in general. Sarah Whitt from Wisconsin[‘s Government Accountability Board] chairs the 1622.2 working group and has been very helpful in attracting other election officials to the IEEE.

The draft standard is in the IEEE balloting process but we are taking pains to make others in the election community aware of it so that they could also comment (IEEE agreed to make the draft publicly available and to waive any subsequent fees they would otherwise charge on the standard)…

As you know, I work at [the National Institute of Standards and Technology (NIST)] where we have been working for some years in the voting system standards and certification area…… NIST [is focusing also] on assisting the states, especially in areas having to do with election infrastructure. Thus, the common data format effort was started with the intention to make election equipment interoperate with regard to data definition and data format. We began working with the IEEE as a means for establishing guidelines and standards that election officials could require of manufacturers if they so choose. The IEEE process was especially desirable in that we could more easily involve manufacturers and subject matter experts such as election officials and individuals such as Kim …

[NOTE: IT GETS WEEDY HERE BUT KEEP READING – ed.]

The 1622.2 draft standard is a good piece of work not only because it had direct input from the audiences I’ve mentioned above, but also because the XML import/export format it contains was based on a well thought-out data model that is being used for other common data format standards underway such as for electronic pollbooks and voter registration databases. It addresses election management systems in general – pre-election, election night, and certified results. It handles geopolitical geography across many different types of jurisdictions quite well (thanks mainly to Kim and the Associated Press and input from states such as Sarah’s Wisconsin that have more complicated geography), and it contains the capability for associating multiple identifiers with the data elements so that they can be cross-referenced. By this, I mean that a district, for example, can have a variety of different identifiers associated with it, perhaps for a county, across the state, and perhaps nationally, with the intention of eventually guiding states to using more consistent identifiers that will then make it significantly easier for election analysts. We have heard consistently that data aggregation across counties and across states can be quite difficult, with sometimes each county having its own set of identifiers for geographies such as districts that span multiple counties, and we believe that the 1622.2 format will assist significantly in helping counties and states to unify identifiers and produce election data that can be reported and digested readily.

Because the 1622.2 format is, at heart, simply a low-level import/export format, other applications such as [Pew’s Voting Information Project (VIP)] can use it as a base format – we have, by the way, written translations between the 1622.2 format and VIP’s, and found that the translations work quite well. Pew has agreed to review the draft standard, with my intention being to ensure interoperability with … the VIP schema.

We anticipate that the 1622.2 standard will eventually be adopted as part of the next set of voting system standards… I mentioned that Ohio published their November results in the 1622.2 format – Ohio would like to use this format also to make election management systems from different vendors work together smoothly so that data can be easily aggregated up at the state level. I’m especially indebted to Matt Masterson [of Ohio – and EAC Commissioner!] for his help here.

Comments and voting are available to members and non-members of IEEE alike at
http://grouper.ieee.org/groups/1622/groups/2/ – the deadline is Thursday, January 15, 2015.

This kind of work doesn’t get the attention of high-profile lawsuits or legislation, but it’s incredibly vital to the future of the profession. I’ll be taking a look at the standard before January 15 and I encourage you to do the same.

Thanks to John and Sarah and everyone else who has gotten the IEEE standard this far – you are doing true election geek work!