Comprendre la révolution numérique

Accueil > Rubriques > Politique > Civil society electronic democracy projects : three case studies

Civil society electronic democracy projects : three case studies

lundi 9 juillet 2012, par Sylvain Firer-Blaess

Homo Numericus suit les travaux de Sylvain Firer-Blaess depuis plusieurs années. Celui-ci s’intéresse à la dimension politique des dispositifs de discussion et d’édition en ligne. Nous avons eu le plaisir en 2007, d’éditer une partie de son mémoire de master qui présentait une analyse particulièrement éclairante de l’ingénierie éditoriale sur laquelle repose l’encyclopédie contributive Wikipédia. Sylvain y posait la question, passionnante, des modèles politiques qui s’y déploient. Quelques années plus tard, il poursuivait son enquête en s’interrogeant sur la possibilité pour Wikipédia de devenir à son tour un modèle politique pour une démocratie électronique. Voici maintenant qu’il nous propose un nouvel essai rédigé dans le cadre de sa thèse, moins spéculatif et plus factuel, sur les plateformes de démocratie électronique qui fleurissent ici et là, en particulier dans le sillage des partis pirates, singulièrement en Allemagne où cette formation politique suit une courbe ascensionnelle extraordinaire.

Cet article est rédigé en anglais. Merci à ceux qui seraient intéressés pour en proposer une version française, de bien vouloir contacter l’auteur.

This paper concerns the development of software designed for mass decision-making processes ; constructed not by a State or a corporation, but by groups of people who come from the so-called “civil society”.

The idea of a possible "electronic democracy" is as old as the birth of the Internet itself. The possibility of an electronic Agora, where citizens could debate and vote, is very often included in the idea of modern representative democracy moving toward a more direct democracy. This more direct democracy can be called "participatory democracy", or "strong democracy" from the model described by Barber in his book of the same name (1984). So far, previous attempts at electronic democracy have not met with great success (Coleman et al., 2009). Its realisation is still today at an embryonic state, limited by technical problems (e.g. how to assign a completely secure identity to every citizen[not clear thés means username] ; or how to protect data) as well as by a significant lack of political will from national governments.

Recently, however, a constellation of electronic democracy software projects, rooted in civil society, has appeared. The aim of these projects is to promote deliberative democracy by providing the tools to enable it, and by setting a direct example. If they are not state-supported, a possible success of their inner organization and outcomes could show that such an option for governance is not beyond conception ; they could serve as an example for future legal implementations. Current key projects are the Argentinian Eudemocracia, the British debatewise and virtual parliament, the Canadian Openpolitics the Italian Telematics Freedom Foundation, the American WhiteHouse2, Votetocracy, DemocracyLab, and the transnational Vilfredo.

In this paper we explore three of these projects : the Germans Adhocracy, Votorola and the Canado-German Liquid Feedback. Since the intent of this article is mainly descriptive we will not employ theoretical tools. The gathering of information has been conducted through Skype interviews of one member of each project respectively. Each interview lasted for around an hour. The names of the interviewees are not disclosed in order to assure anonymity. Other sources such as the web-pages of the projects, and press coverage, were also browsed in order to acquire a more objective point of view. The projects studied and the persons interviewed were not meant at first to be of the same nation, in fact we chose the two last projects after asking the first interviewee what were the most promising other projects he had in mind. This bias can give the appearance of a certain homogeneity in the characteristics of these project that might not reflect the truth of other electronic democracy projects. [This should be checked in subsequent studies].

We describe each project one by one, and then draw some conclusions. The descriptions follow the same structure : first a part on the origins, participants and organization of the project ; second we examine the ideology, or political philosophy, behind it ; then we draw the functionalities of the software ; then we talk about the experiments and uses that have been realized with the software ; and finally we finish with a general review of the project, and the plans of the project’s team for the future. In our concluding chapter we will draw some similarities between the three projects and conduct a diagnosis of their development.


Adhocracy is a web-based interface software developed by the German Liquid Democracy Foundation (Liquid Democracy eingetragener Verein, or Liquid Democracy e.V.). The Foundation has the goal of promoting and implementing the concept of Direct Parliamentarism. Adhocracy has become the most successful project of our three case studies since it is currently being tried out by the German Bundestag.

1.1. Origin, participants and organization

The Adhocracy project comes from a German bachelor student in media, Friedrich Lindenberg, who began to work on this project in summer 2009. Friedrich is today working for the Open Knowledge Foundation and is still the main developer of Adhocracy. In September 2009 Friedrich met people in a meeting of the Chaos Computer Club in Berlin who had been working on a similar project since summer 2006. They knew one another through the German Pirate Party, but later disaffiliated their work from the party. Friedrich and the Berlin group decided to work together and to fuse the two projects in one in December 2009. To support the project they created an official foundation called Liquid Democracy in the summer of 2009 when the first version of the program was released.

The people working on the adhocracy software are a mix of political scientists, sociologists and programmers. Most of them are students, academics and software engineers in their late twenties or early thirties. There are 27 members in the Foundation, but the most active member count is 5-7 individuals. This core group do not collaborate with other electronic democracy projects. The members work without pay and even paid for the hardware from their own pockets. In June 2011 they received some funding from the ISPRAT Foundation (Interdisciplinäre Studien zu Politik, Recht, Administration und Technologie e.V.), a foundation that “promotes the modernization of public administration through the use of information and communication technology” [1]. They will also receive support from the Internet und Gesellschaft Co:llaboratory group [2], a platform of Internet experts, who will develop a new user interface for Adhocracy.

1.2. Political Philosophy

The political concept at the origin of the project is “direct parliamentarism” (“Direkter Parlamentarismus”), which has been constructed by the members of the Foundation themselves. For the Liquid Democracy Foundation, current national representative democracies are not working properly, and they should be perfected with the addition of mechanisms of direct democracy. The recent progresses in technologies of information and communication (TIC) and the general increase of the level of education now permit the implementation of such mechanisms. Therefore, the aim of the foundation is to materialize this possibility of a more direct democracy by the creation of a software enabling direct democracy mechanisms, and by lobbying for its implementation. Interestingly enough, the foundation does not support the idea of a pure direct democracy. On the contrary they opt for a representative democracy that would be strengthened by the addition of some levels of electronic direct deliberation. Their claims against a pure electronic direct democracy are that there is the need for checks and balances in a governing system (one cannot leave all the power to the citizens), because software is fragile and can be easily pirated, and because the Internet is not accessible to all.

Direct parliamentarism is a model of democracy that “combines the advantages of a parliamentary system with the possibilities of direct democracy” [3]. Current direct democratic systems such as referendums permit every citizen to vote but not contribute to the writing of the propositions. Parliamentarism enables a debate-based decision-making process but that is only open to the citizen’s representatives. Under direct parliamentarism, every citizen could decide whether or not to participate in debating and shaping the norms. They could also vote directly or decide to delegate their votes to a person considered as more competent. As it is unlikely that every citizen can focus on all political matters and all projects of norms, a system of vote delegation in Adhocracy permits to delegates one vote on one subject matter to someone else. For instance, one can delegate one’s voting on matters of ecology to Greenpeace, while delegating one’s votes on cultural matter to a different party or NGO. One can at any time re-take control of its vote on any subject matter.

Direct parliamentarism is not only applicable to the State but to every organization. The positive outcomes of the implementation of such model are manifold. Apart from enabling a “democratization of democracy” by allowing more people to discuss and construct norms, it would also severely limit the power of special interest lobbying, since the power to vote would be dissolved in many more voters than the current number of representatives allowed to vote in current representative systems.

1.3. Software Functionalities

The Adhocracy software enables and facilitates mass decision making processes. It is accessible through a web-based interface. On this web-based interface, there are two kinds of pages, or ’places’ : the norms pages and the proposal pages. An organization (a political party for instance) can add each of its norms in a dedicated page of the program. Anyone who has subscribed to the software can also add a proposal page, that can be discussed in common. Each proposal page is structured in the same manner : first a title, second a short description of the proposal, third a place for discussion. The proposal can be linked with already existing norms that are related to the topic, and one can write alternative versions of these norms that would fit the proposal. People can discuss different parts of the system, like the proposal descriptions, the norm pages, and the alternative versions of norms. People can rate every comment written, which permits an easier access to the most interesting comments. As we have seen before people can delegate their vote on a specific subject to another user, or on a specific area covering several subjects. Proposals can be constructed collectively thanks to a wiki system.


The German Bundestag agreed to use Adhocracy as a test project in the Bundestag’s commission for Internet and Digital Society (I&DS), that was created in March 2010 [4]. This commission will have to present recommendations of law-making to the Bundestag on the summer of 2011. There was as of May 2011 around 1500 users who had subscribed to the program. Adhocracy is also used by the socialist party Die Linke, and by the city hall of Munich. As of July 2011, two new important organizations were going to implement Adhocracy : the Sozialdemokratische Partei Deutschlands (SPD), and the federal Youth Council [5].

1.5. Review and plans for the future.

Looking at the past two years, from the release of the first version of Adhocracy to now, my interviewee thinks that the project has turned out to be very successful, as the programming has been done quite rapidly and as it is now used by several organizations, including on the level of the State.

Some journalists have criticized the rather low level of citizen participation within Adhocracy’s implementation in the commission for I&DS, with only 1500 registered users. Some have explained this lukewarm result by the low publicity that has been given to this implementation [6]. Our interviewee for the Liquid Feedback project, gives another explanation : people do not spend time on Bundestag’s Adhocracy because they believe that it does not effect real decision making. Indeed, the Bundestag did not want to implement a mechanism of identity verification, with the result that users can potentially open several accounts and vote several times. This seriously undermines the legitimacy of any recommendations that would come from the Adhocracy system. In fact, the Bundestag did want to use this implementation of Adhocracy as a test, but did not want to allow it to make real decisions, and was therefore not at all considering it as a serious materialization of participatory democracy.

Still, the Liquid Democracy Foundation benefited from a fair media coverage because of the experiment of the Bundestag, and its members have been invited to many conferences to present their project. For the future, they hope that more and more organizations will use Adhocracy. Their main technical problem currently is the user interface, as they are struggling to make it more user-friendly. The next step for the foundation will be to internationalize : the documentation and parts of the program are still in German, which will require translation. My interviewee is very optimistic and think that the project will become more widespread during the next years.


Votorola is an electronic democracy software which has the specificity to regroup users in “rooms” of 20 places for debates, and connects these rooms through a process of vote delegation. Votorola is the least successful project of our case studies, mainly because of the lack of developers and because of the complexity of its design.

2.1. Origins, participants and organization.

Michael Allan, a 46 years old Canadian programmer, began work on Votorola in 2006. Michael gradually found people interested in his project and who were eager to work with him. There are today three to four core workers. The person I interviewed is 25, has a background in philosophy but not programming, and began to collaborate with Michael three years ago. They met through the Internet while they were working on a common former electronic democracy project, no longer active, called metascore. With so few people actively involved in the project, the organizational structure is loose if not non-existent : “everybody does what he likes most”. The group has not formed any legal association either. They have not benefited from any external founding.

2.2. Political philosophy

Similar to the view of the Liquid Democracy Foundation, for my interviewee the rationale behind Votorola is that modern democracy today takes the form of a lower form of democracy, namely representative democracy. The purer and better form of democracy, i.e. direct democracy, is out of sight as large number of citizens within a nation-state makes it unmanageable. Then the emergence of the Internet now permits the realization of a more direct democracy as it allows the management of mass decision-making processes. However, like the Liquid Democracy Foundation, the perfect State democracy could not go without the election of representatives. However, my interviewee also expressed the same thoughts as the two other interviewees, that one of the reasons not to rely solely on electronic democracy is that informatics is unsecured, and that relying only on it could endanger democracy.

The members of the Votorola project do not have a homogeneous ideological background when it comes to the construction of a better democracy. Michael Allan was strongly influenced by Habermas (which explains the specificities of the software as we will see later), while mu interviewee took conscience of the importance of bettering democracy through spirituality and meditation. He claims that if political power was really in the hands of the people, then big issues such as pollution and global warming would be solved much more rapidly and easily.

2.3. Software functionalities

Votorola offers the same basic functionalities of voting, drafting and discussion as Adhocracy. The big difference resides in the design of discussions and the functionality of vote delegation. While in Adhocracy all can delegate to all with no restrictions, in Votorola one user cannot receive more than twenty direct delegations. This does not prevent some voters from being responsible for more votes by the play of the several degrees of delegations (when someone delegates his vote to someone who also delegates his vote to another person). This limitation is the materialization of Michael’s idea of communicative delegation, against classical delegation. For the founder of Votorola, an electronic democracy software must permit rational debates between citizens. But such debates cannot happen if people are too numerous. My interviewee criticizes for instance the current functioning of Adhocracy in the Bundestag’s commission, because each thread of debates rapidly became too long to follow, and as a result this discourages people from participating. Therefore for Votorola’s members, the solution to mass debate is to prevent the agents from writing through the Internet in one big forum, but to divide the citizens into small “rooms” of debate which cannot exceed twenty agents. These rooms are constituted through the delegating mechanism, so that every person who has directly delegated their votes to the same person for a given topic are regrouped into the same discussion room (delegate included). This should permit the realization of rational debates, but should also ensure that everyone delegating a vote would end up with the same position on a given topic. If a debate does not end with the creation of a consensus, then the members are supposed to rearrange their delegations in order to find a group with whom they agree. One downside of this system is that debates therefore take place between persons who already agree on a general idea, so that polarized positions are very unlikely to meet.

2.4. Experiments

The first prototype (alpha) version of Votorola was released in June 2011, therefore no real experiments have been conducted yet. However a beta version of Votorola had been used to help the construction of the site metagovernement [7], a common information and communication platform for all electronic democracy projects. The members of the Votorola project consider that the experiment was successful.

2.5 Review and plans for the future.

For the future, the Votorola team wants to test the current software in a small-scale organization that is still to be found.

My interviewee thinks that the long time that separated the birth of the project from the release of the first alpha version of the software is partly explained by the lack of developers working on the project. He is extremely frustrated that the people willing to work on electric democracy projects are atomized between more than a dozen of different projects and software (one lists 19 different projects on the metagovernment page [8]). For my interviewee strong individualism and strong attachment to individual projects explain this atomization : Developers are often unwilling to give up their own project in order to merge into a bigger one. Also the development of an electronic democracy software is open to many choices and options, and many developers think their method is the good one, and are reluctant to make the concessions required for the building of a common project.

3. Liquid Feedback

Liquid Feedback is the electronic democracy software of the German Pirate Party, and the members are trying to implement it for other political parties. It is more focused on proposition making than debates compared to Adhocracy.

3.1. Origins, participants and organization

My interviewee is 44 years old and works for an IT company. He and three of his colleagues (one of them in his forties, the two others in their late twenties) discovered the idea of liquid democracy in June 2009, a concept created by the German pirate party. At the time the Berlin Pirate Party (a local chapter of the German pirate party) was wishing to implement its idea of liquid democracy to its internal organization, and my interviewee and his colleagues decided to help them. They tried to find existing software and realized that none was meeting the needs of the party, so they decided to create an original software. In January 2010, the software was ready to run a beta test for a party’s convention. The test was successful and two months later the party put the use of Liquid Feedback in its by-laws.

Currently, two people are working on further developments of the software, with up to ten collaborators working from time to time with them. A larger group including beta testers and persons giving ideas and making suggestions contains around 50 persons.

The group is working alone, but have many informal connections with other electronic democracy projects, and notably with the two other projects we have described earlier. Members talked a lot with the Votorola people, but they thought that the delegation tree system was too complex to work. They also discussed with the people of Adhocracy but were not agreeing with some features of the system, Adhocracy being more centered on discussions than Liquid Feedback.

3.2. Political philosophy

My interviewee has the same basic idea as the others, that what they are trying to achieve is a synthesis of representative and direct democracy. However, Liquid Feedback is primarily designed for parties and other organizations of the same scale. Theoretically it could be implemented at the State democracy level, but my interviewee thinks that replacing the existing representative democracy by an electronic direct democracy now would be too early a move, as the general population would not be ready to take care of the political matters of a country.

Also, other difficulties would challenge such an implementation : First, voting by way of computers has been ruled out by the German Federal supreme court. Second, the need for identifying citizens, to link each electronic account to one citizen, in order to prevent on individual from voting several times, requires some work from the State. This problem has prevented Liquid Feedback experimentation on the national level. Indeed the project had been approached by the Bundestag for the experiment concerning the committee for IT and Society described earlier, and the members decided that they would agree to participate in the experiment if this matter of identification was dealt with. The Bundestag, as we know, was not interested in fixing this problem because they didn’t want to allow the system to arrange real decisions. So the liquid feedback project turned down the Bundestag’s proposition. Third, the need for anonymous voting, a pillar of modern representative democracy, is in contradiction with the need for security : anonymous electronic voting would be easier to hack and to skew. Liquid Feedback chose security against anonymity so that votes and comments are signed and public, and so that anyone can view the votes of others. This would obviously not be accepted in the current system of State democracy. For my interviewee, one of the solutions would be to give to each citizen a nickname, which would relatively protect anonymity as well as permitting reliability. However there is still a chance that someone discover whose nickname belong to whom, therefore the anonymity problem is not completely solved with this solution.

3.3. Software functionalities

Liquid feedback presents the general features of an electronic democracy software : proposition, discussion, voting, and delegation. Delegations can be given for every decision, for a specific area, or a specific issue.

Liquid feedback includes an original process of decision-making called “qualified feedback”. When someone presents a proposition, the users have the option to say that they will agree and vote for this proposition if it is changed in such and such way. Other persons agreeing with these possible changes can opt in this proposition change, which makes visible how popular proposition changes are. If the changes are not made, people have the possibility to write a counter proposal that will be linked to the first one.

Any propositions concerning the same topic is grouped into an “issue page”, which is linked to a discussion page. Any new issue must be approved by at least 10% of the people registered in the general topic embedding the issue (form instance “economic policy” or “cultural policy”), otherwise the issue page will disappear after a certain time. In order to deal with all the propositions in a same issue, the liquid feedback team has added a preferential system that permits voters to list these propositions in a preferential vote : if the proposition I put first on the list was not accepted then my vote will come automatically to my second choice.

3.4. Experiments

Today Liquid Feedback is completely implemented within the Berlin Pirate Party, which had no strong structures of power and hierarchy before the software’s implementation. The software is currently evaluated by other regional chapters of the party and by the federal German pirate party, but its implementation was met with much more resistance on this level. There is indeed strong resistance from some board members (the people with power in the party) to implement it, even if more than 80% of the members of the federal party voted in favor of Liquid Feedback’s implementation in May 2010. for my interviewee, the people in power are lobbying against it in order to keep their political importance within the party. Their claim is that such a software is a threat to data protection and privacy, but a survey showed that it was not a matter of concern for the great majority of the party members. They also claim that the board members have been elected to rule, and that therefore they should retain their autonomy on the decision-making processes without being subjected to the “dictatorship of the inactive members.”

Liquid Democracy is under an MIT (Massachusetts Institute of Technology) license, so anyone can use it. Apart from the German Pirate Party, some sections of the pirate party in Austria and is Brazil are using it. Several chapters of the German green party are interested in it, as well as the social-democrats, and the Democrats in California, but nothing has been implemented there yet.
3.5. Review and plans for the future
My interviewee and his friends joined the Pirate Party in June 2009 but left it in April 2011 because of the tensions around the implementation of the software on the federal level (they were getting involved too much in power plays), and because they want to offer Liquid Feedback to organizations beyond the Pirate Party.
Overall, the coding wasn’t hard to construct. Databases are already coded using MySQL. The politics were much more complicated. While it worked very fine and fast with the Berlin Pirate Party, as we have seen the project met much more resistance for the federal Pirate Party.
For the future, the developers want to review the use of Liquid Feedback and gather advice in order to make it better. Also a better interface is being developed by a group of people from southern Germany’s Pirate Party. They hope that more organizations will use it, and for the long term, they hope for a State democracy implementation in what they call interactive democracy, where the citizens and their representatives would be able to discuss with one another.

4. Results

In this final part we point out some similarities between the three projects and discuss the issue of atomization.
4.1. Similarities
First we can see some similarities in the origins of the three projects and about their members. All of them have been initiated by rather young, politically motivated men [9], working benevolently in their spare time. Two distinct generations work on these project ; the majority is composed of late twenties / early thirties years old persons, young enough to be aware of the potentialities of the Internet, and old enough to have the skills necessary to develop such software. A minority is composed of early to mid forties years old persons, all of them professional software engineers, and can be considered as the forerunners for putting into practice the political possibilities of new technologies of communication.
One non-intuitive result is the presence of non-programmers at the core of two of the projects, Adhocracy and Votorola. Adhocracy and Votorola has social scientists and/or political philosophers who helped building the concept as well as the design of the software. The non-programmers can also help with managerial tasks such as public relations. The three projects also have a “periphery” of occasional contributors testing the software and reporting bugs who can also be non-programmers.
The three projects share the same ideal in terms of political philosophy. The ideological justification of each project is based on the idea that it would offer a synthesis between representative and direct democracy. Modern direct democracy techniques (such as referendums) permit everyone to vote while making it impossible for everyone to construct the proposal and to debate with one another, because of the too high number of citizens. Modern representative democracy permits voting, proposition making and debates, but is only accessible to a very limited number of citizens who are the representatives. The electronic democracy proposed by our studied projects would enables each citizen to participate in the three political actions of proposal-making, debates and votes.
Finally, each of the interviewees adopts a “realistic” stance to the question of a complete national electronic direct democracy. For each of them, to replace the current representative democracy by such a system is not possible nor desirable. For my Adhocracy interviewee, a State needs checks and balances, and giving all the power to the people would be dangerous. To my Liquid Feedback interviewee, the western population is not mature enough and do not have enough time to bear alone the weight of national political decisions. Finally, each of the participants also rule out the possibility of a pure electronic direct democracy because of technical concerns : an electronic system is too easily hacked.
4.2. The issue of atomization
The issue of the atomization of the electronic democracy movement, that is of the presence of multiple small-scale projects, is a critical one. The discussion platform [10], which has the function of regrouping all the electronic democracy projects, lists 19 different ones. None of them are exactly the same as each has its own specificity. We need to know whether this distribution of the work force is hampering the progress of the idea and of the implementation of electronic democracy. For the Votorola interviewee, member of the least successful studied project in terms of experimentation, this atomization through many different project is a huge waste of energy and is detrimental to the realization of electronic democracy. For our two other interviewees, on the contrary, this diversity is beneficial as 1. the resulting competition gives to the projects’ members an incentive to work harder and better and 2. it permits to offer different services that would apply for different situations.
In some Free Software projects, such as the different versions of the Linux OS, diversification of software with the same general aim but different designs has also given the user more choices, but the development is usually slower and more prone to bugs than if there was only one project. However, this possibility of having only one project is impossible as the development of such software is dependent on the voluntary work of the developers who will not work on a project with which they don’t agree. But the differentiation in electronic democracy projects is quite different from the differentiation in the Free Software movement. Indeed, the later is most of the time a mechanism of “forking”, where a group of programmers split after a certain time for reasons internal to the organization, and where the forking projects born from the split retain the same original code of the mother project (Rossy 2004 : 16f). On the contrary, electronic democracy software have been developed in parallel, and do not share any code with one another.
What are the causes of such atomization ? One hypothesis is that there is a lack of communication between people with the same ideas. However we discovered that the members of the projects were aware of the existence of one another. As all of the three projects have been developed in Germany, it will be interesting to know in subsequent studies if members of projects in a given country are aware of projects in other countries. So it seems that the division does not come from a lack of communication but rather because of an irrevocable disagreement about what should be an electronic democracy software, about its design and functionalities. For my Votorola interviewee, strong individualism and strong attachment to individual projects in the population of software engineers is another explanation. Finally, the development of an electronic democracy software is open to many choices and options. Since it has never been done before, one doesn’t really know which design is best. The multiplication of implementations should give to the electronic democracy movement the experience it needs to figure out the right design.


The programming of electronic democracy software is not a difficult task : the code is not that challenging to write for the needed functionalities of the software, and developers can use already developed free software to complete their own ; for instance LiquidFeedback used the free software MySQL for the running of its databases. More difficult, in fact, is the implementation of the software within organizations. Votorola has yet to find an organization to implement and test the software. Liquid Feedback is facing strong resistance from the people in power within the federal German Pirate Party for its implementation. Adhocracy is becoming quite successful in its implementations, but the Bundestag experiment has been executed at the cost of the disappearance of the delegation functionality and the absence of identity checking.


Barber, Benjamin R. 1984. Strong democracy : participatory politics for a new age. University of California Press ; Berkeley, California.
Coleman S., Blumler J.G. 2009. The Internet and democratic citizenship. Theory, practice
and policy.
Cambridge University Press ; Cambridge, England.
Rossy, Maria Allessandra. 2004. Decoding the “Free/Open Source (F/OSS) Software Puzzle” a survey of theoretical and empirical contributions. Department of Economics University of Siena 424.

[1 (Accessed July 2011, Translation by the author).

[2 (Accessed July 2011).

[3 (accessed June 2011, translation by the author).

[4However, a critical feature of Adhocracy, namely the delegation system, was switched off for this experiment.

[7 ; accessed July 2011.

[9This exclusivity of gender is problematic but should not really surprise us, as women are usually barely present in the world of software engineering.