You know when something happens and you find almost every single aspect of it incredibly frustrating? I had that this weekend with almost everybody involved in the voter registration process.
Let’s start at the start.
- EU elections are coming up (somewhat unexpectedly).
- More than perhaps prior elections, these may be quite deeply political in their own right rather than being read as a protest vote against the government of the day.
- Not enough people who can vote in these elections do vote, and sometimes that is a pipeline problem of them not being registered to vote.
- Registering to vote unless you are UK, Irish or Commonwealth citizen is a faff. You need to declare, using a bit of paper, whether you are voting in your home country or your country of residence. There are easy ways to do this, but the UK hasn’t.
So, a number of groups have started running apps that help register EU citizens; a process which involves paperwork being sent to the electoral services officer in your local authority.
Problem. Lots of them have said that they won’t accept non-ink signatures, so that the forms sent in by these apps will not be valid. There is a provision that bounced applications need to be followed up. But, this is inefficient and puts the burden on local authorities (who have received real terms cuts to budgets between 30 and 50% since the start of austerity policies). And democratic services often bear the brunt of cuts, for example in Northamptonshire (although other issues are also at play there): https://www.bbc.co.uk/news/uk-england-northamptonshire-47967416.
These sites also scoop the registration process for the “easier” category of UK citizens, rather than directly sign-posting them to gov.uk/register-to-vote
These sites have been asked to stop. https://www.theguardian.com/politics/2019/may/04/site-claiming-to-help-eu-citizens-register-to-vote-is-shut-down
The electoral commission is cautioning people against using them. https://www.electoralcommission.org.uk/i-am-a/journalist/electoral-commission-media-centre/to-keep/eu-citizens-urged-to-avoid-unofficial-electoral-registration-websites
But the people running them continued to do so for several days, despite significant risks that people will not be correctly registered in time because electoral services won’t accept their auto-sent forms in 100% of cases. This is really awful and should be a stain on their consciences to have the hubris that because they have created an app that they believe does a good thing, it should continue even when it is proved to be harmful to the public good and people’s democratic options. If you are responsible for depriving someone of a vote because of a poorly designed service whether you are in government or a civic tech volunteer, that should be a heavy burden.
Designing services for people requires the ability to know when you’re not doing it right and to be able to either iterate quickly or turn it off. Public trust is damaged by online services that do this sort of shit badly. In a period of a crisis of confidence in institutions, fucking it up is not really an option.
Designing public services is not a hobby, it is the construction of infrastructure. The reason we do it with research, iteration and time is that often there is not a margin of error. No number of people being disenfranchised is acceptable. This isn’t to say that people can’t build where government is failing: they absolutely should. Just that they need to learn the lessons of what that construction process looks like and stop just doing a shit hack day, putting a hat on it and calling it a service. Take time. Do it right, it is too important to get wrong.
On the other hand.
I am also surprised and dumbfounded that the register to vote process hasn’t been designed for these issues. I am surprised that after almost seven years, there have not been significant improvements to the process of registering. I am surprised that the obvious flaws in local government managing registration have not been addressed. I am surprised that parts of the process still have no online component and rely on paper. I am surprised that the code is now closed source and hidden so that the trust of the community in the quality of the code and the ability for oversight is hidden (https://www.digitalmarketplace.service.gov.uk/digital-outcomes-and-specialists/opportunities/9497?utm_id=20190503qa)
In the time since individual electoral registration was passed and the new service launched, there have not been serious attempts to meet common user needs around postal voting, voting outside the country (unless you’re in the armed forces), proxy voting or even being able to check your registration status. This is before we get to the problems with a lack of centralised data on candidates, polling stations, results or lists of democratic representatives.
This was an entirely avoidable situation. Government has been too secretive, too slow and too complacent as to improving the process of registering to vote. Now, inevitable startups have come along and are doing a worse job, because government has not opened the API nor improved their own offering nor applied any design thinking to the process.
The proliferation of apps also demonstrates how easy it will be for an attacker to create fake voter registration sites that are just data scrapes or are targeted at opposition groups in order to undermine registration drives. If the electoral commission cannot take them offline when they are merely cockups, what hope do we have when they are conspiracies? Research from Facebook presented at TicTEC 2019 showed that a huge number of state actors trying to interfere with the 2018 midterms in the USA used misinformation about registration (fake signups), polling day (advertising the wrong day), and “online voting” (when there was none) to suppress turnout. Maybe electoral services is something that cannot be done by people outside of government because a market is confusing and opens the door to misuse.
What next?
Government:
Design the process of registering to vote and voting to be centred around the user. This may involve changing primary legislation. At the very least, it could include a no-paper-unless-you-need-paper approach. The user need is for a seamless process that doesn’t break when you have the temerity to be an edge case. This is not optional. The user need underlying everything we do is “because democracy”.
People who made these sites
My suggestions
- Ask for help before publishing, it is not the community’s responsibility to psychically know that you’re fucking it up in secret.
- Test before publishing. Test with users. Test with civil servants receiving your forms.
- Use this as a way of applying pressure for a system that meets users’ needs, because I’m not sure that a market in voter registration technology is actually what we need. Political power is a thing that is not subverted by a hack day project. Real change takes political pressure.
