After nearly two years of development, Boulder’s first-of-its-kind electronic petitioning system is scheduled to begin serving citizen campaigns in the 2021 election season. For some, “Boulder Direct Democracy Online” (BDDO) is an exciting next-step in advancing democracy to meet modern technological potential; for others, especially those who participated in 2020’s pandemic-disrupted democratic process, BDDO presents equity concerns and generally feels too little, too late.
“We need a system that works for people,” Eric Budd, co-chair for the Bedrooms Are for People (BAFP) campaign, says. This spring, BAFP was one of many campaigns to receive incorrect petitioning guidance from the City, which prevented the placement of its citizen-initiated measure on the 2020 ballot. (See News, “What will happen to Boulder’s 2020 initiatives?” July 30.) “I’d like to be hopeful for the system, but the City hasn’t given us any reason to be hopeful.”
Boulder voters initially approved an electronic petitioning system in 2018, and ever since, City staff have been wading through the creation of a secure software platform that can handle highly sensitive information and thousands of user interactions. When petitioners like Budd asked for access to an electronic petitioning system early this spring to help gather signatures while social distancing, Boulder’s City Council determined the system in-development was not yet ready, and declined to entertain temporary alternatives, citing security concerns.
“As we know, an insecure system could wreak all kinds of havoc in an election,” explains Mayor Sam Weaver, who cast the deciding 5-4 vote against a temporary electronic petitioning system in April. The breach of private voter information, for example, or inauthentic petition endorsements could distort election outcomes.
Bringing aspects of the traditionally analog democratic process up to speed with modern technology has been an exercise in both due-diligence and patience, Weaver says. As the first U.S. municipality to vote to adopt an electronic petitioning system, Boulder is emerging at the forefront of the nation’s slow crawl toward election innovation.
“In the United States, elections have lagged behind many other fields in terms of the use of technology,” says David Kimball, co-author of Helping America Vote and member of the Massachusetts Institute of Technology (MIT)’s Voting Technology Project, a leading research effort that analyzes election technology, election administration and election reform across the country.
The lag is, in large part, due to funding, Kimball says. “The federal government generally doesn’t help fund the costs of running elections — local officials have sunk costs in their existing staff and technologies, and so often don’t have the resources to switch to something newer,” he says. “It’s moving in the direction of newer technologies, but it’s been somewhat slow and gradual.”
Some local pro-democracy advocates, however, are suspicious of delays in implementing the new BDDO system, as the process has stretched longer than anticipated. “There’s always pushback against any increase in democracy,” says Evan Ravitz, a longtime political activist in Boulder who spearheaded the original 2018 electronic petitioning initiative. “There is nothing the ruling class hates more than democracy,” he says.
Ravitz was part of a 2017 working group, appointed by the City manager and directed to analyze Boulder’s elections and campaign finance policies, which drafted the ordinance that now allows for electronic petitioning. After the initiative passed with overwhelming community support in 2018, Ravitz spent most of 2019 trying, yet failing, to communicate with City staff about the implementation of the BDDO project: its details, its hold-ups, its progress.
Ravitz grew particularly concerned when procedural elements began in early 2019 without due notice. According to Ravitz, the working group was not notified of the ordinance’s first reading to City Council in February 2019, and then only given four days to prepare for its final hearing the following month. Later that spring, Boulder received an offer for a free electronic petitioning system build by the software nonprofit MapLight, which aims to reduce money’s influence on politics and currently runs software for the California Secretary of State among other high-profile government and nonprofit clients. In Boulder, MapLight wanted to develop an open-source code for electronic petitioning so others cities could also use it.
After a visit from MapLight President Dan Newman in May 2019, the City rejected its free offer and opened an official request for proposal (RFP) process in July 2019. MapLight submitted a proposal, as did several other software companies. Before consulting the working group, in December 2019 the City contracted Arizona-based Runbeck Election Services to build Boulder’s new electronic petitioning system for $490,000, which includes $244,000 for the initial development plus three years of maintenance at $82,000 per year.
“Staff selected Runbeck principally because the company had more experience with secure election systems and they had existing relationships with several secretaries of state, including Colorado’s,” Shannon Aulabaugh, communication manager for the City, explains in an email.
But by the end of 2019, Ravitz had accumulated many more questions, and most he’s still trying to get answered. As of press time, there is no public information about the project on the City’s website.
Weaver says the process for building and implementing the BDDO system has been heavily guided by security concerns, following a necessarily cautious path. “We’ve actually been pretty methodical about planning out the testing,” he says, assuring the system will continue to be tested until complete satisfaction of City staff.
“We’re definitely, cautiously moving more and more to a digital world,” David Siegel, vice president of software development at Runbeck says. He’s watched the election industry evolve first-hand over the last few years. “The country isn’t quite ready for it to be fully there yet, because there are concerns, but we’re trying to take steps at Runbeck that move us in the appropriate direction.”
There was little for Runbeck to build off when designing BDDO, as the only other electronic endorsement system in the U.S. (in Arizona) is limited to serve only candidates seeking election, not citizen initiatives. Petitioners in Massachusetts were granted temporary permission to collect digital signatures during the pandemic, but they used the electronic signature platform DocuSign and county clerks around the state were tasked with comparing the digital signatures to voter records and confirming an individual’s endorsement. (See News, “Courts rule,” Aug. 20.)
In Boulder, Runbeck’s system will go beyond what exists in Arizona and Massachusetts. BDDO will be directly integrated with real-time voter data, allowing only authenticated registered voters to endorse petitions.
Constructing a software system able to continually update and protect private citizen data, as well as withstand thousands of unique user interactions, was a complicated task, Siegel says.
“There are cyber attacks that happen all over the place, and they’re getting more sophisticated, and so we have to get smarter around our ability to build software that thwarts that,” he says, calling it “an ongoing challenge and an ongoing goal.”
Compared to other industries that have been using technologically savvy-but-secure online systems for years — like banking, hospitals, real estate — the magnitude of an election’s impact on society is part of what sets the security bar particularly high, says Charles Stewart, director of the MIT Voting Technology Project and professor of political science.
“A lot of these (e.g. banking) transactions, first of all, usually only involve one person or one company, so if something goes wrong — like the system goes down, there’s fraud, there’s something that was wrong — it only affects one person or a limited number of people,” Stewart says. Most commerce transactions are not anonymous, he adds, whereas some election information is intended to be anonymous, and any error has the possibility of affecting thousands of people at once. “If something were to go wrong [in an election] it may be difficult if not impossible to recover.”
Thus far, Siegel says BDDO system testing, which began in August, has been “normal for a project.”
“We’re hoping this could be a game-changer if we can show that this is working great and it’s well accepted in the City of Boulder. We’re hoping that this is an opportunity for other people. I mean — think about it with COVID right now,” he says, noting the company’s sales team is currently in conversations with potential new customers who want to bring electronic petitioning to other municipalities.
Beyond security, other concerns have also influenced the BDDO project and its implementation speed, Weaver says — concerns like legal correctness, equity and voter education. “Direct democracy itself is an interesting beast, right?”
Introducing an electronic petitioning system has the potential to change the landscape of local laws and governments, and plays right into the “robust debate about whether encouraging more initiatives is a good thing for democracy,” Stewart says.
The citizen initiative process, which only 26 states including Colorado allow, is a way for constituents to amend and create new laws and city ordinances without the support of legislators, like the City Council.
“In most of the country, legislators usually regard [initiatives] as kind of going around their backs,” Stewart says. “In a lot of cases, people who make the rules for the laws about election administration are probably reluctant actually to make it easier.”
Ravitz argues, however, democracy should be as citizen-oriented and direct as possible. He points to Colorado’s Constitution, wherein citizens are guaranteed the right to adopt, amend and repeal laws. “People’s power is foremost,” Ravitz says. “And then representatives, we delegate them some of our power.”
Most localities, such as Boulder, have the authority to determine local petitioning rules — like how many signatures are needed and when they must be submitted. In theory, the rules should make the citizen initiative process easy enough so it’s accessible to anyone, but tough enough to weed out insincere campaigns.
“The whole point of this process is to show enough community support to get a vote,” Chelsea Castellano, community organizer and BAFP campaign co-chair, says. “It shouldn’t be any harder than what it is today.”
Erring on the easier side is “more democratic and allows the public to have more say on more issues that make it to the ballot,” Kimball says. This generally grants voters more agency over local laws and public policy.
Conversely, making the process more difficult helps counteract the fact that citizen initiatives can become laws and dictate City actions without going through the rigorous legislative process. In the legislature, “there’s a pretty lengthy process there in which potential laws are heard and considered, debated, amended, made better. There’s not much of that in the ballot measure process,” Kimball says.
Sometimes campaigns do work with legislators like city council members or city staff to fine-tune ballot measures and potential ordinances, however — as two campaigns have done for this year’s election. “So we have a shot at trying to avoid those unintended consequences if the petitioners are willing to hear us out — sometimes they’re not, or they’re very fixed on how they want something done,” Weaver says.
As Massachusetts experienced this spring, the opportunity to endorse petitions online increased participation rates in the endorsement process. In Colorado’s case, Kimball says, “I think electronic petitioning might make it easier to qualify measures for the ballot.”
Weaver agrees. “This will make it easier to change the City charter,” he says. “If it’s too easy to get on, and a lot of bad ideas get on and they become law, then things can happen that people don’t expect or intend … kind of like TABOR is at the state level.” TABOR, the controversial 1992 amendment, limits the amount of revenue the State can retain and spend, with the surplus monies returned to taxpayers.
“Complicated issues or complicated petitions that have a lot of elements to them are hard for voters to analyze,” Weaver continues. “So the real challenge with direct democracy, in my opinion, is that it’s hard for voters to actually sort through the ideas and figure out if they’re good or bad.”
The only solution to that at this point is to provide more information about each measure, Weaver says. Thus, voter education is the “most important” part of the BDDO system implementation, Aulabaugh explains.
“We believe that this will give access to a broader range of members of our community. It also should provide people with the ability to make more informed decisions,” she writes in an email. “With paper petitions, voters must rely on what the circulator says and the language of the petition. Having the opportunity to review a petition online gives voters the ability to research and obtain additional information before endorsing a measure.”
As it stands, however, citizen initiative campaigns will have to choose between pen-and-paper and electronic endorsement methods — they will not be allowed to use both for the same petition. Castellano worries this will “narrow the potential opportunities to reach people” and could make the petitioning process more difficult for some, while giving others with more financial acumen an advantage.
Exclusively online petitions would challenge campaigns to reach citizens that fall outside a campaign’s social network. The bulk of the near-8,000 signatures that BAFP collected this spring came from engaging with strangers out in public, says Budd, who spent hours standing in community parks and at grocery stores. Budd contends it’d be possible to use electronic tablets when petitioning out in the community, but the financial burden of staffing volunteers with tablets is “another hurdle that would give me some pause,” he says.
“Having a combo, you get a lot of benefits of both,” Castellano adds. Being able to leverage technology while also honoring the familiarity of the current process: “That would be a really positive change.”
As of press time, it’s still unclear if charter amendment campaigns like BAFP would even be allowed to use the BDDO system, as the rules guiding Boulder’s charter amendment process faced intense scrutiny this spring when City officials failed to deliver clear instructions. It remains unclear whether the City charter or the state constitution governs charter amendments; if it’s the latter (as City Council asserts), then the BDDO system would not be available for citizen initiatives seeking charter amendments, as the state constitution does not allow for use of remote, online petitions. If it’s the former (as the City attorney maintains), then BDDO could be used by all. Regardless, citizen initiative campaigns involving municipal ordinances will be granted access.
After BAFP’s traumatic petitioning experience this spring, even if allowed, Castellano says, “We’re not going to use the electronic petitioning system because we do not trust the City will execute it properly, and we also don’t trust that they’ll defend it legally.”
BDDO is anticipated to be ready by January, when the next petitioning season begins, but exact details on how campaigns can use it to track endorsements is still under discussion. As it goes into effect, Weaver says the City will continually monitor how the BDDO system affects the ballot. “If we start getting loads and loads of referred measures, we might want to go back and consider: Do we have the right signature requirement if electronic has made it much easier to get measures on the ballot?”
While wary of petition composition and inadequate voter education, Weaver remains optimistic. “To me, [BDDO] is exciting, and it will be something other cities and states can look at,” he says. “We always like to try things out that empower our folks to be engaged, as well as to make it convenient to be engaged, because that’s always a challenge.”
This article is part of a series analyzing online signature-gathering’s role in the future of direct democracy in 2020. Reporting for this series was made possible, in part, thanks to the Solutions Journalism Network. Para leerlo en español, por favor visita aquí.
