Why choose free and open source software rather than proprietary software to support digital democracy initiatives?

by mako | June 25, 2021 | News

The current health crisis has revived the debate on civic tech (civic technologies) with the deployment of the TousAntiCovid application, putting it in ethical terms. The public discussions that the application has generated have contributed to a certain awareness of the place and role ofopen source and free software within the technologies present in our daily lives.

Indeed, we remember the fears relating to the use of data or the questions of transparency that this application could raise. The CNIL then took up the matter in December 2020 in order to give its opinion on the draft decree amending the decree on data processing known as "StopCovid". The French landscape is indeed increasingly marked by these reflections, particularly in the context of the rise of civic technologies.

Recent events and social movements characterised by a certain institutional mistrust have contributed to a restructuring of the "relationship between public debate and private commitment, direct representation and new forms of democratic expression" (Marie-Laure Denis, President of the CNIL since February 2019). In this sense, civic tech has been particularly mobilised in the context of participatory approaches initiated by public institutions in order to involve citizens in the public debate.

But the use of digital tools for democratic purposes must be able to guarantee respect for transparency criteria and rights relating to personal data. This is what is at stake in the distinction between open source or free software and proprietary software. This is also the challenge of what some people call technopolitics: "the integration of (...) democratic principles at the beginning of the technical development (...) of digital tools", which translates into the inclusion of democratic principles in the platform's code.

The current situation in civic tech, marked by the transition to open source of players in the sector, has thus invited us to clarify once again the contours and challenges of the distinction between proprietary software and free and open source software.

The Open Source Politics team and its co-founders (Valentin, Virgile, Alain and Olivier) have been campaigning for more than 5 years foropen source and free software to become the rule in a French landscape that is still predominantly proprietary.

The ground gained by free and open source software on the civic tech market will be able to fuel a debate and questions that we have already started within the free software and digital commons movement.

In this respect, we are not only driven by the idea of mobilising free and open source software but also and above all by the idea of contributing to a true digital commons, i.e. a "resource produced and/or maintained collectively by a community of heterogeneous actors, and governed by rules that ensure its collective and shared character"(Labo Société Numérique).

In order to better understand this commitment, this article attempts to explain as clearly as possible the differences betweenopen source and free software and proprietary software by providing an accessible analytical framework for qualifying such projects.

Free and open source vs. proprietary software

	Free and open source software	Proprietary software
Paid user licence	No	Yes
Opening the source code	Yes, hence the name "open source".	No. The source code is closed to access.
Examples	Free Office, Firefox, Linux, Android, VLC etc.	Office package (Word, Excel etc.), Adobe Suite, etc.
Freedom to run (use) the program for any purpose.	Yes, everyone can.	No, only the owner can do this or by granting permission.
The freedom to study how the programme works.	Yes	No
The freedom to redistribute copies.	Yes	No
The freedom to improve the programme and to publish its improvements.	Yes	No
Risk of dependence on a publisher	Low if the software is mature and has a community	Strong, which can be risky if it is a small publisher
Security	Mature software is audited and can rely on many contributors	Subject to audits but the results are not necessarily made public and the results are not replicable
Reversibility possible	Yes	Complicated by the licence
Interoperability of software with each other	Easier to implement.	Made more complex.
Pooling of investments	Guaranteed and possible on a large scale if the software has a mature ecosystem of contributors.	Yes, but limited to the capabilities and will of the publisher.

Open source vs. proprietary infographic — Free and *open* source vs. proprietary software infographic

Focus on licences :

Free and open source software is governed by free or open source licences.

The well-known formula in the librarian communities generally states:

"Free as in freedom, not as in free beer

which refers to the fact that although these licences guarantee users 4 major freedoms (execute, study, modify, redistribute), this free use is conditional on having the necessary resources (time, computer knowledge, hosting) to install, configure and use the software oneself.

In the absence of such resources, it will be necessary to remunerate the work of service providers who have built up know-how and a service offer in contact with their clients. Many types of licences and models exist for free and open source software, generally divided into two broad categories:

Copyleft (the opposite of copyright) which guarantees users that the major freedoms of free software will be respected but prevents restrictions from being put in place (later closure of code, unpublished modified versions).
Example: GPL (General Public Licence)

Non-copyleft , generally referred to as permissive since they allow restrictions to be placed on modified, often more complete versions, which can be distributed under a proprietary license.
Example: the MIT license, and the open core

Proprietary software imposes a so-called proprietary user licence which sets out the conditions of access and use of the software. If it is a paying licence, the publisher is free to choose the terms and conditions it imposes (per user, per organisation, etc.).

Netflix, for example, operates on a monthly subscription basis that allows one or more users to use the service on different media (phone, TV, computer).

"When it's free, you're the product"

Another popular expression in the open source community is that many proprietary software products are free to use, such as Facebook, which makes its money by using its users' data to sell targeted advertising.

Focus on the opening of the code:

The question of the openness of the code is the criterion that gave its name toopensource. This source code can be compared to a recipe, which includes ingredients and procedures for making a dish: if you do not have this recipe, and do not have the list of ingredients to be used or the manufacturing process, it will be impossible to reproduce or modify this dish. In this sense, proprietary software does not give access to its source code, whereas open source software offers this access transparently and free of charge.

Focus on the issue of sustainability and scalability:

Free and open source software offers much greater guarantees of reversibility than proprietary software. As the source codes are public and generally well documented for most free software, it is easy to find another provider in case of bankruptcy or disagreement. If resources are available within the organisation, internalisation is quite possible via a transfer of skills.

This criterion is all the more important in the civic tech sector since it is mostly made up of SMEs and VSEs offering innovative solutions that are by nature risky, and since local authorities tend to commit themselves over one or more years.

The free and open source model avoids increased dependence on a single provider whose medium-term viability is not guaranteed.

Focus on security issues :

If any computer system is potentially exposed to security flaws, free and open source software that is sufficiently mature is generally audited as much as proprietary software, in addition to exposing its source in open access. This feature allows for a broader and more thorough audit as a plural ecosystem of contributors collectively ensures the highest level of security. To put it simply, more pairs of eyes (active contributors) can thus detect flaws and propose corrections according to a protocol dictated by the community.

While proprietary software benefits from occasional audits, their results are not always made public and are de facto not reproducible, as the source code is not open. Finally, the transparency of the patches made following the audit is not always guaranteed and sometimes requires another audit.

Pooling of investments :

While it is possible to pool investments in the context of the deployment of proprietary software, this pooling is richer in the context ofopen source since, once developed, the software is co-enhanced without geographical limits by a large and very diverse community of actors, who modify it according to the new needs of its users.

The need is then defined by the users and for the users, unlike proprietary software where development is dependent on the capabilities and willingness of the publisher to implement them.

The free and open source model allows not to be limited by the capacities of a publisher who would not be able to meet the challenges and needs of a very large institution. As was the case for the Conference for the Future of Europe platform, the Decidim code being open and free, the European Commission was able to use several development providers (including Open Source Politics) simultaneously in order to meet its numerous needs.

The fact that Decidim is open source has made it easier for the EU institutions to collaborate with each other, but also, crucially, with external service providers. By using a proprietary solution, the EU institutions would have been limited to the solution provider, whereas by using Decidim, they were able to "build a dedicated team just for this purpose".
European institutions use Decidim to enable the Conference on the Future of EuropeOSOR (Open Source Observatory)

Let's look at the open source figures again:

Conclusion

While the differences between free or open source software and proprietary software may seem trivial to end-users who are not necessarily experts in these matters, they should be at the centre of the debate when it comes to software where they act to exercise their citizenship.

The field of civic tech, which is supposed to (by definition) put current technologies at the service of increasing citizen power for better democratic functioning with more open governments, has to make choices in the tools used. These tools can be common goods owned by citizens for a civic purpose defined in a democratic way, or they can be owned by private companies that would be allowed to define the civic purposes of these tools according to an economic imperative.

However, this is not a Manichean choice either. Different software models are possible with their advantages and disadvantages. In a framework that we would like to be democratic, it is important to know the nature of the tools that we are making available to citizens, particularly by opening up the code.

In the absence of any reflection on the subject, we run the risk of seeing the proliferation of business models based on the resale or commercial use of data obtained during public consultations outside the framework openly consented to or understood by the citizen at the time of their initial participation.

These questions led us to choose Decidim to support our clients. More than free and open source software , it is a true digital commons that brings together a diverse community of actors and contributors and that today powers the platforms of 200 institutions worldwide.

We will see in a future article what criteria guided us in this choice and how to analyse and compare different free and open source software projects in a very simple way.

Citizen participation in a network society

by Pauline | June 10, 2019 | News

Citizen participation in a networked society

This article on citizen participation is the translation of the introduction of the Decidim administration guide, published in March 2010 on the occasion of the release of version 0.10 of the platform (downloadable here).

This introduction is of particular interest since it explains the vision of the founders of Decidim. You will find here the theoretical resources mobilized to build the Decidim framework. They inscribe the platform as the heir of a long intellectual tradition; however, you will see that Decidim deeply renews this tradition and updates it by considering the new challenges of the 21st century.

Photo by Marc Sendra martorell on Unsplash

The future of a networked society

Information and communication technologies (hereinafter ICTs) and associated practices are bringing about irreversible transformations in the social and political world. From the small residents' association to the most intense election campaigns, from a neighbourhood organisation or rally to the European Union, political relations are increasingly determined by the use of digital tools and technologies. It seems that the future of democratic participation and collective action will be through the development of digital platforms and hybrid processes, which renew traditional practices and combine them with digital practices (Fuchs, 2007).

This transition coincides with the decline of representative systems in recent decades (Norris, 1999; Pharr & Putnam, 2000; Tormey, 2015), which has contributed to the questioning of the legitimacy and meaning of democracy itself, reduced and often identified with this system (Crouch, 2004; Keane, 2009; Streeck, 2016). Several authors have used the term "post-democracy" to refer to the decline in the power and meaning of representative institutions, which ranges from globalization to the political disaffection and desertion of citizens (Brito Vieira and Runciman, 2008; Keane, 2009; Rosanvallon, 2011; Tormey, 2015). Various attempts to improve participation have failed to reverse this trend (Keane, 2011; Tormey, 2015).

Photo by Avel Chuklanov on Unsplash

This long-term political crisis emerged after the economic and financial crisis of 2008, and is directly related to it. Thus, millions of people mobilized against this crisis have not only demanded real democracy, they are actually experiencing and building it. The key step in this process is the 15M network movement. It is in a context of technological hypermediation that ICTs, used in the 1980s and 1990s to accelerate financial flows and globalization (Castells, 1996), have become crucial spaces and tools for a multipolar reappropriation of politics and democratic experimentation (Martinet Ros et al., 2015).

After four years of many successes and failures, new political citizen initiatives succeeded in May 2015 in taking power in the main Spanish cities, including Barcelona. In fact, they followed in the footsteps of countries such as Iceland, where the economic crisis has led to a period of citizen reappropriation of institutions and fertile democratic innovation, based on an intensive and creative use of ICTs.

Since 15M, most of the experiments aimed at introducing new forms of participatory and deliberative democracy (Barber, 1984; Habermas, 1994, 1996; Della Porta 2013) have used technology as an intermediary. As can be seen from the Icelandic case (and others, such as the Finnish example), democratization processes such as citizen mobilization and empowerment require techno-political coordination (Rodotà 1997; Martinet Ros et al., 2015) to achieve maximum depth and diversity. Technopolitics emerges from the politicization of technologies and the technological re-assembly of politics as well as from the co-development and co-production of technologies.

These techno-political deliberations and participation take different forms; digital and face-to-face practices, spaces and processes connect and mutually nourish each other on several levels. These participatory devices aim to increase the number, variety and parity of individuals taking part in the common government of the city, thereby expanding and enriching the areas, forms and periods in which they occur and helping to improve collective intelligence (Levy, 1997), capable of dealing with the complexity of contemporary urban life. Technopolitics must overcome the many limitations of what has been called "digital democracy" (Hindman, 2008) by first freeing itself from the "techno-centric" and "techno-optimist" narratives around digitally-assisted participation.

New participatory mechanisms are being built in a context full of opportunities, albeit perilous. The Government Programme for 2015 and the Municipal Action Plan (MAP) 2016-2019 established for the city of Barcelona give pride of place to participation and, more specifically, to innovation and the development of new models of participation. The MAP, whose construction has brought together thousands of people, corresponds to an equivocal social demand calling for a profound questioning of the democratic system and participation mechanisms.

However, this dynamic occurs in a context defined by : a) the social, political and economic exclusion of large parts of the population; b) increasing difficulties of access to participation resulting from the economic crisis situation; c) the crisis of legitimacy and effectiveness of the representative democracy regime and public authorities; d) the immense technological dependence on private infrastructures and services; e) a political and legislative context of opposition to direct democracy, social independence and territorial sovereignty; f) an abysmal institutional disadvantage in understanding social complexities using behavioural data analysis techniques and models that large technology companies and digital services possess.

Data monitoring and digital infrastructure for democracy and citizen participation

In the context of new configurations of information capitalism (Castells, 1996), often referred to as "data capitalism" (Lohr, 2015; Morozov, 2015) or "surveillance capitalism" (Zuboff, 2015), the new digital infrastructures of democracy run the risk of contributing to dynamics that are contrary to the principles of privacy and technological sovereignty.

Photo by Randy Colas on Unsplash

Proprietary, closed and opaque platforms geared towards the exploitation of social activity for profit act in an undemocratic manner and increasingly occupy social life. This model is particularly dangerous in view of the new democratic infrastructures and arrangements we are calling for.

Compared to the model of private and proprietary infrastructure, the model of public communes, which we believe the development of decidim.barcelona should be inspired by, is oriented towards the development of platforms whose design, ownership and organization are free, open, participatory, shared between public agents and citizens (organized or not). Through this model, not only the code of the platform but also the data it generates are managed and made available in a common and public way. The opening of all sectors to participation, the establishment of commons as a political principle (as opposed to the private sphere and even the public sphere -étatique - Laval & Dardot, 2015) seems to be a sine qua non condition for participatory mechanisms to be really functional.

Participation must therefore be recurrent: it must help to define and establish the structural conditions for its own existence and influence the design, development and management of participatory platforms, consultations and the results (i.e. data) generated in this framework.

Placed in the hands of large digital service companies, the algorithmic organization of social life and of the subject that concerns us, political participation, poses a risk to democracy and technological sovereignty that only an effort to produce common public services in the digital infrastructure sector can counter. Only platforms based on free, open, transparent, secure and common software offer sufficient guarantees when it comes to building better democracies. The democracy of the future must therefore be built with democratic infrastructures.

This conclusion is perfectly consistent with the philosophy adopted by Open Source Politics since its inception. The use of free software, which we have made the fundamental principle of our activity, puts into practice our desire to develop digital commons in the service of democracy. We have explained this choice at length in a previous article, accessible here.

The Social Contract of Decidim, a founding text

by Pauline | June 10, 2019 | News

We reproduce here, for the first time in French, the "Social Contract" of the Decidim platform, translated from Catalan through English. Any translation necessarily involves choices, which may in some places alter the original intention of the sentence, but we have tried to restore the spirit of the original as much as possible.

The choice to title this document "Social Contract" is a significant one, since this concept is one of the best known in political philosophy. Initially theorized by Grotius in the 17th century before being popularized by Hobbes, Locke and Rousseau, the social contract is a hypothesis for answering several fundamental philosophical questions. The aim is indeed to provide a conceptual framework to explain the foundation of society and to understand why human beings submit to rules to which they have not explicitly chosen to submit.

The idea behind the adoption of this notion by the founders of Decidim is therefore to assume the development of a new political functioning through the adoption of this platform. It is therefore the mark of a renewed understanding of our participation, as political individuals, in society. This renewed conception of the citizen's political weight stems directly, in the case of Decidim, from the close relationship of the project's leaders with the Indignant Movement, which explicitly wished to redesign the organization of political power in order to achieve a more open democracy.

Contract Social

Charter valuing democratic guarantees and open collaboration.

This text is the social contract that all the members of the Decidim project commit themselves on honour to respect.

The use and development of the Decidim platform, by any institution or group of any kind, implies full agreement and commitment to this Social Contract.

Free software and open content

The code of the platform, as well as that of the modules, software libraries or any other code developed for its operation and deployment will always be free and open source software, under the Affero GPLv3 license or a more recent version[https://www.gnu.org/licenses/agpl-3.0.en.html] when the code is new and under a license compatible with the one mentioned above when the code is reused.

Similarly, content, data, APIs and/or any other interface deployed for interaction with any type of user must follow open and interoperable standards (e.g. OpenID, RSS, Ostatus, etc.), always with the aim of striving for compatibility with the most widely used open standards.

In order to ensure transparency and citizen collaboration in participatory processes, content, text, graphics, fonts, audio, video, or other design elements will be published under a Creative Commons By-SA license[https://creativecommons.org/licenses/by-sa/4.0/legalcode].

The data available on the platform, especially all data that can be systematically collected using scrappers or other mass consultation techniques, will be published under Open Data Commons Open Database License[http://opendatacommons.org/licenses/odbl], in standardized and accessible formats (such as CSV, JSON, etc.) and, as soon as possible, with tools that facilitate data analysis and visualization.

Transparency, traceability and integrity

The platform and its current and future configuration, development, deployment and use must necessarily ensure and optimize the transparency, traceability and integrity of documents, proposals, debates, decisions and any other participatory object, mechanism or process.

By transparency we mean that any data related to these participatory processes and mechanisms is available for downloading, analysis and processing, always in the standards and formats most commonly used to share information (accessibility, multi-format, etc.).

The principle of transparency is a necessary condition for the control of participatory mechanisms and processes, but it should not be extended to the processing of personal data or serve as an attack on the privacy of platform participants.

We understand traceability to be the ability to easily and in detail find the history (past and future) of the development of proposals, plans, regulations or any other object of participation or decision included in a mechanism or process. The platform must at all times show how, why, by whom and with what guarantees a piece of a participatory process was rejected, approved or blocked.

By integrity we mean the authenticity of a specific content, and the assurance that it has not been manipulated or altered without this change having been clearly recorded and that it is visible and accessible. The requirement of integrity amounts to the non-manipulation of the proposals and results of participatory processes or mechanisms.

Equal opportunity and qualitative indicators

In conjunction with the guarantees defined above, the platform promises to provide equal opportunities for all persons, both with regard to the proposals and any other contributions that the platform welcomes. The platform offers equal opportunities for participation in all processes (proposals, debates, etc.): anyone can see them, discuss them, comment on them, evaluate them, treat them, without discrimination of any kind. The digital identity of the platform users will thus always be personal and non-transferable. The verification process that confers decision rights on the platform will also be unique, and it will be the responsibility of the administrative entity in charge of the platform to prevent the impersonation of a person or entity.

The platform must promote, with the aim of ensuring its democratic character, the use of qualitative indicators developed on the basis of data obtained through the various participatory processes and mechanisms and through the activity of users. The sharing of the settings of the different modules as well asopen data will be emphasized in the choice of these indicators.

Equal participation of citizens is one of the fundamental principles of any democratic system; the platform must not only ensure equality of opportunity regarding uses and functions but also access rights. Thus, the organisation in charge of the platform commits itself to act to promote access to, and support of, the platform for all citizens in the same way. Appropriate tools and resources for the platform are available to all without distinction.

Data confidentiality

The confidentiality and privacy of the personal data that individuals may provide to participate in any of the features and/or participation opportunities that the platform offers must be guaranteed at all times. Under no circumstances shall personal data be transmitted to a third party. Personal data will not be used more than what is strictly necessary in the context of user registration and improvements to navigation on the platform.

As soon as the technology of the platform makes it possible, the expression of political preferences or wishes during the decision-making process should remain inaccessible even to the administrator of the platform or the server(s) hosting it.

Accountability and Follow-up

Responding to all requests and contributions as quickly as possible must be a commitment to citizens. Another is to monitor the results of participatory processes and respond to those who request them. Finally, the last commitment is to study the integration of indicators to monitor the participatory process once it has been completed, in order to systematically evaluate its progress.

Continuous improvement and inter-institutional collaboration

Periodic evaluation mechanisms will be put in place to facilitate the improvement of the platform.

Priority will be given to collaboration and exchange of experiences between the institutions included in the project, with the aim of improving, repairing and building new developments to continuously improve the platform.

To this end, a collaborative atmosphere will be fostered for the development of improvements that aim to benefit the entire platform, allowing coordination between different actors if necessary.

Terms and conditions of use

All the points of this charter must be reproduced in the text of the license agreement that each organization that integrates Decidim into its services establishes with the users and must never be contradicted.

Open Source Politics is a company that develops participatory democracy platforms for public, private and associative actors. Contact us if you wish to engage in a consultation mechanism or a participatory budget using civic-tech tools!

Why and how do we choose to build an economic model based on digital commons?

par Virgile Deville | Juin 10, 2019 | Actualités

When we created Open Source Politics, in the heart of the democratic ferment of spring 2016, we used to conclude our presentations by noting that civic-tech had not yet proven anything and that the first challenge to ensure its progress would be to find a sustainable economic model.

Two years later, a first selection naturally took place. On the one hand, the citizens' initiatives that had an electoral horizon were at the end of their experiments; they will form the basis for sedimentation in the next iterations, with a crucial need for access to new funding. On the other hand, several companies have begun their growth phase by successfully marketing platforms and applications to public institutions and private players.

Since we are often asked about the OSP model, we have taken the time over the last few months to analyse the different approaches to our market being structured. An opportunity to reflect on our own specificities and to anticipate the long-term consequences of the political and economic choices currently being made.

Let us first point out that the search for a cost-effective model is not mandatory: for projects of an associative nature, relying essentially on voluntary and militant contributions, the call for philanthropic donations and/or public subsidies may suffice. One example is the association Regards citoyens, which regularly warns of potential abuses of civic-business.

On the other hand, for official approaches to participatory democracy to benefit from the potential of civic-tech, it is necessary to invest in the development of ever more effective tools and in professional methodological support. This is the path we have embarked on with OSP - without abandoning our associative actions for all that.

In the United States, the Knight Foundation lists eight variants but, for our part, we identify at this stage 4 major funding models for our industry: raising funds, selling data, selling licenses, selling skills. While they may all prove to be viable and lucrative in the short and medium term, these models will certainly not have the same democratic consequences in the long term.

Raising funds

This is the classic financing model for a start-up to accelerate its growth. By projecting himself on the future economic success of a company, a venture capitalist (also called venture capitalist or business angel ) will inject a lot of money in exchange for shares in a company. This cash injection allows the company to recruit new employees, invest in research and development, implement a more ambitious communication plan and stifle competition in the logic of being the dominant player in the market to eventually pocket a stake almostmonopolistique - the winner takes all.

At the end of 2016, at the world summit of the Partnership for Open Government held in Paris, we expressed our fear that French civic-tech, in a reversal of international trends, would turn away from the creation of digital commons and move almost exclusively towards the financing of proprietary software.

Our diagnosis is in the process of being carried out since some of the most visible French "civic-tech companies" have raised several million euros over the last six months. They have thus been able to double their workforce in a few months and intensify their communication, sometimes jointly, to institutions and the general public.

Raising funds is not a problem in itself, quite the contrary, but it is actually only temporary funding to accelerate the implementation of a company's real business model. Therefore, the key question is: what is the business model that has convinced public and private investors to engage with these civic-tech players?

Sell data

As early as June 2015, on the occasion of a test of the American application Brigade, which was presented as the "Tinder of Democracy", the economic potential of the political big data was perceptible. The platforms that collect our opinions in their databases in the form of responses to micro-surveys or signatures on petitions are veritable gold mines behind our backs.

Even if they would not do so today, what guarantees do these companies give us that they will not exploit this data for commercial purposes tomorrow, when the level of supply and the need for cash are too irresistible? Political decision-makers, journalists and major economic players, who are already investing fortunes in the opinion measurements carried out by polling institutes, are only waiting for this: tools that make it possible to precisely target a segment of the population in order to address them with the content they will like in terms of their political background and thus ensure the success of an election or lobbying firm.

Two concrete experiences with the discretionary power of these platforms have lifted our last suspicions.

During the recent "Digital Democracy" consultation, which we moderated and summarized for the National Assembly, a third of the overall traffic recorded on the platform came from a direct link to its own proposal that Change.org shared with 1.5 million fans on Facebook and emailed to its 500,000 users most interested in institutional issues. Logically, this proposal was by far the most popular (almost 20% of all votes cast out of a total of 1700 contributions). In a single targeted message, Change.org had more impact than a month's worth of daily communication by the National Assembly on its social networks and a dozen interviews with the President of the institution and several MPs, which were nevertheless relayed by our largest print, radio and television media! It's great for many causes that a platform like Change.org has reached such a critical mass, but such great power imposes great responsibilities.

In the context of missions to accompany local public consultations, we have also had the opportunity to broadcast "Facebook Ads". These are sponsored publications for which we were able to calibrate the audience with great precision: in exchange for a few dozen euros, we could place the invitation to a public meeting or the link to a questionnaire in front of the eyes of several thousand Facebook users living in this or that neighborhood, corresponding to this or that age group and having shown an interest in this or that subject through their likes.

If the sample is large enough, the investment is considerably more efficace - notamment with young people citoyens - que than distributing leaflets on the market or sending mail to mailboxes. The problem is that Facebook deliberately limits the scope of the messages to encourage us to add a few euros in exchange for a larger display. Petition platforms work the same way: pay 10 euros to have your petition sent directly to 1000 additional potential signatories. And so on and so forth.

The adage is now famous: "On the Internet, when it's free, we're the product."

In an "attention economy" where it is increasingly difficult to get a civic message out of the bubbles of already convinced and involved insiders, can the definition of the democratic agenda now only depend on paid filters imposed without transparency or counter-power by private platforms?

Sell licenses

A mistranslation of "free software" misleads many people who solicit us: it is not because software is free that it is free.

Without even mentioning the development of the software, using it has costs for deployment, configuration, hosting and maintenance. Conversely, once it is developed and outside of the above-mentioned costs, duplicating software has a zero marginal cost. Since the development has already been financed and carried out, anyone can benefit from it. In exchange, it is necessary to invest in the next developments, which will benefit everyone in return. Conversely, in the case of proprietary software, it is necessary to pay a license fee for a software that already exists, in order to make the initial investment profitable as in the case of a manufactured product. By the way, in the case of a dominant position tending towards monopoly, chances are that you will pay more and more since you have no alternative.

Thus the French State has paid more and more to use the same proprietary platform. Instead of increasing the skills of public administrations in the management of a basic solution and investing in its amélioration - quitte to entrust the latter to development companies privées - la public authorities agree to pay, licence after licence, for a platform of which they have no control over the source code or the evolution strategy. In a funny way, it is already a public investment, via a participation of the Caisse des Dépôts et Consignations, which partly finances the development of this solution that deprives the State of its sovereignty in terms of online participatory democracy!

Using the Software as a Service (SaaS) offer from a proven company is a comfortable choice that takes the responsibility off the decision-makers and technical teams of ministries and local authorities. According to nous - et we say it as citizens beyond the competing interests of our entreprise - un major risk of privatization of tools and skills that must instead be shared with the greatest number.

Should the State instead create its own platform or buy the rights of its preferred provider? This would be potentially just as serious, in the unforeseeable event that a government with liberticidal practices used such platforms to record the opinions of participants or to modify their contributions a posteriori. This is notably the position of our Barcelona inspirators who designed the Decidim platform of which we are partners. Virgile Deville developed this argument at the beginning of December 2017 duringa conference of the think tank Decide Together around the relationship of institutionalization or independence between civic-tech and representative democracy.

Every design and integration choice made by a technical team on a platform has, even unconsciously, an impact on users. Let's take an example: knowing the result of a vote before participating changes our behaviour. With free software, we can have a debate on whether or not to give access to this information to participants - quitte possibly to develop both options. In the case of proprietary software, a module for displaying positive, neutral or negative opinions is part of a global offer to take or leave because these decisive choices have already been arbitrated by the developers who, in the end, control the meaning of your participatory process. Code is Law.

Code is law, translated into French by Framasoft

Selling skills

What to do then? How to finance truly democratic digital tools? The process of creating a viable common good is certainly slower, but in the long run it is considerably more virtuous and resilient.

The National Assembly has just set an example: in the fall of 2017, it called upon Open Source Politics to advise it on the configuration and use of a DemocracyOS instance that its technical team learned to use and deploy on its own secure servers. She is now free to consult on DemocracyOS as much as she wants with her own resources. Another advantage is that the National Assembly has been able to use Open Source Politics to analyze contributions without us having access to the database being consulted, and therefore without us having any means of altering citizens' contributions. The National Assembly drew conclusions from its use and issued specifications for desirable functional evolutions with a horizon of several months. A body démocratique - le Bureau de l'Assemblée Nationale - a validated some of these developments and ordered their implementation. The National Assembly's investment will benefit any institution and any citizen group that wants to use it, anywhere in the world.

This is what a digital commons is. This is what a public investment in a public code should be.

publiccode.eu

What is the point of Open Source Politics if our customers can do without us as soon as the transfer of skills has taken place? In reality, this model is totally consistent with the model of selling multiple services: hosting and maintaining a tool for customers who do not have sufficient internal resources, developing new functionalities when our customers finance these improvements which will be mutualized, and finally supporting the use of these technologies through training, communication support, strategic consulting, moderation and analysis ...

Since the spring of 2016, we have broadened our scope of expertise to include the facilitation of collective intelligence workshops and automatic language processing, in order to understand all the steps of an open and modern democratic process, both online and offline. We are not selling a captive asset; we are sharing know-how. The participation platform is only a tool, part of a process that still faces sociological and cognitive barriers that exclude a large part of the population from participation. Just as platforms do not evolve in a snap of the fingers, citizen participation is not something that can be decreed overnight.

We must, to use a formula resulting from a working meeting with Nancy's town hall, give citizens power, meaning and time. Power, so that they can really decide: 2,300 Nancy parents and teachers took part in the first online vote on the evolution of school rhythms; the mayor followed their decision. Meaningful, so that they understand the processes they are taking part in: thanks to intense fieldwork, 14% of tenants voted in the first participatory budget of the RIVP - unrecord. Time, so that habit is created and trust is gained: for the past two years, Nanterre City Council has been able to conduct more than ten successive consultation campaigns to gradually broaden the profile of the citizens involved.

Our common code

We approach the challenges before us with great humility. Civic-tech platforms are still largely perfectible, and democratic practices need to be fundamentally rethought. Since day one, we have wanted to build a company that is like us and aligned with our values. We still have so much to do, but we are not alone and we know what the collective makes us capable of.

Around us, great partners are committed to ensure that the model of a truly open democracy exists and we thank them for their soutien - citons for example Entr'ouvert and its user relationship management solution Publik that we want to make interoperable with our platforms, La MedNum which supports new models for the development of innovation in partnership with the territories or Medias-Cité which carries the creation of APTIC vouchers to disseminate the skills of digital mediation actors to all those who need them.

Around us, about twenty French-speaking institutions are getting ready to join a community of users who are thinking and investing together in the future of our tools... sharing with the town halls of Barcelona, Helsinki and Turin, with the governments of Argentina or Belgium.

Around us, and we have already introduced you to some of them, researchers and activists from all over the world are mobilizing, not to build new economic rents, but to make 21st century democracy our common good.

Open Source Politics is a company that develops participative democracy for public, private and associative actors. Contact us if you wish to engage in a consultation mechanism or a participatory budget using civic-tech tools!

@ValentineChaput

@OpenSourcePol

Pourquoi la civic tech doit miser sur les communs numérique ?

par Valentin Chaput | Mai 27, 2019 | Actualités

La civic-tech française risque de se détourner de la création des biens communs numériques

Notre génération aspire à créer un monde plus collaboratif. Les enjeux de notre époque ne nous laissent de toute façon pas le choix. Nous devons changer en profondeur le fonctionnement de notre démocratie si nous ne voulons pas qu’elle soit emportée à court terme par la défiance, la colère et le renoncement. Le numérique, qui nous offre la promesse d’abaisser les barrières d’accès à l’information et à l’échange, est une partie de la réponse. Mais les modèles classiques résistent et, en croyant les dépasser, nous aggravons parfois leurs torts. OuiShare l’a observé sur l’économie dite « du partage », absorbée par la croissance fulgurante de géants comme Uber qui ont rapidement préféré la lucrativité à la transformation sociale. Nous risquons de voir exactement le même phénomène s’appliquer à la civic-tech française, que nous célébrons en grande pompe du 7 au 9 décembre dans les plus beaux palais de la République lors du sommet mondial du Partenariat pour un Gouvernement ouvert que la France préside cette année.

Nous sommes collectivement responsables d’avoir laissé grandir la confusion qui entoure la civic-tech, cet objet politique non-identifié derrière lequel nous nous sommes réfugiés avec espoir et enthousiasme. Nous avons décliné un jargon fleuri composé d’« open gov », de « hackathon », d’« open data », d’« API », de « do it yourself », de « crowdsourcing » et de « proxy voting » sur la « blockchain » qui rend nos projets littéralement incompréhensibles pour la très large majorité de la population que nous voulons toucher. Nous avons par ailleurs été piégés par nos propres définitions de la civic-tech, si englobantes qu’elles ne permettent pas la distinction entre plusieurs réalités techniques, économiques et finalement éminemment politiques.

La civic-tech concerne l’ensemble des plateformes et applications mobiles spécifiquement conçues pour renforcer l’engagement citoyen, la participation démocratique et la transparence des gouvernements. Ces solutions accompagnent tout le cycle de vie d’une politique publique, de l’idéation à l’évaluation. Il est cependant nécessaire de creuser le sujet pour obtenir une typologie plus objective des modèles et des acteurs.

Le Gouvernement ouvert repose par définition sur un espace de collaboration, un trilogue qui doit s’engager entre les institutions publiques, les structures organisées de la société civile et les citoyens dans leur diversité. Les intérêts et les moyens de chacun sont naturellement différents, parfois divergents. Les gouvernements et administrations désirent améliorer la qualité et la transparence du service public rendu à leurs usagers, et tout signe d’ouverture est une bonne communication en vue d’une réélection. Les citoyens attendent que de meilleures décisions soient prises avec eux pour améliorer concrètement leur existence. Les associations cherchent à valoriser leurs actions, accroître leur audience et leurs ressources. Le modèle économique d’une start-up du numérique est lui aussi assez limpide : il faut commencer par investir sur fonds propres ou en levant des fonds pour proposer le meilleur produit, être le dernier à survivre à la phase d’accélération et ainsi s’imposer comme un monopole de fait, quitte à racheter des concurrents en cours de route pour mieux s’imposer. Il n’y a qu’un Airbnb, qu’un Facebook, qu’un Netflix, parce que tous les autres sont morts ou marginaux. Winner takes all. La question qui se joue en ce moment en France est de savoir si nous devons soumettre la civic-tech aux mêmes modèles économiques ou si la démocratie justifie une exception.

Les institutions françaises prisonnières des logiciels propriétaires.

La diversité des initiatives civic-tech françaises qui foisonnent depuis dix-huit mois a maintenu une apparence de complémentarité. Elle a désormais été décrite sous forme de catalogues homogènes par tous nos principaux médias locaux et nationaux, qui pour la plupart n’ont pas poussé l’analyse au-delà des éléments de langage corporate. Entrons dans le détail.

Certaines plateformes sont « scalables », c’est-à-dire que le coût marginal d’un nouvel utilisateur tend vers zéro, comme l’illustre Jeremy Rifkin dans son analyse de l’économie des plateformes numériques. Ainsi, Change.org ne doit pas ré-investir de ressources — en dehors de serveurs plus importants — pour passer de dix à dix mille pétitions, de dix à dix mille signataires. Il en va de même pour l’application GOV qui veut « uberiser » les sondages grâce à une application qui lui permet de collecter les avis d’un nombre croissant d’utilisateurs sans dépenser plus d’énergie alors qu’un institut classique doit reproduire et analyser des centaines d’entretiens téléphoniques dont le coût unitaire ne varie pas. La contrepartie est la centralisation et l’uniformisation des plateformes. Facebook propose les mêmes fonctionnalités à tous ses utilisateurs. C’est un modèle qui marche pour des outils de mobilisation, dès lors que l’on considère que l’usage n’est pas différent pour un candidat de gauche ou de droite, qu’il gère une base militante de dix ou de dix mille personnes. Cela conduit Nation Builder à équiper à la fois la campagne pro-Brexit et la campagne anti-Brexit, la campagne de Jean-Luc Mélenchon comme celle de François Fillon. Ces plateformes sont des outils d’action au service des intérêts particuliers qui s’affrontent dans la vie politicienne — sans connotation négative, mais par opposition au système politique institutionnel et public — et il convient donc de les regrouper sous le terme plus précis de « pol-tech ».

Un autre pan des civic-tech, celui qui concerne la prise de décisions et leur évaluation, dépend justement de l’initiative des gouvernements eux-mêmes. Certains ont les moyens de développer eux-mêmes des outils (comme la mairie de Paris pour son budget participatif), mais la majorité fait appel à des prestataires privés. On parle alors des « gov-tech », au modèle hybride : il est important que chaque gouvernement dispose d’un outil sur-mesure et puisse garantir la sincérité et la protection des données individuelles qui sont récoltées, mais les types de participation sont récurrents — appel à projets ou idées des citoyens, consultation sur une décision publique, cartographie collaborative, budget participatif, portail d’accès aux données publiques… Les mêmes plateformes peuvent donc être dupliquées modulo une légère adaptation contextuelle. C’est ici que deux modèles entrent en concurrence : les logiciels libres contre les logiciels propriétaires.

Plusieurs entreprises françaises se sont créées sur cette opportunité. Spallian s’est partiellement reconvertie dans la vente d’applications de signalement « Tell My City ». Fluicity développe une application mobile de communication entre une municipalité et ses administrés. OpenDataSoft propose une solution intégrée pour que les collectivités créent facilement leurs portails open data — désormais une obligation légale. Cap Collectif commercialise des plateformes de consultation. Ces entreprises font de la gov-tech et à mesure que les cas d’usages se multiplient — particulièrement en période pré-électorale — leur qualité et leur rentabilité augmentent. Ils attirent des investisseurs privés qui entrent au capital ; OpenDataSoft vient par exemple de lever 5 millions d’euros pour déployer sa solution partout dans le monde. Les dernières améliorations techniques sur ces plateformes sont indéniablement intéressantes.

Le modèle propriétaire s’accompagne toutefois d’une série d’inconvénients :

Le manque de transparence pour commencer. La puissance publique n’a pas accès au code source qui fait tourner ces plateformes. Pourquoi se préoccuper de ces détails techniques — qui, avouons-le, dépassent de très loin la compréhension de la majorité des décideurs — tant que la plateforme marche ? Si le code a valeur de loi, selon la démonstration faite par le Pr. Lawrence Lessig (Code and Other Laws of Cyberspace, 1999), une plateforme numérique n’est jamais neutre. Elle est le fruit des choix techniques et idéologiques de ses concepteurs. Lorsque nous ne maîtrisons pas son code, ce sont les auteurs de ce code qui nous maîtrisent.
L’abandon de la souveraineté ensuite. Publier rétrospectivement un jeu de données issu d’une plateforme non auditable n’est pas une garantie suffisante que les données n’ont pas été manipulées. Quand bien même nous n’aurions pas de raison de douter des intentions des éditeurs actuels de ces plateformes, le fait que ces entreprises puissent être rachetées à moyen terme par d’autres acteurs est une menace que tout acteur public doit prendre en considération dès lors qu’il engage sa responsabilité dans un processus de récolte d’opinions citoyennes et de concertation démocratique.
L’absence de collaboration durable enfin. Tous les gouvernements ayant les mêmes besoins, ils représentent une manne promise à ces entreprises. Au lieu de mutualiser ces besoins, les institutions — et donc in fine les contribuables — payent et repayent chacune à leur tour des technologies existantes. Une partie des gains sont certes réinvestis par ces entreprises, mais les améliorations ne profiteront qu’aux prochains clients. L’argent public ne finance pas le développement de biens communs librement réutilisables, mais des modèles économiques classiques. Dans le modèle propriétaire, il faut que chacun le sache et le comprenne, aucune mutualisation technique n’est possible.

Code is law, translated into French by Framasoft

Le défi de faire émerger la “common-tech” en environnement fermé.

Après avoir distingué la pol-tech et la gov-tech, nous obtenons une vision plus claire des technologies de la citoyenneté stricto sensu. La capacité des citoyens à maîtriser et utiliser par eux-mêmes ces outils pour s’informer, s’organiser et prendre des décisions collectives est dans l’ADN de la civic-tech, définie aux Etats-Unis comme « the use of technology for the public good ». Peut-être faut-il évoluer vers une définition des « common-tech » pour délimiter plus précisément la création de ces communs digitaux, qui correspondent davantage à ce qui existe à l’international. Car des alternatives libres existent pour les mêmes besoins :

« Fix my Street » est une plateforme open source de signalement développée au Royaume-Uni par mySociety.org.
La nouvelle association Open Knowledge Estonia que j’ai eu le plaisir de rencontrer à Tallinn se lance dans le développement d’une application open source « 2 minutes 4 my city » qui connecte les citoyens avec les actualités et décisions locales.
DKAN est un portail d’open data open source créé aux Etats-Unis à destination des collectivités.
Democracy OS est une plateforme de consultation open source initiée en Argentine et désormais déployée en France.

La création de logiciels libres pour la démocratie est en train de se généraliser à travers le monde :

La Commission européenne impose que les logiciels qu’elle finance, comme ceux du programme D-Cent, soient open source.
L’administration Obama a ouvert le code de son application officielle de pétitions « We the People » et vient de lancer le portail code.gov qui libère le code de toutes les plateformes gouvernementales américaines.
La nouvelle ministre taïwanaise du numérique Audrey Tang a animé depuis des années les hackathons g0v.tw autour du développement de solutions open source.
Le pionnier des outils de lobby citoyen est la plateforme Meu Riodéveloppée en open source au Brésil.
Les Islandais dont nous saluons les pirates et le modèle démocratique ont créé un portail open source pour Better Reykjavik.
Les élus Podemos à Madrid ont investi dans le logiciel Consul qui est utilisé par le portail decide.madrid.es pour les concertations et le budget participatif de la capitale… et d’autres villes espagnoles, qui ont ainsi accès au même outil.

La civic-tech française est à contre-courant.

Your content goes here. Edit or remove this text inline or in the module Content settings. You can also style every aspect of this content in the module Design settings and even apply custom CSS to this text in the module Advanced settings.

code.gov, le portail qui donne accès aux logiciels développés par l’administration américaine.

Le modèle open source repose sur des licences qui définissent les conditions de libre accès, utilisation, transformation et commercialisation de plateformes qui sont codées de manière ouverte et collaborative. Les fichiers qui structurent les applications sont accessibles publiquement sur des plateformes comme GitHub ou Gitlab, et des notices vous expliquent comment déployer et configurer gratuitement des instances indépendantes que vous pouvez héberger sur vos propres serveurs et adapter à vos besoins. De là vient une incompréhension manifeste autour de l’open source : ce n’est pas parce que l’accès est gratuit que le développement l’est aussi. Le paramétrage technique, la traduction, l’ajout de fonctionnalités nécessitent du temps et des compétences de développement — parfois plus que pour une solution propriétaire qui existe déjà et dont le coup de duplication est infiniment plus faible que le prix de la licence d’exploitation que l’entreprise vous fait payer. En revanche, l’amélioration ainsi financée bénéficie à tous les acteurs de la communauté. Partout dans le monde. Ainsi, en choisissant Democracy OS pour développer le portail participez.nanterre.fr, la mairie de Nanterre a investi dans une amélioration de l’ergonomie de la plateforme qui a été réutilisée jusqu’au niveau du gouvernement argentin. La diffusion de l’open source est libre : la métropole de Reims a fait appel à un prestataire privé pour mettre en place une instance de consultation Democracy OS sans même que l’association ne soit au courant. De la même manière, n’importe quelle collectivité, n’importe quel projet associatif disposant en interne de la compréhension technique nécessaire peut utiliser Democracy OS. Il existe des dizaines d’alternatives développées à travers le monde : Discourse (Etats-Unis) pour des forums participatifs, Loomio (Nouvelle-Zélande) pour des prises de décisions adaptées aux organisations non pyramidales, Ushahidi(Kenya) pour de la cartographie collaborative, etc. Dans ces contextes, la plus-value provient de l’expertise déployée sur le terrain grâce à l’outil, et non du dangereux mirage d’un solutionnisme technologique qui prétend qu’un outil unique va tout changer.

Nous sommes convaincus qu’il y a un modèle économique pour ces common-tech. Nous sommes en train de l’expérimenter avec Open Source Politics en faisant un travail de curation et d’adaptation des meilleures plateformes libres dédiées à la démocratie. Ce modèle est probablement moins rentable pour des investisseurs à court terme, mais beaucoup plus pour les citoyens à moyen terme. Et donc pour la démocratie à long terme. Le chemin prendra nécessairement plus de temps à réaliser son plein potentiel. Les institutions préfèrent souvent le confort de la relation avec un acteur privé plutôt que la collaboration avec une communauté encore peu structurée. Mais elle existe à travers le concept de hackathon permanent que nous avons lancé début 2016 au sein de l’équipe Open Democracy Now et nous rencontrons de plus en plus de développeurs heureux de s’engager pour une civic-tech libre.

Dans son ouvrage de référence sur les nouveaux modèles de pair-à-pair, Michel Bauwens explique qu’un commun a peu de chance de triompher s’il est isolé face à des concurrents privés, mais finit toujours par l’emporter s’il s’allie avec des acteurs publics ou privés qui apportent une stabilité et une rétribution au travail de la communauté. Comme le détaille ce brillant article d’Uzbek & Rica, le défi pour la puissance publique de comprendre et de collaborer avec l’émergence des communs dépasse le cadre de la civic-tech et concerne tout le secteur de l’innovation. La responsabilité des dirigeants réunis lors du sommet mondial du Partenariat pour un Gouvernement ouvert dépasse donc largement l’exercice de communication.

Le siècle des communs, à lire chez Usbek & Rica

A ce jour, la civic-tech n’est qu’un passe-temps pour la classe moyenne urbaine désabusée par le spectacle de sa représentation politique. A de très rares exceptions près, nos initiatives ne sont pas inclusives et ne touchent pas les citoyens des quartiers populaires et des périphéries qui forment les bastions d’abstentionnistes et de votes extrêmes. La tâche est immense tant le fossé à combler est profond, tant les fractures seront longues à cicatriser. D’autres villes ont réussi à le faire, comme Medellin en Colombie, passée de plateforme de la drogue à la démocratie participative en vingt ans d’actions vertueuses. Nous manquons la cible car nous n’avons pas les moyens de passer à l’échelle. Il faut un soutien fort en faveur du développement de nouveaux outils numériques capables de se répandre de plus en plus facilement sur tout le territoire, afin d’intensifier nos pratiques démocratiques dans les écoles, les associations et les entreprises, de multiplier les consultations et les redditions de comptes transparentes, d’équiper les collectifs citoyens locaux qui sont les seuls en position d’associer les exclus. C’est la condition de la transition démocratique.

Ces outils existent déjà pour la plupart. Nous mettons à l’honneur leurs auteurs venus d’Allemagne, de Taïwan, d’Estonie ou de Malaisie lors d’une soirée de la société civile ouverte ce mardi 6 décembre. La boîte à outils du Gouvernement ouvert (ogptoolbox.org) qui est développée par Etalab depuis un an est mise en ligne lors du hackathon international au Palais de l’Elysée le 7 décembre puis au Palais d’Iéna le 8 et 9. Aspirant le contenu de nombreux référentiels internationaux, ce site donne accès à une information détaillée sur les bonnes pratiques mises en oeuvre à travers le monde.

La civic-tech française a énormément grandi depuis dix-huit mois. De nouveaux leviers sont en train de se mettre en place. Un incubateur va naître grâce au soutien d’Axelle Lemaire. Il a vocation à être hébergé à terme par le « Civic Hall » voulu par Anne Hidalgo à Paris. Tant que nous n’aurons pas de garanties sur les critères techniques et politiques exigés pour en faire partie, nous regarderons ces deux initiatives avec la vigilance qui est attendue de la société civile dans une démarche de Gouvernement ouvert. Cette posture peut paraître idéaliste à l’heure où l’innovation démocratique souffre d’une réelle précarité économique, mais si ce n’est pas pour nous rapprocher d’un idéal démocratique plus libre et plus ouvert que nous nous battons, alors à quoi cela sert-il ?