Voting technology: past, present and future

Election technology has a long history. When voting to grant citizenship, for example, ancient Greek voters would throw black or white stones into a clay jar. The jar would be broken open at the conclusion of voting, and the number of stones of each colour would be counted.

Organising a vote in the ancient world was logistically easier than organising a modern national election, given that the electorate was typically small and concentrated in one city, but the principles behind the ballot box (or ballot jar) at least was established. Perhaps the intervening millennia have more innovations to offer?

19^th century voting machines
In the 19^th century, American democracy was young and the industrial revolution was in full swing. Against this background, we might expect the patents of the 19^th century to provide early insights into how technology could improve our electoral systems.

An early voting machine was developed by Thomas Edison (see US90646), describing a system in which legislators would move a switch to a “yes” or a “no” position to vote for or against a motion. Indeed, several early US patents relate to systems allowing legislators to vote from their assigned desks. Here voting is in relatively small numbers by officials voting from assigned voting stations, and record keeping is often a primary motivation (rather than secrecy).

While some inventors were devising machines to replace traditional ballots, other inventors attempted to improve the security of traditional ballot boxes, to address fears of ballot stuffing and other malpractice. For example, US21684A provides a glass ballot “box”, allowing voters and observers to watch as cast ballots leave the voter’s hand and land on the pile of ballots. Observers could watch as the box fills with ballots over the course of voting to confirm the integrity of the vote. Unfortunately for the glass blowers of the mid-19^th century, and fortunately for those responsible for transporting ballot boxes, this invention does not seem to have been adopted widely. Other inventors proposed ballot boxes that track the number of ballots cast, to assist in the detection of tampering. US203525A proposes one such system.

US21684A: transparency in politics

Some developments were very much of their time. For example, the advancement of women’s suffrage at the turn of the 20^th century was often gradual and uneven (particularly in the US, where women’s enfranchisement could at some points vary by elected office and between local, state, and federal elections), and the problem of providing voting systems that accounted for the resulting complexities gained new prominence. Some inventors designed new voting machines that could quickly switch between a mode for men (with full functionality), and a mode for women (with restricted voting options), such as J. Ransome’s voting machine of 1913 US patent US1067679.

Voting mechanisms and hanging chads
During the 20^th century, it became increasingly popular in the US to use mechanical or electronic machines to cast votes and count results. However, it can be a challenge to maintain trust in such devices. Solutions of varying degrees of usefulness have been proposed. For example, in US3094276A, proposes photographing a vote count displayed on a voting machine before and after voting, to provide “incontrovertible” evidence of the result.

Technology is often introduced to reduce the cost of running an election or to reduce the time needed to count the ballots. For example US3201038A is a 1962 patent directed to what would become IBM’s “Votomatic” punch card system. Voters would express their intention by punching holes in designated parts of a ballot, and a machine would detect the holes and tally the result.

Famously, the “Hanging Chads” of the Bush/Gore US presidential election in 2000 exposed some flaws with punch card systems. This incident motivated many inventors to devise their own improvements to the voting system. US2005127179 and US7431209B2 both explicitly refer to that election.

Just a few years before the US election, perhaps having foreseen the flaws of punched card voting systems, the inventors of GB2287811B proposed an alternative system. Their patent describes a voting booth having a number of parallel passages. Each passage leads from an entrance corridor to an exit corridor and corresponds to one of the candidates (or an abstention option). A voter would walk down the passage corresponding to their chosen candidate, and a sensor would detect and register the voter’s choice. In essence, rather than making a mark, a voter would choose a path.

GB2287811B: How many roads must a man walk down?

Double voting and voter identification
“Vote early, vote often” is a phrase (probably wrongly) attributed to Al Capone. If voters can be identified when voting, this fraud can be prevented. However, a balance to be struck between accessibility and security. The use of any ID has proved controversial in recent UK elections, for example.

Concerns around double voting are well represented in the patent literature. For example, in 1869, Austin B Culver was concerned with the possibility of double voting by concealing one ballot within the fold of another as it is inserted into the ballot box. In his solution (see US96086A), voters or election officials would fold a lead band around ballots before they are placed into the ballot box, so that the presence of multiple ballots within one band could be used as an indicator of double voting.

Many patents including ID checking features have been proposed. For example, in US5878399A, the inventor proposes checking an ID card and fingerprint (or even DNA) data against a database to identify a voter before they can vote.  Meanwhile, US4373134A  proposes using the magnetic strip of an ID card to identify a voter. The patent envisages voting electronically at convenient kiosks with card readers.

Remote and online voting
Voting often takes place on a single day, with voters often being assigned a single location in which to vote. To ensure that everyone is able to vote, postal and proxy voting is commonplace. Some argue that this is a potential source of fraud. Whether this is true, it is important that voting is secure, and seen to be secure. Technology certainly has a role to play here.

In a General Electric patent dating back to 1940 (US2206702A), voters would listen to a radio broadcast. The radio broadcast would prompt voters having a particular preference to connect a designated load to the power grid. The votes would be counted by measuring the effect of these loads on the grid.

More practically for a general election, perhaps, many developments have been proposed to improve postal voting arrangements. In US4641240A, voters are presented with stickers corresponding to each candidate, which they may then fasten to their ballot. The stickers each have a candidate name and a machine-readable barcode, which are read by a vote counting machine. The ballots, along with the stickers, are posted to the voter. Meanwhile, US2009160174 proposes a “Vote by mail” arrangement having a secret vote code for additional security.

US4641240A: Stickers for voters

Online voting systems can potentially address some concerns with postal voting. A well-designed system can offer good protection against fraud, whilst offering convenience, traceability and speed.

US3525811A describes a remote voting system including a connecting system between remote voters and a polling place. Meanwhile, US2005263593 suggests the use of ATMs or the like in election voting. This builds on existing financial networks in an attempt to provide a voting system with high security and traceability.

The future?
With the ability to track changes in a way that cannot readily be tampered with, blockchain technology is arguably well suited to assisting with election security. Not surprisingly, there are many patents in this space. These include this Blockchain Technologies Corp proposal for receiving and counting votes using the blockchain and this Panasonic patent application for anonymous blockchain based voting.

Or perhaps the future lies in virtual reality? For example, US2020027297 proposes a system in which a virtual or augmented reality using voter may “see a virtual ballot, and mark selections on the ballot using gestures”.

The desire for voting systems that a secret, secure, trusted, fast, and cost-effective has led to many potential solutions, but perhaps the need for electoral systems to have the confidence of the electorate means that simple solutions that are understood by all may be the best options.

Indeed, the immediate future may place more emphasis on security and in-person voting, and less on speed and cost-effectiveness. Perhaps it is time to revive those black and white stones?