Let me publish here excerpts of our (@almereyda and I) paper for comments and feedback.
Full paper prepared for the “City as a commons” conference :
20151105_IASC2015Bologna_Article_final.pdf (306.4 KB)
Let me publish here excerpts of our (@almereyda and I) paper for comments and feedback.
Dear all, please feel free to leave comments inline via the following link:
In the top right you will find a small arrow button, which gives access to extended features when selecting text in the PDF.
#Action researching TransforMap
The TransforMap process formally started in Germany at the end of 2013 when commons activist Silke Helfrich called activists of various communities of practice to unite in putting together a map of all alternatives. This call happened in a context where various related communities (commons, social and solidarity economy, transition network, collaborative economy, etc.) have increasingly shown a shared will to increase convergence in all domains. This resulted in a kick-off meeting in Munich in March 2014 gathering over thirty participants belonging to various communities spanning from the commons and collaborative economy to community gardening and OpenStreetMap. Throughout 2014 and 2015, TransforMap developed into a collective slowly aligning its understanding of the issue at stake and a possible solution. Limited funding was provided by various foundations enabling the organization of several meetings, the setup of a communication infrastructure and some first technical development. Until October 2015 and the start of more substantial funding, contributions have been largely volunteer ones. It is important to note that no legal entity was created, funding being so far managed by an existing third-party non-profit.
##Action research for knowledge commons
Relatively early on TransforMap was framed by its participants as a commoning experiment (Bollier and Helfrich 2012) intended at crafting an incremental solution for federating mapping efforts of alternative economies. This process implies considerable learning from the participants. Having integrated the premise that sustainability scientists have the responsibility to collaborate with other actors in knowledge systems to co-produce knowledge (Cornell et al. 2013), we are early participants in the TransforMap initiative, however not the initiators. Defining TransforMap as a research object at the interface of science and society, our approach is definitely process-oriented rather than descriptive-analytical (Wittmayer and Schäpke 2014). As participatory action researchers, we recognize the existence of a plurality of knowledges in a variety of institutions and locations as opposed to tenets of an ‘objective reality’ (Kindon, Pain and Kesby 2007). We align with Wittmayer & Schaepke’s understanding of action research as “the collaborative production of scientifically and socially relevant knowledge, transformative action and new social relations through a participatory process” (2014, 2). Importantly we see in such definition the ingredients for the self-organization of a (knowledge) commons: collective action – collaborative production – and self-governing mechanisms – participatory process (Hess and Ostrom 2007).
The existing literature on participatory action research has not specifically looked at the challenges of mainly digital processes. As Hess and Ostrom stress (2007, 43), new information technologies have redefined knowledge communities and disrupted the traditional world of information users and information providers, leading to institutional change at every level of the knowledge commons. In this context the Institutional Analysis and Development (IAD) framework is seen as well suited for the analysis of resources where technologies are developing at an extremely rapid pace (Hess and Ostrom 2007). This is the case in the fields of collaborative mapping and open data where TransforMap is situated from a practice point of view. Previous efforts to adapt the IAD framework for digital commons have already been conducted (Fuster Morell Nov 2013; Schweik and English 2013). Those adaptations retain three top analytical containers to study variables: the resource level (or technology attributes), the community, and the governance (or rules-in-use); those variables influencing outcomes of actions arenas where participants interact.
In the reflective part of this paper where we descrine the TransforMap proposal, we rely mainly on the work of Fuster Morell (Nov 2013) that defines Online Creation Communities (OCCs) as a particular type of online community whose goal is to build and share a specific knowledge resource. Our choice is rooted in the fact that Fuster Morell (Nov 2013) includes the design of the participation platform in the study of governance (code is rule) while Schweik and English (2013) tend to leave that aspect as part of given technical/resource attributes. Indeed, in their study they look at software being developed on existing collaboration platforms. In the case of TransforMap, the foundational step of assembling and further evolving a platform for participation (or collaborative infrastructure) has been, and is still, at the core of the governance process, crystalizing tensions among the community and strongly influencing participation.
##A word on research material
In this conference paper, we rely on various sources of insights. A preliminary step initiated by the TransforMap collective has been to identify existing maps or inventories that list and locate initiatives that belong to what we defined as alternative economies. This was done using a collaborative method, starting with a shared online document and later setting up a semantic wiki to open the effort of collecting such resources. From the exercise we have formed a broad overview of the existing landscape. In order to understand in more depth the motivations and challenges of the communities and individuals behind those resources, we have conducted a series of interviews for ten different map initiatives. Early analytical results were shared and discussed with the broader TransforMap collective online and during a face-to-face meeting. In addition, multiple threads of conversation have been informally initiated with a broad variety of actors: initiative networks, individual activists, geodata experts, linked open data specialists, etc.
 See blog post at https://commonsblog.wordpress.com/2013/12/13/mapping-the-alternatives-kiezmap/
 This can be accessed at: http://mmm.3oe.de/wiki/Mappings_Overview.
 Those reports can be found at https://discourse.transformap.co/c/community-building/research under the title “Field reports”. The
face-to-face meeting happened on 17-18th of June 2015 in Munich.
#Characterizing the problem: identifying the social dilemmas in mapping alternative economies
If such a data commons as what TransforMap envisions has not yet emerged, there may be (good?) reasons. Commons are resources shared by a group that is vulnerable to social dilemmas (Hess and Ostrom 2007). While the latter were archaically thought as an inescapable tragedy requiring government intervention (Hardin 1968) or privatization of resources (Olson 1965), a vast amount of empirical research has shown that such dilemmas can be overcome through self-organization and collective action (Ostrom 1990). Here we propose first to characterize those social dilemmas – the status quo in mapping alternative economies – including some that may be specific to data commons. The knowledge required to make this assessment if largely the fruit of the collective process initiated within TransforMap itself.
Scientists in search of data are probably the greatest freeriders of all. They often ask data owners (e.g. networking organization such as Transition Network or Anstiftung & Ertomis) to access their database or go directly to scrape data on the web. In those cases, they rarely contribute back to enriching the data. One of our interviewee express irritation and talked of extractive science. In terms of mapping infrastructure, the quasi entirety of the maps we listed are either using the Google infrastructure for free – in that case it belongs to Google’s business model – or using open source software without republishing their own instance of it for others to benefit from. Having too low resources (technical or financial) the individuals or organizations behind those maps cannot invest in contributing back to the open source mapping infrastructure.
De facto, mapping initiatives – from scientists or activists – are not connected to one another. Restrictive licenses (including those specifying non-commercial use) are undermining reuse and interconnection of data. Worse, many maps do not have license on their content, preventing use, or making it time-consuming. In some cases data is published in formats that are not easily reusable (e.g. PDF). With data, enclosure cannot be reduce to a binary dilemma – open or closed. There may be various levels of enclosure, or, in contrast, openness. Those are summarized by Internet’s founder Tim Berners-Lee in his 5-Star scale for linked open data and often used aside Open Knowledge’s definition as a reference conceptualization for open data.
In the era of Google and Facebook, digital data is systematically being commodified, often with users ignoring its use. Unknown algorithms use personal data to drive internet users to increase their traffic and the number of clicks as that is directly correlated with revenues from advertising. As soon as 2010, Google’s CEO Eric Schmidt, predicted that it will be very hard for people to watch or consume something that has not in some sense been tailored for them (Holman 2010). This filter bubble as it is called by Pariser (2012) raises serious questions regarding the discoverability of information that does not conform to the mainstream logic of profit making: e.g. alternative economies. From our inventory of maps a majority (around two thirds) were using Google Maps as it is the technological solution that is the most easy to use while involving no financial cost. The remaining third use the open data compliant OpenStreetMap in combination with various open source mapping libraries.
Opening data collection to the crowd inevitably raises the problem of vandalism and spamming. Login before editing is generally the most common but not definitive solution. Another type of data corruption is connected to the fact that some service providers may include in their terms of services restrictions to use data that was generated on their platform. This is the case of Google geocoding: generating geo-coordinates with Google geocoder legally prevents users from using such data on another platform. The OpenStreetMap community explicitly warns contributors not to do so. An issue linked to the combination of separate databases is the redundancy of data. Eliminating duplicates may require time-consuming manual work and for large datasets discourage such aggregation.
##Liability & control
Data falling under privacy restrictions such as personal contact data (email, address, etc.) involve legal liabilities that will reduce the ability to share entire database of initiatives across platforms. In this vein, sharing data may make the right of people to remove personal data from the web difficult if not impossible to implement and deter initiatives to connect data (this was clearly stated in one interview).
##The autist dilemma
Like text, maps are not a neutral representation of reality (Harley 1988). Combining different maps (understood as databases) will therefore often encounter issues linked to conflicting ontologies and the lack of common data vocabularies. In practice, even the will to combine data may be lacking as researchers or activists and practitioners work on specific themes and specific places and may not see the benefits of collective action, or, alternatively identify collaboration as undesirable as it may imply some alteration of their ontologies and vocabularies. We called this the autist dilemma.
##Goldfish and fossils
Current inventories and maps of local initiatives are generally blind to time. On the one hand, data may be collected during a short span of time (typically for research project) and then left to rot on a shelf, digital or not, and slowly fossilize, never being updated. On the other hand, collaborative maps mostly do not use versioning, each new edit of an entry erasing previous data, preventing to see evolution of the data through time. We call this the goldfish and fossils dilemma.
#The TransforMap proposal
The progressive identification of these social dilemmas led to the incremental development of a TransforMap proposal as an answer in order to get closer to realizing the potential of linking and opening data about alternative economies. We attempt to restitute that proposal here as we understand it as of October 2015. We structure that presentation following the IAD framework as refined for digital commons (Schweik and English 2013; Fuster Morell 2013). This involves three packages of variables: the shared resource (the commons as such), the community of commoners, and the governance (rules-in-use).
The TransforMap proposal necessitates distinguishing three levels of shared resources being produced as a commons. While the motivation of the effort lies in pooling data (1) about initiatives and allow its vizualization, this cannot be achieved without developing a stack of protocols, standards and apps (2) to establish those connections to avoid building a unique and central database as another silo. Unfortunately, the complexity of such an enterprise and its digital nature makes necessary the assemblage of an ad-hoc collaboration platform (3) for the negotiation and development of all those features. This means the resource is a permanent work in progress.
- TransforMap data
The core of TransforMap’s proposal is to convince data owners (networks, research projects and institutions, associations) to pool together existing and scattered data resources, and further enrich them collectively as a commons. This also means enabling individuals (the crowd) to become co-producers. Such data involves basic information (name of initiative, address, description, etc.), but also more elaborate attributes such as keywords, information on the main mode of interaction characterizing the initiative (e.g. sharing, co-producing, etc.) or belonging to an existing network or community of practice (e.g. Transition Network, P2P, etc.). However, the TransforMap proposal is to avoid too much centralization in data management. Thus, only basic information is centralized and stored in an existing map commons: OpenStreetMap. This central point is used as a hub to link various other databases for extended data attributes (e.g. keywords or needs addressed by one initiative).
In addition, aggregating separate data sources implies that they should converge along two lines:
Combining different data sources generally implies to combine different licenses together and sometimes (re)assign a license to one or more datasets. As the number of data sources rises, combining multiple licenses become impossible. TransforMap’s goal is to maximize the usability of data by encouraging data owners to choose the most open type of license: i.e. the exclusion of any use restrictions through the release of the data under public domain – what is sometimes called the zero-license. Aligning licenses is a socio-political process involving argumentation, negotiation and cultural change. This simply requires intense social interaction and a powerful narrative.
Aggregating data is straightforward when different datasets are structured the same way. When they are not, it becomes a very challenging endeavor. The proposal of TransforMap is to encourage each data owner to align with the OpenStreetMap structure for basic geodata. Then, the aim of the collective is to catalyze exchange between those data owners in order to incrementally align data ontologies. Through exchange with previous efforts to do so in the field of solidarity economy, we have acquired an understanding of the significant challenge it represents. Such alignment of various data owners will be very slow and will involve a lot of social interaction. Eventually, this process is similar to the development of a common language. To happen, it needs focused interaction between the concerned individuals from various communities and organizations. To do so there is a need for a dedicated space. This is covered in point 2.
- TransforMap infrastructure for mapping
For individual users TransforMap will consist of a web app allowing filtering of initiatives along pre-defined terms, adding and editing points of interests (i.e. initiatives), visualization of further data for each point, an import function to save data on a local device, and a print option to bring the map onland – as opposed to online.
For database owners, TransforMap provides a discussion space along with in the future various tools to make data ontologies apparent and allow comparison and matching. A set of specifications and guidelines for data modeling is being co-developed with the perspective of enabling exchange of data across platforms. The long-term aim of TransforMap is to incrementally catalyze the development of linked open data vocabularies specific to alternative economies. Linked open data vocabularies are a core component in enabling automatic data exchange, but need considerable interaction between actors who manage data collection platforms. Therefore, TransforMap’s proposal is to put together the necessary infrastructure for such interaction.
- Collaboration platform for TransforMap
Producing such infrastructure for mapping (from software to standards) requires putting together a collaboration platform sustaining the TransforMap process itself. At first, this was completely overlooked by the participants. Then, quickly it appeared that the complexity of the task could in no way be addressed through out-of-the shelve infrastructure such as emails or cloud task-management platforms (i.e. Trello). A forum was set up using open source software to provide the space for the development of a community that would collaboratively gather and aggregate data (1), and discuss the required infrastructure (2). Existing external infrastructure for software development (i.e. Github and Taïga) are then used. One central point of the TransforMap forum is the “Who’s Who” category where the members of the collective presents themselves and encourage interested people/organizations to do so as well and expose their interests, skills and potential area of collaboration. This plays the role of an awareness hub where we identify who the commoners are (or could be).
As for any commons that has to be produced, the TransforMap data and infrastructure commons require the creation of a community: the community is forged not given. Here it is described as the TransforMap collective.
The core of the TransforMap collective is made of about twenty individuals that may or may not represent other organizations and repeatedly interact online or onland. Understood in a broader sense, the TransforMap potential community reaches far beyond as all owners of databases and maps of alternative economies are targeted by the community development process. In addition, enthusiastic mappers with a specific interest in grassroots innovation are also potential participants. Hackers and FLOSS developers – in particular those working towards linked open data and open mapping tools – are also seen as key targets in establishing the TransforMap community. Eventually, researchers are also encouraged to get involved. With this background, we observe that in the eighteen months since its kickoff, the collective did not significantly grow, seeing some original members leaving and new ones joining.
TransforMap has a relatively high heterogeneity considering skills sets. It gathers activists from various horizons (community gardening, collaborative economy, commons, solidarity economy), (geo)hackers promoting FLOSS, and some researchers making the dialogue sometimes difficult, but also ensuring a large spectrum of capabilities. As it is often the case with digital commons (Schweik and English 2012), the TransforMap collective displays a strong gender bias towards men (about 4/5). The community itself is aware of that fact and sees it as a problem and tries to engage more women in the process with limited success. So far the community concentrates mainly in Germany and Austria with gradual expansion towards the USA, France, UK, Finland, Italy, and Portugal among others.
All participants are free to enter and leave the community.
The TransforMap collective has been very active on building relationships on various fronts. On the data side, communities – and more specifically individuals
involved in data management and communication – representing various strands of alternative economies (e.g. commons, collaborative economy, transition movement, solidarity economy, common good economy, P2P economy) are in regular contact with the TransforMap collective until they themselves directly collaborate. From the infrastructure side, TransforMap relies on many existing software commons: from its forum software (Discourse), to all the components of its mapping infrastructure (OpenStreetMap, Leaflet, Wikidata, etc.), the TransforMap collective uses and, sometimes, contributes to the development of code. In that sense, it is difficult to draw a line to what the TransforMap community of developers really is as it relies on and contributes to other existing software and data commons: it is embedded in an open source ecosystem (Schweik and English 2012).
##Rules-in-use and governance of TransforMap
The original mission and scope of the community was to a large extent defined by Silke Helfrich as she called (in German) in December 2013 activists from the “#peer-to-peer, #Ouishare, #SolidarityEconomy, #Commons, #Degrowth, #DIY, #Maker, #FLOSS, #TransitionTowns, #Economy for the common good, […] all who can or want to think freedom, sustainability, and fairness together” to the creation of a map of local initiatives across communities .
Cultural principles/Social norms
Helfrich’s initial call contained already most of the cultural principles framing the TransforMap process: freedom, sustainability and fairness. Being anchored in the commons tradition it refers to cultural principles associated to Free Libre and Open Source Software (FLOSS) community: knowledge should be open, and made accessible – code is open and published. In addition, participants of TransforMap have adopted a very flat power structure in circles. The norm being that anyone may express an opinion.
Design of the platform of participation (where regulation is embedded in the code)
So far the only platform of participation that has been established is the forum. There, all content is open to everyone, except some topic deemed sensitive such as funding that are only accessible to a limited number of trusted users having a profile on the forum – the definition of a “trusted user” was not formally defined and is being left to the discretion of forum administrators.
The planned mapping infrastructure will be decentral and partly embedded in existing commons (OpenStreetMap, Wikidata). This has the implication that some components of the wider constellation of TransforMap’s mapping infrastructure will be subject to regulation embedded in the code of those existing platforms. Beyond that, there is still a lack of consensus among the TransforMap collective whether editing should be allowed without a login and whether there should be individuals with more editing privileges (e.g. current data administrators from existing maps). But generally, the decentral model implies that participating data curators/owners will retain agency and be able to use their own ontologies and display customized subsets of data.
Self-management of contributions: autonomous condition of participants in allocating their contribution to the building process.
Participants in TransforMap allocate themselves to tasks. This has had the implication of a very unpredictable and (de facto) slow development and no systematic overview of task to be implemented. It has sometimes caused tensions in particular with individuals willing to have a more traditionally planned project management.
Formal rules or policies applied to community interaction.
The principle of consent – the absence of opposition – borrowed to the sociocracy method is agreed to be the core formal rule for decision-making. Transparency of communications is also a formal rule, but has been hard to implement, individuals often sticking to mainstream (i.e. closed) patterns of communications and not actively sharing information.
There is consensus to publish as much data as possible under public domain as it is seen as the type of license that allows the maximum reuse of data. However, embedding TransforMap in existing commons such as OpenStreetMap will imply that part of the data follows those license terms (i.e. ODbL – Open Database License). In addition applying a public domain license will be difficult to implement while adding existing data sources that have already specific licenses, often excluding commercial use. Such licenses have been recognized by TransforMap to be problematic as commercial use may be unintended (e.g. advertisement on a blog) or deter potential contributors to invest resources. Convincing data owners to assign a zero-license (public domain) to their data is a very intensive social process as it conflicts with the general practice among the grassroots of using non-commercial licenses, or, even more often, the practice or not licensing data.
When it comes to the software infrastructure, licensing is unproblematic within the TransforMap collective: each component has its own license and are systematically open licenses allowing reuse in the FLOSS tradition. However, there has been the case that one former participant had to invest considerable amount of time in converting pieces of code from a proprietary stack to an open source one. That participant eventually left the community, stressing in particular the lack of recognition of the time invested.
Decision-making and conflict resolution systems with regard to community interaction.
Decision-making among the TranforMap community relies on the consent principle. In practice, some individuals, because of their role (e.g. signature right on funding contracts and payments, deployment and maintenance of collaboration infrastructure) have acquired much more power in the community than others, creating conflicts with members that felt excluded, or “betrayed” and took distance with the process. At the time of writing, the absence of a formal conflict resolution system seems to hinder collaboration.
The TransforMap collective is a self-provision assembly meaning that there is no dedicated infrastructure provider (Fuster Morell 2013). The collaboration infrastructure (i.e. website, forum, owncloud instance, mailing list servers) has for been first hosted by one of us, author of the paper. Progressively, and to limit personal liabilities as well as securing the process, the provision of the participation infrastructure has been formally transferred to the German non-profit association Ecobytes, specialized in providing digital infrastructure for communities in the field of sustainability and being operated by members of TransforMap.
In its first year of existence, the TransforMap collective managed to secure some small funding to partially compensate work from its community members. For practical reasons, the Austrian non-profit association Get Active took the role of being the legal entity applying for funding and receiving funds. This implies that the TransforMap member having the signature right (and liability) for this association concentrates de facto a significant amount of power, which crystalized some tension and conflict in the community.
For Schweik and English (2012) the key measure of success of an open software commons is whether commoners stay or leave. In that sense, is TransforMap successful? On the one hand several original members have left the collective and the initiator, Silke Helrich, has announced this summer that she will take distance from the process. Various reasons have been motivating original members to drop their involvement: from the fact that TransforMap grew far beyond the initial motivation of creating “a single map with all alternatives” (a commons activist), to a lack of acknowledgement of contributions (a software developer), to expression of distrust in individuals and the collective management of financial resources (a commons activist), to the difficulty in adapting to new digital tools for collaboration (various activists), to a too complex goal (a developer). On the other hand, the collective welcomed new individuals, compensating to some extent the defections. As action researchers, it is probably where forming an objective assessment is the most difficult as we are part of the community dynamics and have not performed a systematic investigation of the reasons that pushed individuals to leave and others to stay. Still, it is our view that passed the first successful semester the TransforMap collective is kind of stalling in building a core and growing community of contributors.
Now the same counts for
Very interesting paper,
can i share it with Olivier Frerot ?
Thanks! Just seeing now your message. Sure, please do.
We will be revising the paper (including shortening it) to submit it to a journal towards the end of January, so feel free to share your reactions/comments/critiques here!
could you have a look to this post and let me know your feelings and advice ?
I think it could be useful to support crowdmapping