Morning,
As promised, here’s a deeper dive into the state of our elections systems infrastructure. I’ll start with an overview of the market, then I’ll connect that to our political system, and finally, we’ll take a look at how this is leaving all Americans (and some certainly more than others) behind. I mean that we “the people” are left behind because the truth is that our infrastructure and the market place that funds said infrastructure is broken.
There are three primary vendors of Election Systems in the US.
Dominion Voting Systems
Hart InterCivic
Together, these three companies own and control 92% of the market. These companies are all partly or entirely owned by private equity firms with some evidence that they have ownership ties to Russia and China.
I know, it’s hard to believe that the software and hardware that runs our elections could be in any way owned by our adversaries but it’s the truth.
States and counties receive government funding through a program called HAVA that was created after the debacle that ensued during the Florida vote recount in 2000. This funding was used up quickly and it’s been over a decade since states have had any funding to upgrade their systems.
If you were not convinced in 2000 that our system was broken, take a look at this article by Robert F. Kennedy Jr from 2004.
Or, look at how Republican Geoff Duncan won the election in Georgia more recently where 100’s of thousands of votes went uncounted with clear and verifiable discrepancies. He was running for Governor while simultaneously being in charge of elections in the state.. sound like a conflict of interest?
That combined with the consolidation and control of the market by three companies leaves little incentive for the companies to improve or upgrade their products.
Finally, the sales process into any state or county has a multi-year cycle with the winners greasing the wheels by purchasing old systems and lobbying efforts.
All these factors combined and it becomes extremely difficult for competitors to enter the market with a better product.
But all of this is not even the worst part of the state of the market. Because all of these vendors’ software and hardware is closed source, we have no way of knowing what the code that is running on each machine is actually doing or if it’s been tampered with.
To further complicate issues, the regulatory and political environment is making it even harder for new systems to be built.
So who does this impact?
Really it impacts every voter in America but we have seen a higher impact on voters who come from minority groups.
Despite all of these challenges, I believe there can be a business built around Open Source election system. Such a system would be modular in nature so that states and counties would have the ability to use some or all of the system to meet their needs.
Open Source Election Systems need to be
Free - to use and customize for anyone who has the skill.
Transparent - in the software development process and those who are contributing regardless if they are individuals or a company.
Verifiable - meaning the software and hardware is verified in certification, software, and outcomes.
Secure - the software and hardware is proven to be secure and that the development process and deployment process is secure as well.
The contracts as they are structured between closed source vendors and the states and counties they serve are perfect for an Open Source business model as they rely mostly on deployment, maintenance, and customization of the election system.
RedHat is perhaps the best well-known company to have an Open Source business with a similar model. This company was sold for around $34 Billion to IBM.
I’m not saying you can build a massive multi-billion dollar company ontop of Open Source election systems. But with a projected $900 million a year in spending on Election Systems, it seems like you could build a multi-hundred million dollar a year business by bringing a differentiated Open Source product to market. It will take a decent chunk of capital to develop, the right team, and the right backers (ideally, a grassroots fundraising effort filled with tiny donation).
Best of all, elections officials across the country are aware that they need new and better systems. San Francisco is leading they way. Their Elections Commission and the office of technology have been working for several years to scope out and secure funding for an AGPL v3 Licensed system.
Los Angeles also attempted to build an Open Source system but has run into two serious issues.
In its first test during the California primary, the system failed to work in many ways. Frustrating voters and elections officials. I have yet to determine what happened and am working with several technologists and local officials to see if we can help.
Though this was pitched as Open Source and funded with tens of millions of taxpayers dollars, the system has yet to be licensed. IE they have not picked a license and are just promising to make it “Open Source”. This is unacceptable. True Open Source technology must be developed in the open from the start. To regain trust and credibility, this system needs to be licensed AGPL v3 so that others can contribute to the project. This would certainly help with solving issue #1.
Should SF or LA succeed, this would pave the way for many other counties to utilize their system. Furthermore, it would create a cottage industry of local IT providers who could assist in the maintenance, customization, and deployment of election systems nationwide at a fraction of the cost of the close source vendors.
There are many reasons that after 20 years of known issues with our election systems that we still have yet to build an Open Source system that works. But despite the high barriers, we must persevere if we are ever to ensure our elections can be trusted.
I hope this finds you safe and well.
Best,
Alex
Want to reach out, hit me up on twitter @afkehaya
Below are a list of projects building components for open source election systems. It is not a complete list so if you know of others please let me know.
OSET - I am in touch with their team and am doing my due diligence to learn more. My main concerns are their funding source and their choice of open source licensing. I’ll report back on what I learn from their team. An earlier verision of this article stated that they had funding from MSFT. OSET reached out to me and informed me that is not the case! I’m speaking with their team this week and will know more then.
Voter Information Project (VIP) - I am getting help on technical due diligence on this project.
Rank Choice Voting Software - This is AGPL v3 licensed and seems promising
TAC Results Reporter (SF) - this is a project built by the Technology Advisory Council in SF. They are a group of technologists who are working to advise the Election Commission on Open Source voting. They are solid and should be followed.
Open Count - Developed by Berkely I’m still researching this one and need to dig into their whitepaper.
Voting Works - was a promising project but it’s hard to tell if it’s still active
Galois - Recently secured a $10 million DARPA contract to develop secure hardware for elections. This project is exciting and I’m hoping I can connect with their team to learn more about their progress.
Prime III - A project by Juan Gilbert that to my knowledge is the only Open Source voting tool in use. Juan is a great guy and worth consulting on anything related to this topic.
Election Gaurd - this is a project built by Microsoft that is meant to verify results. It’s tough to tell where this will go or what Microsoft’s motivations are in developing it.
Research/News:
A list of recommendation by TAC to the city of SF
https://www.salon.com/2019/08/30/probe-of-missing-georgia-votes-finds-extreme-irregularities-in-black-districts-and/
https://www.esquire.com/news-politics/politics/a24847675/geroge-election-voting-machines/
https://www.ajc.com/news/state--regional-govt--politics/mystery-missing-votes-deepens-congress-investigates-georgia/x4OTY0ylxfA0Z0Rg6wjkyN/
https://publicpolicy.wharton.upenn.edu/live/files/270-the-business-of-votin
https://www.rollcall.com/2020/04/02/new-election-security-funds-wont-come-easy-for-hard-hit-states/
https://thehill.com/homenews/house/482569-senate-gop-blocks-three-election-security-bills
https://www.propublica.org/article/the-market-for-voting-machines-is-broken-this-company-has-thrived-in-it
https://www.latimes.com/california/story/2020-03-10/election-los-angeles-county-voting-problems-dean-logan