I like to bring ideas to the groups I join, and I want to see them taken up: discussed, improved, maybe acted on. When an idea of mine goes nowhere, I try not to take it personally. An organization needs its members' ideas as much as members need to be heard, and a group that can't carry them loses something it depends on. An organization that can carry them is conductive: it takes a member's idea through a fair evaluation and a decision into the work that follows, and tells the member what happened. It conducts in both directions: a member's idea travels up to a decision, and a decision travels down to the members who carry it out. I argue that chapters built this way should be the template for movements that want to last, and that public interest technology groups, which are most fluent in digital tools, should lead the way.
I start with the idea of a conductive organization and why ideas matter. Then I describe what happened to one idea of mine in a large online community, and how common that is. The solution has four parts: finding the others, developing governance, implementing the Participatory Governance Framework and setting a measurable goal. I close with two short cases from tech, then with what public interest technology groups can do now.
I was looking for a way to describe organizations that move ideas quickly, from the member who has one to a decision and the work that follows. I tried several terms before I came up with conductive.
I found that Hubert Saint-Onge and Charles Armstrong had used it twenty years earlier. Saint-Onge led knowledge management work in banking and insurance, and Armstrong ran S.A. Armstrong, a manufacturer often cited as one of North America's most innovative knowledge-era companies. In (2004), they describe the successful organization as one that increases the quality and flow of knowledge, within itself and across its network of suppliers, customers and collaborators. For them, conductivity belongs to the whole organization. Its culture, structure, strategy and systems either let knowledge move or block it, and they call its structures “the custodians of conductivity.” Their concern was the customer: knowledge flowing in from the market so the company could serve it better.
The word could be misread as conducive. Conveniently for us, conductive is conducive: an organization that lets ideas travel is one where people keep offering them.
The more I read, the more I found our thinking aligned. Their four elements each have a counterpart in a civic group. Strategy, which they treat as the foundation the others align to, becomes a goal the members share. Culture is built as members find each other and learn to work together. Structure is the group's chapters and governance. Systems are the tools it uses to handle ideas.
Armstrong and Saint-Onge applied the conductive model to business and customers. I'm applying it to organizations and members.
Ideas keep an organization alive. Alan Robinson and Dean Schroeder studied more than 300 organizations for (2004), and found that the ones with more ideas per person tended to do better. Many organizations never asked for ideas. Others asked but had no way to evaluate them or act on them. In a sequel, The Idea-Driven Organization (2014), they estimate that most of an organization's potential for improvement lies with the people closest to the work, not with its managers.
Toyota is the classic case. In 1951 it started its , modeled on a suggestion program its managers had seen at a Ford plant the year before. By 1973, its 43,000 employees were submitting more than 250,000 suggestions a year, and Toyota adopted more than 70 percent of them. Few American car buyers know the system exists, though it is well known in management studies.
Most Americans know the process better from politics. Campaigns have traditionally been a bit about personality, about vision, but also about ideas. Franklin Roosevelt ran in 1932 calling for : try a method, admit it if it fails, and try another. Michael Bloomberg, who built his fortune on a financial data company, brought the same habit to City Hall. He seated his staff in an open bullpen modeled on a trading floor, so ideas could pass from desk to desk. In 2011 his administration opened the NYC Simplicity Idea Market, an online forum where city employees could post ways to save money or improve service and rank one another's ideas, and he promised to implement the best ones. Through his philanthropy, he carried the approach to other cities with the Mayors Challenge, launched in 2012, which funds the best ideas from city halls.
“Ideas are easy. Execution is everything,” the venture capitalist John Doerr has said. At a place I used to work, I recall the CEO saying on a companywide call that he was open to new ideas, but qualified it: you had to be able to implement them. Lacking a team then, I was sunk. I heard a variation of this when corresponding with a local law clinic. I'd been sharing some ideas, and then finally asked whether some incoming students could help me with them. Why wouldn't they volunteer on my ideas? The director set me quite straight, telling me roughly: ideas are great; ideas with funding are better.
Pardon my naiveté: I've long had the mindset of the open-source software world, where one person proposes an idea and others build on it, without the organizational friction. In 2002, Yochai Benkler described this as , in which large numbers of people contribute to shared projects like Wikipedia and open-source software without a manager assigning the work. The internet could gather ideas at a scale no suggestion box could match. Few civic groups have put that promise to work.
Conductivity is easy to describe, but a bit harder to find. I've long been devoted to using information technology for the betterment of society; the field is broadly called public interest technology. One newer group advocating it is All Tech Is Human (ATIH).
Last month they hosted the Responsible Tech Summit at the Times Center in Manhattan. It was easy for me to get to, so I went. It was a fast-paced conference, with several panels and more than 20 speakers, and I talked with perhaps a dozen more people, mostly in brief networking. It was the sort of conference to inspire thoughts, though not necessarily to develop ideas. Fortunately, ATIH has an online community on Slack, with 16,000 members worldwide. I joined it and posted to its general channel to thank the organizers. I also passed along a question I'd heard from my conservation friends: what can we do about the digital waste in cloud storage, like all the photos people take and never look at again?
Someone gave a response; another reacted with a lightbulb emoji. And that was that. Well, I would guess that among all those members at least someone could nudge it forward. The ATIH Slack, like many similar large online groups, seems to suffer from the bystander effect. It's like Yogi Berra's line about a popular nightspot: “Nobody goes there anymore. It's too crowded.” People post and rarely get an actionable response.
My post is one example, and I haven't measured how often this happens. Surveys suggest it is common. Gallup asks employees whether their opinions seem to count at work, and only about one in four worldwide strongly agree. In public life the gap is wider: in 2023, Pew found that 85 percent of Americans said most elected officials don't care what people like them think.
Organizations rarely report how many ideas come from their members, or what became of them. Toyota is the exception, publishing its counts as early as the 1970s.
The solution follows Saint-Onge and Armstrong's four elements: culture, structure, systems and strategy. Culture starts with finding the others. Structure comes from developing governance. Systems are chosen with the help of the Participatory Governance Framework. Strategy is a goal the group can measure.
A person with stalled ideas looks for allies. In a large community, that starts with knowing who “we” are: a shared name, and some idea of who else is in the room.
Here's what I'd like from groups: a mission statement I can remember. I'd like a member directory to search out others (a LinkedIn group can supply this). I'd like to know why people have joined the group.
Many people join a group, as I did, hoping to be recognized for what they bring. Most would settle for bandmates: people who notice their part and play alongside them. That kind of recognition happens in small groups, where members know one another by name. A chapter is that small group.
With local chapters, I could build rapport with people near me, and we could develop ideas like the cloud-storage question together. With topical chapters, such as one for environmental conservation, I could share the idea with people already working on those problems. In the big general channel, it got no traction.
Chapters do not have to be geographic. Many organizations call their topical chapters working groups. The Internet Engineering Task Force does most of its work this way: each working group has a charter, chairs and a mailing list, and it closes when its work is done. Wikimedia recognizes thematic organizations alongside its national chapters. Public interest technology needs both kinds: local chapters to work with towns, and working groups on topics like AI policy or digital waste that draw members from anywhere. Either kind starts the same way: by getting the right people into the same room.
One of the first digital tools built for finding the others was Meetup. Scott Heiferman started it in 2002, after reading Robert Putnam's Bowling Alone (2000) in the weeks after 9/11. The idea was to use the internet to get people off the internet and into a room together. That made Meetup an early model of hybrid organizing, with groups found online that met in person. Meetup groups did the first step well. They had no built-in way to elect a leader, make a decision or assign the work that followed. The organizer paid the fee and owned the group. Putnam asked at the time whether groups like these could ever match the cohesion of a church congregation.
Theda Skocpol's research explains what was missing. In (2003), she shows that America's large membership groups, from the late 1800s to the mid-1900s, were federations. Local chapters elected their own officers for short terms and sent delegates to state and national bodies. Members learned to run meetings, hold votes and influence officials. After the 1960s, those groups gave way to professionally run advocacy organizations that wanted donors more than members. Putnam counted the decline in joining. Skocpol traced it to the loss of self-governing chapters. By that measure, Meetup brought back the meeting and left out the governance.
Consider one of the longest-running service clubs, the Rotary Club. The first club met in Chicago in 1905, and the name came from rotating the meetings among members' places of business. Members share a name, Rotarians, meet weekly, and weigh their decisions against the “Four-Way Test” (four short questions about truth, fairness, goodwill and benefit to all concerned). It also has a formal path for ideas. Any club can propose a resolution to the yearly Council on Resolutions, which meets online, or a change to Rotary's rules at the Council on Legislation, held every three years, where each district's elected representative votes. Rotary has no topical focus on the issues of the day, which makes it seem a bit of an anachronism. What keeps it going is its structure and its culture.
Rotary is one of many traditional organizations that govern themselves. Their formality may also have kept newcomers away, which helps explain the appeal of lighter groups like Meetup's. A chapter that lasts needs both: the easy entry of the new and the self-government of the old. Nathan Schneider points toward this in (2024), where he argues that online communities need the democratic habits of older institutions. A digital movement that wants to last also needs a local unit that people can show up to.
Conductivity also runs downward. A national campaign or a town's request for help has to reach the members who will carry it out, and their questions have to travel back up. Skocpol's federations did both. Delegates carried local concerns to state and national meetings, and national campaigns came back down through the chapters.
Virtual organizing lends itself to simple things, like a lightbulb emoji. It's been derided as “clicktivism.” Hahrie Han, who studied why some advocacy chapters keep their members and others lose them in (2014), found that weaker chapters relied on things people could do alone. Stronger ones built relationships and gave members real responsibility. Establishing a chapter, with regular check-ins, leadership, elections and strategic plans, is how we can .
The need for governance became acute when digitally enabled groups emerged to respond to crises in the 2010s. Occupy Wall Street spread to cities across the country within weeks in 2011. Each encampment ran a general assembly, where anyone could speak and the group aimed for consensus. The assemblies gave everyone a voice, but they could run for hours and still end without a decision. Zeynep Tufekci, in (2017), argues that social media lets protests grow fast, before they build the capacity to make decisions and last.
Some of that capacity did get built. In Wellington, New Zealand, members of the local Occupy group met people at Enspiral, a network of social enterprises, and found they shared a need for a better way to decide. The tool they built, Loomio, was meant to turn open-ended assembly discussions into proposals a group could act on, and its first interface borrowed Occupy's hand signals. Ordinary advocacy needs that capacity from the start.
Elinor Ostrom, who studied how communities manage shared resources, described three levels of rules in a . Constitutional choice settles who runs the group, how leaders are chosen and who takes over when they step down; Skocpol's federations had rules at this level. Operational choice covers the day-to-day work of carrying decisions out. In between sits collective choice: how the group makes its decisions, including what happens to a member's suggestion once it is raised. Most groups, old and new, leave this level informal and handle it at the meeting. A large online community can have a clear organizer and still lose its members' ideas, because nothing at the collective-choice level catches them.
The challenge with the collective-choice level is the wide variety of software platforms that can supply it. Some come from the grassroots deliberation space, like Loomio and Decidim. Others are designed for organizational collaboration, like Asana, Notion and Trello. I was surprised that no common term described them all. I've devised the Participatory Governance Framework to assert a common model for these platforms; I've also created a scorecard to assess thirty of them.
In participatory governance, members can start a discussion, shape it into a proposal, vote on it and record the outcome. The framework's domains follow the path an idea travels in a civic group. Conductivity is how well an idea can be:
Participatory governance ensures that all ideas get a proper hearing, leading to a response or a path forward. The framework's domains run from inquiry handling to task management, so a group can see where each idea stands. Many of these platforms borrow a concept popularized by Agile: the “backlog,” where ideas sit until they're considered. Tools differ most in how well they bring those ideas back up for review, and AI is being built into many of them to resurface the ones that were forgotten.
The same tools can also give a group a picture of its own performance: how many ideas the group heard, how many it answered, how many decisions it made and how many tasks it finished.
When those numbers come up short, the tools are often part of the reason. An organization loses ideas when the format is wrong, as in a meeting that runs out of time or a channel too broad for anyone to answer. Since the 1990s, information systems researchers have called this : a tool helps only if it suits the job in front of it. Bob Sutton and Huggy Rao address this in full in their book The Friction Project (2024). A fit tool removes bad friction, which slows work for no benefit, and keeps good friction, which makes a group stop and think before it decides.
Decision making shows how this plays out. In many organizations, decisions are made in meetings, often under Robert's Rules of Order, first published in 1876. Someone makes a motion, the group discusses it, and a show of hands settles it. The harder part comes after, when someone has to record the vote accurately, file the minutes and get them to everyone who needs them. Over email, even the first part is a headache. A proposal goes out, replies arrive out of order, some members answer only the sender, and nobody can say for sure what was decided or when. Tools built for the job, like Loomio, keep the proposal, the discussion, the vote and the outcome in one place, where every member can find them later.
Groups rarely ask whether their tools fit. Many are happy to do without structure and systems in the name of freedom, perhaps with a little rebellion against the technology that fills the rest of their lives. One of the foundational readings for anybody thinking about starting a new group is Jo Freeman's “The Tyranny of Structurelessness” (1972), in which she assessed the development of the women's movement in the 1960s. Her stark lesson: groups without formal structure still had structure. It was just hidden and unaccountable.
Most volunteer groups run on email because email reaches everyone. A tool for gathering ideas or tracking tasks works only once most members use it, so the safe choice is the one everyone already has. Companies get past this because someone can require a tool, pay for it and train people on it. Volunteer groups rarely have anyone with that authority. So the choice of tools has to be a decision the group makes together, the way it sets a meeting schedule.
A shared goal sounds self-evident. Most group leaders would readily say they have a strategy and goals. What helps is a goal that can be measured, with progress that members can see. Information technology makes that easy to do in public.
Environmental conservation offers a working example. The Sierra Club has had state and local chapters for generations, and helped push through many landmark environmental achievements. Frederic Rich, in , argues that the movement stalled after its 1970s victories. Cleanup of tangible things (water, air) gave way to less tangible problems (climate change), which meant taking on the fossil fuels at the center of the world economy, and environmental activism became a proxy for partisanship. Rich advocated for a “center green” approach of compromises, leading to a focus on bipartisan support for preserving open space through land trusts.
Rich wrote this in 2016, a time of notable political rupture in the U.S. Around the same time, the research of entomologist was reaching a wide audience. After over a decade of publishing research and books arguing for a restoration of native plants, he co-founded Homegrown National Park in 2020. Its goal is 20 million acres of native plantings, about half the lawn in the country. Anyone can put their yard on a public map, sorted by state, county and zip code. The map is the measure: each yard added moves the count toward 20 million acres. It asks for real work: tearing out lawn and planting natives. The Tennessee Sierra Club then set its own targets of 2,000 participants, 2,000 native landscapes and 2,000 acres. One group supplied the goal, and another supplied the chapters. That is conductivity running downward: a national goal reached local chapters, which turned it into targets of their own.
Responsible tech faces a version of the same problem. Its leading issue today, the regulation of AI, is as abstract as climate change: hard to see in daily life, and fought over at the national level, where a local chapter has little to hold on to. What the movement lacks is its own Homegrown National Park, something people can work on where they live and count as they go.
A few days after the summit, I found a candidate goal at my town's conservation board meeting. We were looking at posters one of our members had made with AI, and someone asked, “What is our town's AI policy?” We didn't have one, and we have no technology advisory board that could propose one.
Two efforts by technologists show how this plays out. One is a qualified success, and the other is a series of frustrations in responsible tech.
Wikipedia is one of the most consequential and best-known public platforms on the internet. It is hosted by the Wikimedia Foundation, which also supports the wider Wikimedia movement of volunteer groups around the world. Wikipedia is a rare case of a founder's power passing, step by step, to : Jimmy Wales gave up his special powers over two decades and now holds one seat on the foundation's board. The movement's open user groups need only a contact person and a record of projects, and they can grow into incorporated chapters. A volunteer committee recognizes new affiliates, and a group that stops filing annual reports can lose its status. Editors make policy on the wiki itself, through public discussions that close with a written summary of the consensus. I suspect few movements have copied that last part, since the wiki takes years to learn. It remains one of the largest working examples of participatory governance.
Read against the four elements, Wikipedia holds up well. Wikipedia's strategy is built into its product, an encyclopedia anyone can watch grow. Its culture formed around the shared work of editing, and its structure gives volunteers elected bodies and chapters with clear rules for recognition. Its systems are the wiki itself, where policy is argued and recorded alongside the articles, so a decision can change the encyclopedia the same day.
Responsible tech has had a harder time building an association that lasts. Computer Professionals for Social Responsibility was incorporated in 1983 to raise concerns about computing use in military technology (at the peak of concerns about space-based lasers popularly referred to as “Star Wars”), but it faded in the Internet era, and dissolved in 2013. In announcing the end, Douglas Schuler, a longtime CPSR leader, wrote that “the age of the participatory membership organizations is over.” He pointed to hundreds or even thousands of groups now working on the same concerns, and said he would welcome a new membership organization with CPSR's big-tent reach. Today, America has about ten times as many computer professionals as it did when CPSR started, but we still lack a durable big tent.
Technologists for the Public Good tried a newer model. It formed in 2021 as a professional association for technologists serving the public, and offered networking events, mentorship and conferences to people working across government, nonprofits and academia. I joined in its final eighteen months. It is dissolving this year, and from the inside it never seemed to reach a critical mass of members.
All Tech Is Human has the numbers TPG lacked, with 16,000 people in its Slack community. It brings them together at in-person gatherings like the summit I attended, but it has no membership or chapter model, so the crowd has no structure to work through.
Read against the four elements, responsible tech has had some of the pieces, but never all four at once. CPSR had a committed culture and local chapters, but its strategy aimed outward, at the Pentagon and Congress, and its members could only wait to see whether anyone listened. TPG had a clear purpose and an online community, but too few members to build a culture of its own. ATIH has the crowd and a shared concern, with neither the structure of chapters nor a goal its members can measure. Schuler's wish for a new big-tent membership group is still waiting for an answer.
I looked at this general need in my recent essay, Recognizing the Civic Technologist, including the history and reach of groups that have tried to meet it. If there were ever a time for computer professionals for social responsibility, it's today. We need chapters.
For public interest technology, groups like All Tech Is Human and the Alliance of Civic Technologists could supply those chapters, and they would have a clear job. Code for America and its volunteer brigades were organized around building things. Much of the technology a town needs has already been built, and a lot of it can be bought off the shelf. What towns lack is people who can oversee it, people who can review a vendor contract or help draft an AI policy. At the Responsible Tech Summit, one panelist asked who in the audience was a nonprofit lobbyist, and only a few hands went up. Overseeing a town's technology takes people who know how policy gets made, and that room was short on them. Chapters would need to reach those people too.
The structure can start small. A chapter can begin as a dedicated channel in an online community that already exists, reach beyond the obvious centers like NYC, and meet in person as it grows. It can start at the state level and split into regional and county chapters, the way Rotary clubs spread. Public interest technology has the numbers for it: according to CompTIA's State of the Tech Workforce 2026, about 9.6 million Americans worked in tech in 2025, nearly one in seventeen workers.
For any movement, the steps toward a conductive organization follow the four elements:
For public interest technology, the first goal is close at hand: staffing town technology advisory boards, or creating them where none exist. A handful of places in the U.S. have technology advisory boards, including Seattle, Austin and Needham, Massachusetts. A few years ago I proposed a similar Civic Engagement Advisory Board; it was approved by the town board, but they couldn't get enough volunteers to fill it. Boards like these would benefit from a large professional organization behind them.