Open Source Fix For US Voting System? - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Government // Cloud computing
01:23 PM
David F Carr
David F Carr
Connect Directly

Open Source Fix For US Voting System?

Open source programmers and maverick election officials want to improve the way we vote, register to vote, and count the votes. Wish them luck.

The last surge of investment in voting technology happened a decade ago. Since then the regulatory apparatus for election reform has broken down, and voting machines themselves are starting to fail as well. Every election shines attention on a different bit of dysfunction, from long lines at polling places to cyber security risks.

Open source to the rescue? Maybe, although probably not in time to have much impact on the 2016 election. Silicon Valley's Open Source Election Technology Foundation (OSET) is methodically chipping away at the problem, building credibility with software for voter registration and election-night reporting while also working on the more challenging problems of improving the casting and counting of votes. Meanwhile, a few brave -- and impatient -- county election officials are embarking on their own voting technology design and development projects, which may or may not intersect with OSET's work.

[Does everything have to be partisan? Read Wanted: Honest Algorithms For Voter Redistricting.]

The bottom-line goal of these initiatives is the same: to move away from reliance on proprietary technology while boosting transparency and leaving a reliable audit trail -- making it clear that the victor in any contest really is the candidate who won the most votes.

"[Current voting machines] certainly weren't constructed with transparency and accountability in mind," bemoaned E. John Sebes, co-director of OSET and CTO for the TrustTheVote Project. "There's a lot of demand for that, but not a lot of money for [it]." One of the ways OSET wants to help is by reducing the cost of doing the right thing.

The last purported technological fix to what can go wrong in elections arguably backfired. The 2002 Help America Vote Act (HAVA) was partly a reaction to the Constitutional crisis of 2000, when a disputed vote count in Florida caused the US Supreme Court to decide the election in favor of George W. Bush. That was the election of hanging chads and the Palm Beach butterfly ballot, which raised issues of unreliable analog technology (punch card voting) as well as usability and design. HAVA set new standards and provided federal funds for voting upgrades, but critics saw this as a boon for a handful of digital voting technology vendors and worried that the introduction of touchscreen technologies raised as many questions as they answered.

An election information app created by the Open Source Election Technology Foundation
An election information app created by the Open Source Election Technology Foundation

In principal, a good digital user interface should filter out errors such as voting for two people in the same race (or appearing to do so because of smudged ink or torn paper). Yet stories about poorly calibrated devices recording votes in favor of the wrong candidate abounded, and all-digital systems meant votes could evaporate with a computer memory or storage glitch, leaving no physical record such as a punch card for recount workers to hold up to the light. The outcry over the lack of a verifiable paper trail drove many jurisdictions back to optical-scan voting systems, and the next generation of technology is also likely to use a combination of paper ballots and digital tabulation. The new digital reformers want to go further, making voting systems more affordable and accessible to routine auditing, debugging, and inspection.

The target system reformers are converging on would provide voters with a ballot marking device that produces a neatly marked paper ballot, which then gets fed into a scanner. The voter will get the opportunity to verify that votes have been recorded correctly by the computer system, and the paper is retained for subsequent recounts.

Absentee voters and others voting by mail could still fill out the exact same ballot by hand. At polling places, touchscreen devices and computer-assistive technologies for disabled voters could still be used, as long as they produce a paper ballot for backup.

Greg Miller, co-executive director and chief development officer at OSET, said the group of technologists who started the project were inspired to do something proactive after witnessing another round of election glitches in 2006. "We realized we could spend a lot of time whining about what's wrong with these machines, [arguing that they are profitable for the vendors but not very functional]," he said. "But instead of just talking about what's wrong, we ought to be working on fixing it."

In January, the bipartisan Presidential Commission on Election Administration warned of an "impending crisis in voting technology" in its report on the American voting experience: "By the end of the decade, a large share of the nation's voting machines, bought 10 years ago with HAVA funds, will reach the end of their natural life and require replacement. To address this impending challenge and to usher in the next generation of voting machines, the standards and certification process for new voting technology must be reformed so as to encourage innovation and to facilitate the adoption of widely available, off-the-shelf technologies and 'software-only' solutions."

Unfortunately, while that commission was bipartisan, little else in Washington is. Any change in how elections are conducted carries with it some chance of favoring one party over the other, and suspicions run deep. With Congress blocking President Obama's nominations to the Elections Assistance Commission created as part as HAVA, federal work on standards for election technology is moribund. That's one reason election technology vendors are reluctant to invest in new systems, which might never get certified.

"This is a market that's so screwed up that it's practically not a market," Sebes noted. With an open source project, OSET can invest without expecting to make a profit. Meanwhile, with the federal government gridlocked, state and local officials are reasserting their authority.

"We're trying to break the mold," Los Angeles County clerk and registrar of voters Dean Logan said in an interview. "We want the specifications to be developed outside of the competitive marketplace. We don't want specifications designed around a profit model." Los Angeles sat out the last round of voting technology upgrades, concerned that the equipment on the market wouldn't meet its capacity needs. Since then, the voting technology market has consolidated further and innovation has virtually ground to a halt. At the same time, LA's current voting technology is obsolete, at risk of breaking down, and out of compliance with the law if not for being "grandfathered" in.

All these factors led Logan to decide the only way to obtain technology that met his requirements was to create it himself. That doesn't mean LA will hire the programmers and develop the software itself, he said, but it will set the expectations, select the contractors, and own the resulting code. The famed technology design firm Ideo consulted with LA on the equipment mockups and conceptual designs that now need to be turned into a working system.

"This is where the principle of public ownership comes into play," Logan said. "We're large enough that we have the resources to support a system on our own. Most jurisdictions in the country don't have that." He is not averse to sharing that work with others, or taking advantage of open source efforts, as long as he gets something that the county can control, inspect, and fix without being dependent on an outside vendor. Although he has no formal relationship with OSET so far, he believes its efforts are consistent with its goals and the organization will likely put in a proposal to provide the software, or components of it.

Travis Country, Texas, is working on a similar project. Many other voting officials are looking for new and better solutions but are not brave (or crazy) enough to go the do-it-yourself route. They would be very happy if someone else would make the effort and be willing to share.

With its TrustTheVote project, OSET hopes to provide solutions to election technology challenges that everyone will be able to tap into. Originally known as the Open Source Digital Voting Foundation, the group changed its name to avoid the conflation of "digital" with "web-based." While it draws organizational inspiration from web-centric open source organizations like Mozilla and the Apache Software Foundation, OSET does not promote moving voting to the web, where cyber security issues

David F. Carr oversees InformationWeek's coverage of government and healthcare IT. He previously led coverage of social business and education technologies and continues to contribute in those areas. He is the editor of Social Collaboration for Dummies (Wiley, Oct. 2013) and ... View Full Bio
We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
1 of 2
Comment  | 
Print  | 
More Insights
Newest First  |  Oldest First  |  Threaded View
User Rank: Apprentice
10/8/2014 | 4:48:46 PM
Re: Pooling of resources
The good news about pooled resources is the "perpetual harvest" principle of open source software.  Pooling in this sense is possible by all adopting jurisdictions gaining access to an open source work and benefiting from the enhancements thereof.  LA County is a likely leader in the development and eventual deployment of a new CA certified for use voting system.  That system is intended to be based on software that is open source.  And all counties in CA would benefit from adopting it to the extent of their local requirements and needs.

Incidentally, for those readers unfamiliar, by "open source" we generally mean both the process of development of the software source code (the instructions that tell a device how to run an app or application), AND how that software is obtainable, distributed, improved, etc.

On the issue of licensing source code, once available, I have been engaged in software engineering and development for over 30 years, and things do move quickly, but I am unfamiliar with what a "certifiable General Public License" means other than to possibly refer to a license that recognized by an open source licensing governing body like OSI.  And to that end, I absolutely concur.  I am confident LA County's source code, once available, will be subject to an open source license that any jurisdiction can adopt.

But here is an important point: the commenter also asserts some need to be vigilant against some sort of self-empowering licensing schemes.  While I am uncertain what exactly that means, I can venture a guess that it is a veiled reference to their misunderstanding of the need for an open source software license that can be readily accepted by local governments, who have procurement regulations that (unlike the federal government's procurement regulations) do not support all of the the terms and conditions of today'stypical open soruce license.  

For instance, the TrustThevote Project has run into this in several jurisdictions around the country, and caused them to engage with the very best open source licensing lawyers inthe country... the same lawyers who brought you the license for the Mozilla FireFox web browser, for instance, to develop a "procurement-friendly" variation of the general public license scheme.

Two important points about that: 1] their license is, in fact, a qualified open source license under the OSI definition and further is a derivative or child of the Mozilla Public License, also an OSI compliant license; and 2] the license the TrustTheVote Project uses (and being adopted by jurisdictions seeking to deploy open source solutions that fit within their procurement regulations) has a very important clause that says if you are not a jurisdiction subject to procurement regulations that would prevent adopting a license (like the GPL), then the source code subject to the OPL license can be licensed using the GPL license.  This alleviates the issue of licensing schemes altogether.  And for anyone wishing to download, modify, or further distribute source code from the TrustTheVote Project who is not an elections jurisdiction, the OPL need not apply; the GPL will.

For those interested in this nuance of open source software in local goverment settings, there is an explanatory white paper along with the license itself available at the OSET Foundation web site (links are not permitted here).
User Rank: Apprentice
10/8/2014 | 4:20:03 PM
Re: Pen and paper
@moarsauce123 your comments are on point for the need of a durable ballot of record.  All stakeholder elections officials engaged with the TrustTheVote Project absolutely concur that a paper ballot of record is and must be core to this solution.  The primary reason for technology involved in U.S. elections is the counting of ballots.

It is next to impossible to properly count the tens (hundreds) of millions of ballots flowing from today's elections. Counting machinery is vital for that purpose.  Hand counting requires supervision, audit, and plenty of qualified volunteers ...for those jurisdictions who perform precinct-level counting (similarly true for central count).  The challenge is human error; fatigue and many other factors.  That is why counting machinery entered the picture long ago and will likely remain.  

And the even more important reason for that (in the case of a federal election) is the time-certain nature of our electoral system: remember we must have ALL ballots counted and elections certified by a specific date due to the constitutionally mandated orderly transfer of power at 12:01 EST on a specific date in January.

As to machinery for marking ballots (vs. the very low cost of a pen (pencils cannot be used for obvious reasons ;-) to hand mark a ballot).  We must remember our fellow citizens who have physical challenges (disabilities) that require (accessible) ballot marking devices.  So here again, we see an inevitable requirement for technology to assist.

So, your point about throwing technology at voting is well taken; the approach of the TrustTheVote Project is responsible application of technology where required to facilitate verifiable, accurate, secure and transparent elections.
User Rank: Author
9/19/2014 | 10:32:32 AM
Re: Pen and paper
To @moarsauce's point, Scotland just used a 13-word paper ballot for its big vote.
User Rank: Ninja
9/19/2014 | 7:02:35 AM
Pen and paper
The easiest fix is pen and paper. There is one ballot per voter and the votes are marked on the ballot. Take a look at many European countries that use paper ballots. There is the name of the candidate or party followed by a circle. Any marking in the circle (cross, X, check mark, filling the cirlce, etc) is considered a valid vote. There is a paper trail and the public is free to witness the entire process from casting votes over transporting them to the counting place to the count itself. In most cases volunteers do all that. since every step of the voting is open to the public chances of box stuffing or intentional miscounts are slim. If there is a need to do a recount it will be very easy to do so. Tallying the votes is also rather fast with enough volunteers per precinct.

Folks, there is a solution already available and proven over decades that will once and for all fix the US voting system. The nice thing is, it is fairly inexpensive, does not require any machines, is difficult to tamper with, and voters do not need a training course in order to learn how to operate the gadgets. Why does every problem have to be 'solved' by throwing expensive technology at it?
Brent Turner
Brent Turner,
User Rank: Apprentice
9/16/2014 | 11:33:14 PM
Pooling of resources
The unmentioned element in the battle to replace the end-of-life systems is the smaller counties pooling resources to obtain a certifiable General Public License open source system.  California is positioned to do so.. and waiting for a large county to provide leadership may not be time effective. There are decades of work already in place.. and now it is a matter of political will. Microsoft lobbyists may be stepping down their political pressure.. which will free up the process.  We must remain vigilant against those attempting to nuance the solution by inserting self-empowering license schemes. 
Thomas Claburn
Thomas Claburn,
User Rank: Author
9/16/2014 | 6:03:53 PM
I'm quite ready to be cynical about the potential for fraud with electronic voting systems, but then we've had plenty of fraud and bad outcomes with traditional system. So I guess I'm open to a different take on the same old problems.
Top-Paying U.S. Cities for Data Scientists and Data Analysts
Cynthia Harvey, Freelance Journalist, InformationWeek,  11/5/2019
10 Strategic Technology Trends for 2020
Jessica Davis, Senior Editor, Enterprise Apps,  11/1/2019
Study Proposes 5 Primary Traits of Innovation Leaders
Joao-Pierre S. Ruth, Senior Writer,  11/8/2019
White Papers
Register for InformationWeek Newsletters
Current Issue
Getting Started With Emerging Technologies
Looking to help your enterprise IT team ease the stress of putting new/emerging technologies such as AI, machine learning and IoT to work for their organizations? There are a few ways to get off on the right foot. In this report we share some expert advice on how to approach some of these seemingly daunting tech challenges.
Flash Poll