Revised and updated in October 2026
What first fascinated me about open source was not the software itself.
It was the possibility of organizing a complex undertaking around a generous
assumption: someone outside the room might have something valuable to
contribute.
In open-source software, the underlying code is available
under licenses that permit people to inspect, modify, and redistribute it.
Users do not have to remain merely users. They can examine how something works,
identify a problem, and develop an alternative. Their improvements need not
depend entirely on the willingness of the original creators to make them.
That possibility raises a question that extends well beyond
programming. If people can contribute to the systems they use, could they also
contribute more directly to the systems that govern them?
As an Argentine, I found the political implications
particularly compelling. My interest was not simply in making elections more
convenient. It was in the distance between choosing representatives and having
a meaningful role in the decisions those representatives make.
An Argentine Experiment
Pía Mancini and her colleagues gave this question a concrete
form through DemocracyOS,
an open-source platform designed to help people learn about proposals, discuss
them, and indicate how they wanted their representatives to vote. Mancini
explained the initiative in her 2014 TED talk, How to Upgrade Democracy for
the Internet Era.
They also helped establish the Partido de la Red. Its
central proposition was unusual: candidates would commit to voting in
accordance with decisions reached by citizens through an online participation
process. The elected representative would provide a connection between that
process and the legislature. The platform would not itself pass laws; the
proposal depended on a representative honoring a political commitment.
The party did not win a seat in its first electoral attempt
in 2013. That matters, because an interesting design should not be confused
with a demonstrated transformation of government. Nevertheless, the experiment
posed a worthwhile question: could representation become a more continuous
exchange, rather than a relationship renewed mainly at election time?
What interested me was less the prospect of another party
than the possibility of changing what a party was for.
More Than a Voting Button
There is, however, an important difference between putting
decisions online and opening the process through which decisions are made.
A platform could allow millions of people to vote while
leaving a small group in complete control of the agenda, the wording of each
question, and the alternatives available. Citizens would gain a button without
necessarily gaining much influence over what the button could do.
The more interesting application of open-source thinking
would begin before the vote.
Imagine a proposed change to a public assistance program.
People could support or oppose it, but they could also identify an eligibility
rule that excludes the very households it is supposed to help. Frontline
workers could flag an administrative requirement that would be impossible to
implement. Researchers could challenge a cost estimate. Someone could suggest a
revision that neither the government nor its opponents had considered.
None of these contributions would automatically settle the
issue. They could, however, improve the proposal that eventually reached a
decision.
This is where the analogy becomes useful. The public would
not be treated merely as a collection of preferences to be counted, but as a
source of knowledge that the institution itself does not possess.
Such participation would not have to start from nothing.
Public consultations, participatory budgeting, and other established methods
already provide opportunities for citizens to contribute beyond elections. The
challenge is to connect those contributions more clearly to decisions and to
explain what happened to them. The OECD’s
participation guidelines explicitly emphasize both meaningful influence and
feedback about how public input was used.
Publishing a proposal is one thing. Allowing people to help
improve it, and showing them whether they did, is another.
Democracy Is Not a Software Project
The analogy also has limits.
Open-source software is not necessarily governed
democratically. Projects can have maintainers with substantial authority,
selective contribution processes, or a single person with the final say. Making
code available does not mean giving every user an equal vote over
its development.
Nor should political participation copy those arrangements
uncritically. Technical expertise may justify greater authority over a
particular programming decision. It does not justify greater citizenship.
There is another difference. A software project can split
into competing versions. A country cannot offer a separate legal system to
everyone dissatisfied with a collective decision. Political choices bind people
who opposed them, including people who never participated in the discussion.
That makes safeguards indispensable. Minority rights should
not depend on the mood of an online majority. A highly active group should not
be mistaken for the whole population. Nor should an elected official be able to
avoid responsibility by saying that a platform supplied the answer.
The goal, in my view, should not be to eliminate
representation, expertise, or deliberation. It should be to make them more
accessible to challenge and better informed by the people affected.
A representative could still weigh evidence, negotiate
compromises, and explain a decision. What would change is the expectation that
these tasks belong inside a process the public can meaningfully enter.
Open to Whom?
The most obvious objection is unequal access. But access
means more than possessing a device.
An invitation to participate is not much of an invitation
when it assumes reliable connectivity, confidence with technical language, and
several spare hours every evening. A person working two jobs or caring for a
relative may have essential knowledge to contribute and very little time to
contribute it.
The answer cannot be that they should try harder.
A credible system would need plain-language explanations,
accessible design, assistance where necessary, and ways to participate offline.
The OECD likewise recommends accounting for digital divides and, where
possible, combining digital tools with in-person methods.
It should also allow different degrees of involvement.
Someone might follow a proposal closely because it affects their livelihood,
contribute one useful observation to another, and leave a third to
representatives whose judgment they trust.
Democracy should not require everyone to acquire a second,
unpaid job. Participation should remain optional, and representation should
still work for those who have neither the time nor the inclination to
participate directly.
But designed well, participation need not feel like work at
all. There may be something exhilarating about using twenty idle minutes to
understand a proposal, challenge an assumption, suggest an alternative, or help
improve a decision that will eventually affect your own community. The same
fragments of time we routinely surrender to scrolling, commenting, and reacting
could occasionally acquire a sense of purpose.
The challenge, then, is not to make constant participation a
condition of being represented. It is to make meaningful participation
available enough, easy enough, and rewarding enough that people may sometimes choose
it over doing nothing in particular.
Who Controls the Process?
Even an accessible platform would leave a final question:
who runs it?
Who decides which proposals appear? Who writes the
summaries? Who moderates disagreements? How are participants verified without
unnecessarily exposing their political views? What happens when an
administrator rejects a contribution or changes the rules?
These are not minor implementation details. They determine
how power operates inside the supposedly open system.
An open-source platform makes its code available for
scrutiny. That is valuable, but it does not by itself make the institution
using it accountable. A closed political process can have an open-source
website.
The real test would therefore be what citizens are able to
change. Can they introduce an overlooked problem? Challenge an assumption? Help
revise a proposal? Obtain an explanation when their contribution is rejected?
Follow a decision through to its implementation?
Without those possibilities, digital participation risks
becoming an elaborate suggestion box.
I remain drawn to the original idea, but not because I think
an app can resolve the conflicts of democratic life. It cannot remove competing
interests, unequal resources, or the need to make difficult choices.
What it might help remove is the assumption that most people
belong outside the process until someone asks for their vote.
The most valuable lesson of open source is not that everyone
must decide everything. It is that a system can be built to accept
contributions its original designers could not have anticipated.
For democracy, that would mean treating citizens not only as
the people who authorize power, but also as people capable of improving how it
is exercised.
