Delete. The Virtue of Forgetting in the Digital Age
(Talk based on Viktor Mayer-Schönberger’s book Delete. The Virtue of Forgetting in the Digital Age.)
What once has been put on the Web, it is never again forgotten: people is losing jobs, being banned from entering countries, etc. for having put content online that has afterwards been found by third parties, and these third parties judged the content and acted in consequence. Even if the content was long out-dated, even if the content had been removed, even if the content was addressed to other people than the ones that finally accessed it.
For ages, the problem was remembering and the norm was forgetting. Many activities had precisely remembering as their goal: language, storytelling, painting, etc. And neither the book or the phone made a change by making remembering less hard or less time consuming.
But digital technologies have.
And many institutions backed their power on the capability to remember: through scribes, the printing press, the panopticon, etc.
And it is not only about power, but about time, decision and how some decisions were made at a specific time. Seen from a distance, a decision made long ago might not be easy to understand in present time. Not being able to forget, we lose the ability to generalize and look instead at the detail (taken from Borges). We might them turn from an unforgetting to an unforgiving society.
What can be done?
Privacy rights. We can empower citizens to manage their own privacy. But what we find is that people do not care.
Information ecology. We can just use/store the necessary amount of information strictly needed to perform a task. But the problem is that we cannot foresee the future and know exactly what we will be needing or not.
Digital abstinence. That is, not sharing anything on Facebook, etc. But is that realistic?
Full contextualization. Instead of abstinence, we should put all the information we have so that a reliable context can be built around any piece of information given. It is proposal of a full transparent society, but is it feasible? is it affordable in matters of time?
Cognitive adjustment. Adjust our society, our individual processing of how we evaluate information. But history tells that forcing the human being to change just won’t work, brain rewiring is not something than can be easily done. And, indeed, what would be the appropriate mechanism to do so?
Privacy DRM. Instead of changing humans, changing technology. To ensure that only those that we have given permission to see our content can actually access it. But, then, we have to build a surveillance system that watches our moves and watches that these (registered) moves are not used for wrong purposes. That is a little bit of a contradiction.
Why not reintroducing forgetting?
Expiration date of information. Once the date expires, the content disappears. This would be the digital equivalent of oral communication. It would also link content to a specific time, keeping the sense of context with it.
The Digital Shoebox in the Attic. A way of keeping content alive, but dormant. It has to be explicitly retrieved, thus avoiding serendipity.
Forgetting should be the default, not remembering.
Q: Who defines what we have to forget? A: This is a difficult question. For instance, in a transaction, who sets the expiration date? the customer or the seller? But sometimes there might be benefits in this. e.g. Amazon knows all the books I bought in the last 10 years, but does not know what interests me now: Amazon would be happy to know, and I would have an incentive on updating that information (by setting expiration dates) so that Amazon could do better suggestions to me.
Sandra González-Bailon: wouldn’t it be better to forgive and not to forget? A: That would be nice, but it seems that, psychologically, forgiving and forgetting are very related. Forgiving is thinking that something of the past is not relevant any more. But recalling it again and again makes forgiveness more difficult.
Q: wouldn’t full transparency help in levelling information (i.e. power) imbalances in the world? A: It might work in some cases, but not all cases are about overcoming power imbalances: there are many many personal cases were full transparency would be harmful and not a matter of power.
Mike Jensen: making politicians accountable requires an active memory. We want protective forgetting, but don’t we want some protective remembering for some individuals? A: This is not about an ignorant society that does not remember anything. In ancient times, when remembering was expensive, people remembered what was valuable. So let’s try to shift the default from remembering to forgetting, make remembering the exception, and concentrate in remembering what is worth.
Q: isn’t putting an expiration date an active act of remembering (and not forgetting)? wouldn’t be recalling that something is about to expire recalling the fact itself and then remembering it again (and again)? A: It seems that humans cannot only automatically forget, but also actively forget. So, if I wish to actively forget something, the recalling won’t be activated by the expiration date.
Q: Wouldn’t be negotiation costs of agreeing in an expiration date very high? A: Technology can help in this, and enable information capturing devices with software that talks to other devices (e.g. my e-ID card, with several levels of recording permission) and acknowledges a specific expiration date.
Internet, Politics, Policy 2010: An Impact Assessment (2010)
Political Economy of the Network Neutrality in the European Union
Meelis Kitsing, Department of Political Science, National Center for Digital Government, University of Massachusetts Amherst
Network neutrality: access providers should not charge higher prices for priority delivery, elimination of price discrimination and traffic prioritization, no denial of access to specific services or applications, etc.
In Europe both content providers and network operators supported the final version of the EU telecom packages, while normally (e.g. in the US) they have opposite visions. The reason may be that they provide complementary groups, so regulations on one party might end up impacting the other party. Of course this may be context-dependent and valid only where functional separation prevails. In any case, it is more like a coordination game, like the battle of the sexes game.
But the debate is scarce and narrowed to technical issues, leaving aside ideology.
On the other hand, even if there is an agreement at the European level, regulations have to be transposed at the national level.
Estonia is a small but critical case in pioneering ICT-related legislation, maybe because of the importance of Skype at the international level.
Let’s Get Physical: Methodologies for Framing Critical Internet Policy and Governance Issues from a Sustainable Development Perspective
Don MacLean, International Institute for Sustainable Development
The information society perspective is terrific (more access to more content in less time, etc.), as terrific as terrible is the perspective over sustainable development: the ecological footprint has surpassed the biocapacity of our planet and now we are incurring into an ecological debt (WWW (2008). Living planet report, p.22), though there are several policies that could reduce this ecological debt (íbid. p.23).
Impacts of Internet and ICTs on sustainable development: first order effects (direct), second order effects (indirect) and third order effects (systemic). And some uncertainties: what technological designs and standards to connect everything and minimize environmental impacts, policies to convert first and second order effects into systemic transformation, governance principles, how to connect the Internet and ICTs to sustainable development, etc.
A project identified 10 critical Internet policy uncertainties and explored the impact on sustainable development of policy choices based on government-led, market-led, security-driven, and community-based governance scenarios.
Some recommendations are to consolidate the existing research on relationship of the Internet with sustainable development, survey research on the web 1.0 relationship between second and third order effects (individual behaviour, attitudes, values, economic structures, social structures, government structures).
Canada’s internet policy: Is ‘inclusiveness’ road-kill on the information highway
Mary C. Milliken, University of New Brunswick
Many people do not participate because (a) they have no access but especially (b) they are not included in the design of the participation processes.
In Canada, civil society organizations were excluded from telecommunication policy, though they had been included and active in media policy.
Governments have a very business-oriented approach when regulating telecommunications and broadcast media, and the people have been left aside.
The CBC began using the Internet in order to be really universal, though they didn’t had specific resources to do so. After a restructure, the CBC labels itself as a content provider, and a provider of content that has to be possible to broadcast in any channel or platform.
But the Internet has no attached requirement to be a public service, and be regulated as such. If the Internet had been understood as a broadcasting media, it could have been regulated as other platforms and have attached this public service requirement/criteria.
Policy-making for digital development: the role of the government
Ismael Peña-López, Open University of Catalonia
Internet, Politics, Policy 2010: An Impact Assessment (2010)
Digital Politics Divide: does the Digital Divide still matter?
Andrea Calderaro, European University Institute
From the Digital Divide (Norris, 2001) to the Digital Skills Divide (Van Dijk, 2009) to the Digital Participation Divide.
Wealth factor are still one of the main reasons why people use or do not use the Internet. But also there is an uneven distribution of ownership of web hosts and ownership of Internet domains [I wonder: cause or consequence?].
Concerning the political arena, there are cyber-pessimist and cyber-optimist points of view that need being bridged. The fact is that political parties are also unevenly on line depending on the country.
The reasons for that are the number of internet users, the level of democracy, and the GDP.
Unraveling Different Barriers to Technology Use: Urban Residents and Neighborhood Effects
Karen Mossberger, University of Illinois at Chicago
Uneven access to the Internet may have a negative impact in the opportunities of the people and thus drive them towards social exclusion. And living in poor neighbourhoods, having a low income or lower educational levels are reasons that explain lower access to the Internet.
When asked the citizens of Chicago why they did not had broadband at home, 30% said they were not interested, 27% cost, 9% difficulty.
Per neighbourhood, “not interested” is a reason much likely answered by whites and Asian-Americans (42%), then African-Americans (29%) and then Latinos (19%). By age, older people are more likely (30%) to say that the reason for not having broadband is “lack of skills”, the same ratio when looking at the income.
Neighbourhoods magnify these barriers to access the Internet, because they magnify cot and skill barriers for residents of areas with high concentrations of African-Americans and Latinos. There is a double burden of concentrated poverty.
Amazonian Geeks and Social Activism: An ethnographic study on the appropriation of ICTs in the Brazilian Amazon
Marie Ellen Sluis, University of Amsterdam
Instead of talking about access, talking about what means to have or not to have access: meaningful access. And the same for inclusion and meaning digital inclusion.
Projeto Puraqué is a collective of social activists using ICT as a tool for social inclusion, increasing critical knowledge on regional socio-political problems and issues. ICTs a tool rather than an end.
Examples: opening up the computre to demystify technology and enhance self-steem, raise awareness on e-waste and fostering reuse and recycling as gambiarra alternative,
The project operates in a certain framework that seeks social transformation in the long term and on a sustainability basis. It is the people who decide what is beneficial for them, and the project is a lot about the digitization of what Brazilians do most: social networking.
Indicators of the digital divide and its link with other exclusions
Jocelyne Trémenbert, Institut Telecom / Telecom Bretagne, Université Européenne de Bretagne, Marsouin
The goals of the research is to explore the polymorphism of the digital divide and its links with other forms of exclusion. Is the distance to the Internet different for different types of exclusion? Do we find within the digital divide expressions of exclusion?
Aage, gender, educational level, income, occupational category and localisation enable to predict with +70% accuracy the use of the Internet, especially the occupational category and the educational level. Non-users are often isolated people: the digital divide goes hand in hand with the social divide.
Five types (clusters) of non-users: the users to be (5%), the potential users (19%), probably / hesitants (41%), the resistants (16%), the excluded (19%).
We need new indicators of the digital divide, new elements about the specificities of some categories of non-users, and a new quantitative typology of non-users based on data on inhibitors,motivations, points of view and picturing.
Internet, Politics, Policy 2010: An Impact Assessment (2010)
Online citizen-campaigning in the UK General Election of 2010: how did citizens use new technologies to get involved?
Rachel Gibson, Institute for Social Change, University of Manchester; Marta Cantijoch, University of Manchester; Stephen Ward, University of Salford
Previous elections have shown that the Internet is only for “politics junkies”, and that e-participation has just reinforcement effects.
In news, the Internet is presented initially as a game changer, to become lately a useless tool for candidates and parties.
During the UK 2010 election, there was more people online than during previous elections, and more people went online to look for political information, both at the official and the non-official sites.
4 factors of e-participation: e-communication (info seeking, e-discussion, online videos), e-targeted activities (e-contact, e-petition, e-donation), e-formal (register, party tools, join/start election related social networking site group) and e-informal (forward, post, embed unofficial political content).
Are people engaging/participation in these four factors differently?
Age, online skills and interest in politics are amongst the more important reasons that determine the kind of online participation that a citizen is more likely to engage in. Partisanship strength also has an impact in e-targeted activities and the e-formal factor.
In general, e-communication has an impact in increasing the probability of voting amongst people not interested in politics. Notwithstanding, there is a need to disaggregate e-participation, as not all e-participation is about the same thing, as it happens offline.
Campaigns and Communications: Is the Revolution in Digital Media Changing Political Organization? An Examination of the 2010 UK Elections
Mike Jensen, Institute for Governance and Public Policy, Autonomous University of Barcelona; Nick Anstead, London School of Economics (LSE)
The Network Society has turned upside-down the boundaries of some concepts like political and non-political, places and spaces of flows, etc. How does campaigning fit in this new context?
Facebook comments and twitter messages were analyzed to see the differences between the national campaign and the local level campaign (here the case of Birmingham). And we can see big differences in what happened in Facebook and Twitter between the national and local campaigns.
Concerning horizontal communication, we can see that for some candidates, at the local level there was much more decentralization than at the national level.
Elements of decentralization and localism can both be found at the UK 2010 election. Local candidates make relatively little reference to the national campaign while, at the same time, there is evidence of the existence of horizontal links at both the national and local level.
Towards a more participatory style of election campaigning? The impact of Web 2.0 on the UK 2010 General Election
Darren G. Lilleker, Centre for Public Communication Research, Bournemouth Media School, Bournemouth University
To what extent was interactivity permitted on party websites at the 2010 election? What type of interactivity was permitted? How has permitted interactivity evolved between peace time and previous elections?
Typologies of interactivity, depending on the level of receiver control (high, low) and direction (one-way, two-way, three-way) of the communication flow: Feedback, Mutual discourse, Public discourse, Monologue, Repsondive dialogue, Controlled responses.
We can see that individuals (candidates) are usually more popular in social networking sites than parties. Uploading played a key part on interactivity, sharing creates a network effect, adding info was offered by a minority.
Different parties used different types of interactivity, ranging from monologue to public discourse. In general, though, there was a lot of two-way communication. Surprisingly enough, three parties (Lab, LDem and Green) abandoned many of their interactive features in their websites, while the others, on the contrary, increased the interactivity, a trend that has been thus in the last years. Similarly, some parties are increasingly losing control over the message.
In the UK 2010 election the Internet was mainly used to sell the party/candidate and to manage relationships, but social networks were also used (besides branding) for interaction. Supporters were contacted horizontally and more co-building was created. Notwithstanding, many partisans would share their ideas with other partisans, but not willing to do it with the party.
In general, participation was low, and people would engage elsewhere but usually in trivial if engaging ways. Notwithstanding, very active communities triggered more participation, creating a virtuous circle of high participation.
Me too for web 2.0? Patterns of online campaigning among candidates in the 2010 UK general election
Maria Laura Sudulich, University of Amsterdam; Matthew Wall, Free University Amsterdam
What is the use of the web 2.0 by candidates? How do they use Facebook? What are the elements determining the uptake of online campaigning by candidates in the UK 2010 election? Is the rationale behind launching a campaign website different from the one explaining candidates’ presence on Facebook in the UK 2010 election?
the probability of having a website (or a facebook profile page, a fan page or a group page) was calculated depending on incumbency, marginality, party affiliation, bookmakers odds / implied chance of winning, and whether opponents had a website and the number of these opponents.
A striking finding is that there seems not to be a thorough political strategy to be on Facebook,and results are very inconsistent across different possible forms of Facebook campaign presence. Marginality, notwithstanding, is very significant for both having a website and being on Facebook, as party affiliation is. Implied chance is a significant predictor of candidates websites, but not for Facebook groups; on the contrary, the “me too” factor is not a significant predictor of candidates’ websites but it is for Facebook groups.
Internet, Politics, Policy 2010: An Impact Assessment (2010)
Why Mobilize Support Online? The Paradox of Party Behavior Online
Ana Sofia Cardenal, Internet Interdisciplinary Institute (IN3)
There is an increasing number of Internet users, there have been some very interesting cases of political use of the Internet, and nevertheless, there still seems to be an underexploiting of the opportunities that the Internet offers for political mobilization.
The parties would use the Internet if benefits are higher than costs. Benefits would be increasing expectations of winning office and competition. Costs would be party cohesion (risk of losing control of message when using the Internet), party size (need for more resources and/or support cost of opportunity of allocating resources on online campaigns) and size and importance of extra-parliamentary organization (strategies of recruitment might interfere with online mobilization).
(H1) Large parties will have more incentives than smaller ones to use the net. (H2) Large parties that are in the opposition and compete for office have more incentives to be active online. (H3) Non ideological parties or conversel highly cohesive ideologicals are in better position to use the Internet in their own benefit. (H4) Parties with small extra-parliamentary organizations will also be in a better position than parties with large bureaucracies.
The research analyses the websites of 12 parties and the actions of +1300 online activists.
Findings show that large, non-cohesive, and parties with small bureaucracies have the best (the most interactive and participatory) websites and are the most successful in mobilizing their followers online. In particular, the Catalan nationalist party (CDC) is arguably the one that does best, as it has traditionally been a mobilization party.
Concerning supporters, while it seems that parties do not matter much in offline activism, cyberactivism is more successful in specific parties.
Summing up, party characteristics matter in explaining how parties behave online and what is their impact in cyberactivism.
Internet and Votes: The Impact of New ICTs on the 2008 Spanish Parliamentary Elections
Albert Padró-Solanet, Universitat Oberta de Catalunya
Goal: what is the role of the Internet in party competition: normalization or revolution? And what is the relevance of the context in all the matter?
The dependent variable is party vote or abstention (not participation) and as independent variables there are party vote intention at t-1, campaign exposure (offline, online, political information).
Findings show that online political information exposure differs from offline exposure: there not always is a reinforcement on party vote, and sometimes there is no impact or a negative one on major parties. In other words, e.g. people that stated at t-1 that they would be voting to the PSOE, have a decreasing probability of ending up doing so as their exposure to online political information increases. The probability of abstention, on the other hand, increases as the exposure to online political information also increases. Offline exposure, though, acts in the opposite way, reducing abstention and reinforcing your initial intentions to vote a specific party.
Reasons for this behaviour might be that, probably, online political information is more fragmented than offline political information, but it doesn’t lead to selective exposure (against normalization hypothesis but vs. a Balkanization hypothesis).
Lessons Learned from Obama? The Effect of Individual Use of Party Websites on Voting in the Elections to the European Parliament 2009 in Germany
Pablo Porten-Cheé, Ilmenau University of Technology, Germany
In Germany, parties still spend little on the Net, though they state the importance of web campaigning specially to inform the public, to mobilize young voters, to activate partisans, etc.
What is the impact of political informational (including use of party websites) and interpersonal political online communication on voting? The assumption is that there is a positive impact which leads to more votes.
Findings show that there was a highly significant effect in the green vote in the latest German elections.
Voter Targeting via the Web – A Comparative Structural Analysis of Austrian and German Party Websites
Uta Russmann, University of Innsbruck, Austria
How do parties target their audiences online? Classically, parties have segmented voters in combinations of age, gender, ethnicity, profession, education, ideology and lifestyle.
(H1) Catch-all parties address a more general audience on their website. (H2) Austrian parties address a more general audience on their website, zed features and techniques on their website to specific target groups.
Results show that catch-all parties do target specific groups and client parties address the general public and specific target groups evenly. The behaviour is similar in Germany and Austria. On the single website the general public is addressed more often. After the general public, party members and supporters are the most targeted audiences by political parties. There does not seem to be a clear strategy relating online targeting.
Ute Russmann: is there a profile of citizen that only gets their political information online? Albert Padró-Solanet: people are very heterogeneous in how they find their political information.
Q: could targeting be made by tailoring different candidates within the same party, each candidate shaped according to the expectations of different segments? Ute Russmann: there does not seem to be such a practice. On the contrary, the image of all the candidates of a given party is very homogeneous.
Stephen Ward: could online negative campaigning have an impact in people being informed online voting less? Albert Padró-Solanet: doubtless there is an impact, but maybe is not that much that there is such a thing as negative campaigning, but that people online find much more information, positive and negative, about candidates and parties.
Stephen Ward: in targeting, wouldn’t it be a better channel using direct e-mail or social networking sites? Ute Russmann: they surely are now, but during the German elections in 2008 (which is the object of the research), political parties were not really using web 2.0 tools very intensivelly.
Internet, Politics, Policy 2010: An Impact Assessment (2010)
New ways for policy-makers to interact with citizens through open social network sites – a report on initial results
Matthew Addis, IT Innovation Centre, University of Southampton, UK
WeGov aims at using social networking sites (where the people is)
to engage citizens in two-way dialogues as part of governance and policymaking processes.
Several issues are raised, though, concerning security, privacy, digital identity, hacking, masquerading, bugs and system malfunctions, etc.
So far, though, it seems that social networking sites do offer plenty of potentgial for improved interaction of plicy makers and community, and they are complementary to other models. Notwithstainding, care is needed on what’s legally acceptable, risks need being assessed. On the other hand, there is a high dependency on the major actors (Google, Facebook and other SNS operators). Last, many technical challenges have to be overcome before real dialogue can take place.
Q: Should politicians and/or public servants take part in social networking sites? Are them an appropriate place where to engage in a conversation? A: These issues are taking into account in the project, though no results are available at this moment. Notwithstanding, SNS can be used to create only-politicians workgroups, so they can work amongst them in a new environment without “third parties” peeking in.
Q: Isn’t that a big brother approach to policy making? Are SNS only used for surveillance but not for listening? A: Of course it depends on the usage.
Analysing e-consultations with the help of Computer Assistance
Aude Bicquelet, London School of Economics (LSE)
Text mining is the process of extracting information in large corpora with the aim of identifying patterns and relationship in textual data. Can text mining methods help the analysis of large corpora such as e-consultations?
Alceste software was used to explore petitinons and comments from the citizens. After a first scan, cluster analysis was performed to try and isolate groups of terms and infer from them concepts/meanings.
Text mining methods help in categorization of data, reduction of information, visualization to help and disentangle complex and prolific data, high velocity of analysis.
On the other hand, there are some risks like points or issues missed by the algorithm, missing data, insensitivity to meaning and context, etical issues related to privacy or confidentiality, “dehumanization”, non-awareness of participants being a matter of research, etc.
Ismael Peña-López: Though there might be a risk of decontextualization, text could also be mixed with other data (e.g. on the author of the text) and thus reintroduce context within the equation. Would that be possible? A: Yes, this is definitely an option.
Q: Text mining can be used for “sentiment mining”.
Q: How are these findings used? A: They feed back the whole participation process and they inform the people designing the systems so that they can improve them.
Surfing the Net: a pathway to political participation without motivation?
Rosa Borge and Ana Sofía Cardenal, Internet Interdisciplinary Institute IN3, Universitat Oberta de Catalunya, Barcelona
Goals of the research: To investigate the relationship between internet motivation and political participation, and to see whether the use of Internet changes the importance given to motivation.
The literature is not conclusive, and both positive and negative impacts have been found in Internet over political participation in the literature.
The main drivers of political participation are resources, psychological involvement and recruitment networks. Does the Internet reduce the importance of motivation (psychological involvement)? Does the Internet has any impact on it? When participation is not so costly (due to the Internet), maybe political interest is not so important, as participation is almost costless.
(H1) Use of the Internet will not cause the effect of motivation to disappear, but digital skills may actually be a barrier. (H2) Browsing aimlessly on the Internet and being contacted online will increase the probability of online participation.
First results show that political interest and political skills have a direct effect on online political participation. Specifically, having internet skills does have a direct effect in participation, though it will not make political interest disappear as a reason to participate.
There also is a relationship between browsing aimlessly on the Internet and being contacted online and an increase in online political participation.
[Own ramblings: Pippa Norris stated that citizens were increasingly engaged in “actions” rather than “ideologies”. Isn’t that the next step, where “action” is substituted by “casual politics”? Is there still a role for deliberation? We’re witnessing transition fr endorsement of ideas, to participation in actions to casual politics on the run. Seems like e-politics is less of the Habermassian e-agora and more about e-herding.]
Civil Society and ICTs: Creating Participatory Spaces for Democratizing ICT Policy and Governance in the Philippines
Ian Jayson Reyes Hecita, Florida State University
Goal of the research: analyse the state of Civil Society Organizations (CSO) in Philippines.
Context: failure of institutions, lack of capacity of government, democracy dominated by the elite.
Thus, CSOs are raising many relevant issues in Philippines related to ICTs and society.
But government receptiveness on CSOs can depend on leadership, on tapping government “champions”, the political attractiveness of the policy issue, and the effectiveness of engagement will be more viable at the executive level than at the legislative level. It will also depend on the openness/receptiveness/political will of government agencies to CSO participation, on the level of institutionalization of CSO participation, the resources of CSOs and capacity and skills to engage the government, the critical mass in their basis and policy audience, the need to develop consumerism, their capacity to carry on research and, of course, the legal and regulatory framework.
As a conclusion, CSOs may have a limited impact in terms of influencing policy outcomes, but they may have an important one in brining relevant topics on the table.
Internet, Politics, Policy 2010: An Impact Assessment (2010)