Risk Analysis to Overcome Barriers to Open Data

Sébastien Martin1, Muriel Foulonneau2, Slim Turki2 and Madjid Ihadjadene1

1
Université Paris 8, Vincennes-Saint-Denis, France
2
PRC Henri Tudor, Luxembourg, Luxembourg

Abstract: Despite the development of Open Data platforms, the wider deployment of Open Data still faces
significant barriers. It requires identifying the obstacles that have prevented e-government bodies either from
implementing an Open Data strategy or from ensuring its sustainability.This paper presents the results of a
study carried out between June and November 2012, in which we analyzed three cases of Open Data
development through their platforms, in a medium size city (Rennes, France), a large city (Berlin, Germany),
and at national level (UK). It aims to draw a clear typology of challenges, risks, limitations, barriers, all terms
used by the different stakeholders with diverse meanings and based on different motivations. Indeed the
issues and constraints faced by re-users of public data differ from the ones encountered by the public data
providers. Through the analysis of the experiences in opening data, we attempt to identify how barriers were
overcome and how risks were managed. Beyond passionate debates in favor or against Open Data, we
propose to consider the development of an Open Data initiative in terms of risks, contingency actions, and
expected opportunities. We therefore present in the next sections the risks to Open Data organized in 7
categories: (1) governance, (2) economic issues, (3) licenses and legal frameworks, (4) data characteristics, (5)
metadata, (6) access, and (7) skills.

Keywords: Open Data, open government, e-government

1 Introduction
Open Data has gained a lot of interest in the e-government communities over the last years, leading to the
implementation of many initiatives and platforms to publish Open Datasets in such areas as mobility (e.g., bus
timetables), security (e.g., crime rates), or economy (e.g., statistics on business creations). Open Data is an
essential tool for the dissemination of the open government principles. However its wider deployment
requires identifying the obstacles that have prevented e-government bodies either from implementing an
Open Data strategy or from ensuring its sustainability. We have therefore carried out a study between June
and November 2012, in which we analyzed three cases of Open Data development through their platforms, in
1 2 3
a medium size city (Rennes, France ), a large city (Berlin, Germany ), and at national level (UK ). In addition,
we have studied the context in which the Open Data movement has been developed across Europe, in
particular the type of data that have been opened, and the services that were developed by Open Data re-
users.

In this study, we aim to draw a clear typology of challenges, risks, limitations, barriers, all terms used by the
different stakeholders with diverse meanings and based on different motivations. Indeed the challenges and
constraints faced by re-users of public data differ from the ones encountered by the public data providers.
Through the analysis of the experiences in opening data in the UK and in the cities of Rennes in France and
Berlin in Germany, we attempt to identify how barriers were overcome and how risks were managed.

Beyond passionate debates in favor or against Open Data, we propose to consider the development of an
Open Data initiative in terms of risks, contingency actions, and expected opportunities. We therefore present
in the next sections the risks related to Open Data organized in 7 categories:
 governance,
 economic issues,
 licenses and legal frameworks,
 data characteristics,

1
http://www.data.rennes-metropole.fr/
2
http://daten.berlin.de/
3
http://data.gov.uk/
ISSN 1479-439X 348 ©Academic Publishing International Ltd
Reference this paper as: Sébastien Martin et al “Risk Analysis to Overcome Barriers to Open Data” Electronic
Journal of e-Government Volume 11 Issue 1 2013, (pp348 -359), available online at www.ejeg.com
 Sébastien Martin et al

 metadata,
 access, and
 skills.
2 Studied experiences
We analysed both Open Datasets and services created based on those datasets. We took into consideration
data catalogues from Rennes in France and Berlin in Germany. We used the conclusions drawn by Fraunhofer
(Both, 2012) which supported the creation of the Berlin platform. Analyses of services were only carried out
for Rennes and the United Kingdom, since the Berlin initiative was more recent.

Berlin (Germany) is a good example of Open Data at the local level, with the additional advantage of showing
the relationship between different administrative scales, the level of the city itself and the region (land). Berlin
platform has served as a model and prototype for the whole Germany and even beyond since it is integrated in
in a European research project. It was also prepared with a prospective study by the Fraunhofer Institute to
understand of the data opening process since its early stages. The first datasets are progressively added to the
portal, launched in September 2011.

Rennes, the administrative centre of the French region Bretagne, also provides an example of local Open Data.
The city led the opening of its data in a broader approach to innovation based on digital technologies. Among
the first initiatives in France, this is the first community to open a portal in 2010. It has set up an effective
support for reuse with a reuse competition and has a rather dense network of re-users.

The UK portal is open since January 2010 and centralizes data nationally. The British approach has also
resulted in the publication of reports and scientific articles. The British government has asked each
department to publish its Open Data strategy of opening, each one being inserted into the overall strategy
4
outlined in a White Paper . Data from the United Kingdom have also enabled the creation of a large number of
services that can be analysed to understand how the data were reused.

These case studies were chosen because they are exemplary Open Data initiatives at different geographical
levels and suggest paths to improvement the data opening process and the creation of new services.

Figure 1: Ishikawa diagram summarising risks and barriers related to data opening

4
http://data.gov.uk/library/open-data-white-paper

www.ejeg.com 349 ISSN 1479-439X

Electronic Journal of e-Government Volume 11 Issue 1 2013

Finally we reviewed existing studies and open data experiences as presented in the literature. Overall we
gathered key information on obstacles to data opening, perceived risks, contingency, remediation actions.

3 Risks related to governance issues

Opening data is the result of a political commitment. It raises risks related to the objectives and the
sustainability of the initiative.

3.1 Open Data vs. Open government, a misunderstanding

Yu and Robinson (Yu, 2012) highlight a misunderstanding caused by the confusion between open government
and Open Data, which are motivated by different objectives. In practice, this leads to consider Open Data as a
technical issue, whereas it is clearly not sufficient to engage in a process of open government. An open
government process should seek transparency and a profound change in the way in which public bodies
operate. Yu and Robinson grounded mostly their analysis on the American case and the first project of the
Obama administration in 2009. Within the framework they established, this risk concerns European initiatives
to various extents. In Rennes, the concept of open government is not even mentioned. Transparency is
discussed. However, the initiative clearly aims to create innovative services for citizens and to generate
economic value from data, while strengthening the attractiveness of the city and the region. In Berlin however,
5
open government is one of the stated objectives, of the Open Data initiative .

3.2 Reluctance of civil servants

In some cases, Open Data is perceived as a threat by civil servants. The increased control of citizens may lead
to protests against public actions, based on an adequate or inadequate interpretation of data which is often
de-contextualized. This fear can generate hostile attitudes and finally a reluctance of civil servants to take an
active part in the data opening process. In order to overcome this risk, certain actors have engaged in the early
mobilization of internal and external stakeholders (e.g., Kéolis in Rennes) and of civil society organizations in
order to prevent potential conflicts.

Alonso (Alonso, 2011) also points out the resistance to change in government administrations. He proposes to
showcase the social and economic benefits of opening data and to identify champions.

3.3 Inconsistency of public policies

A lack of consistency and perseverance in political behaviors can also put the initiative at risk. Re-users indeed
need to be confident that the Open Data policy will be sustainable so that datasets will be updated and
continuously available in case they use an API. If Open Data remains the project of a specific team, then it can
be questioned as soon as the political configuration changes. However, for the re-users to find a business
model and implement a sustainable service, it is necessary that data sources have a certain level of stability
and are maintained over time. Therefore, re-users can react heavily to any signal from the authorities. When in
6
France, it was discussed to include Etalab (in charge of the French Open Data platform ) in a wider agency,
many stakeholders expressed concerns about the willingness of the authorities to continue the Open Data
policy. However, if Open Data is rooted enough in the administrative culture and operations, if it is supported
by a cultural shift in public administrations (Davies, 2010), then it is possible to decrease the risks that the
Open Data policy will be overturned.

3.4 The relevant administrative level

Some initiatives have been taken at local level, others at national level. Each local authority makes diverse
choices regarding reuse conditions and formats for instance. Each initiative at local level opens datasets which
are best suited to its particular context. They raise a risk of fragmentation of the initiative at the expense of
the potential reuse of data released beyond the local territory.
A major challenge is therefore to find a balance in the intervention of different political levels, between state
intervention (central or federal) that should ensure the consistency of the released datasets and local
responsibilities. However, the coordination of efforts at European level (e.g., the European Thematic Network

5
« Open Data ist ein wichtiger Baustein des Open Government für eine transparente und bürgernahe Verwaltung. »
http://daten.berlin.de
6
http://www.etalab.gouv.fr/

www.ejeg.com 350 ©Academic Publishing International Ltd

 Sébastien Martin et al

7
on Legal Aspects of Public Sector Information ) and interoperability initiatives (e.g., SEMIC/JoinUp project) can
help overcome the fragmentation of projects in Europe.

3.5 The lack of dialogue between data providers and re-users

Another set of risks relates to the relationships between providers and end or intermediate users. It includes
the lack of dialogue with the users, the lack of information about the updates of already opened datasets, and
the lack of information about the future datasets to be opened.
8
In Berlin and Rennes, a dialogue has been formalized with groups of re-users (e.g., La Cantine numérique . It
9
can always be strengthened through events dedicated to Open Data (e.g., Berlin Open Data day 2012 ).
However, for both Rennes and Berlin, there is no focus on the last two aspects, i.e., the data update and the
new datasets to be released.
Identified risk Mitigation Contingency actions

Reluctance of civil servants Engage in the early mobilization of Ask officials to identify specific issues
internal and external stakeholders, that explain their reluctance.
and of civil society organizations.
Identify champions
Involve as many people as possible
Point out economic and social benefits
Integrate into day to day process

Inconsistency of public policies Favour a cultural shift in the

administrations

The relevant administrative level Find a balance in the intervention of

different political levels

Lack of dialogue between data Encourage regular meetings between

providers and re-users providers and re-users. Dedicate a
specific page for the announcement of
updates and future openings, and
allow re-users to give their opinions.

Table 1: Summary of risks related to governance

4 Risks related to economic issues: costs and return on investment

Despite the proliferation of studies on the benefits that can bring Open Data, few of them assess the costs and
benefits of each type of data, as it was attempted by the study led by the University of Victoria (Australia) for
the various impacts of spatial data and hydrological data (Houghton, 2022). The lack of common standards to
assess both the cost and benefits of opening data puts the sustainability of Open Data initiatives at risk.

4.1 The cost of opening data

The cost of opening data is not always calculated by taking into consideration the same type of resources.
Implementation costs include hardware, software and human resources. They go beyond the mere technical
work to take into consideration the implementation of a process. To overcome the costs of opening data, small
communities can mutualize their expenses or rely on national infrastructures. Table 2 shows an estimate of the
costs incurred for several platforms. The UK, which has already engaged more resources than most other
platforms at national level, has increased its expenses in December 2012 with the provision of a potential
10
credit of £ 8 million for the public bodies that have not yet met their objectives .

7
http://www.lapsi-project.eu/
8
http://www.lacantine-rennes.net/
9
http://berlin.opendataday.de/ueber/
10
http://news.bis.gov.uk/Press-Releases/New-funding-to-accelerate-benefits-of-open-data-684c1.aspx

www.ejeg.com 351 ISSN 1479-439X

Electronic Journal of e-Government Volume 11 Issue 1 2013

Platform Country Scope Scale Cost assessment

data.gov USA General National Around $10 million/year

data.gov.uk UK General National 2010-2011 : £1,2 million

11
2011-2012 : £2 million per year
12
Etalab France General National €5 million/year

Nantes Métropole France General Local € 100 000 (cost of the Portal)

PortalU Germany Environment National €750 000 / year

Table 2: Cost comparison of Open Data platforms

4.2 Benefits and return on investment

Finally, the way in which it is possible to demonstrate a return on investment is still debated. The uncertainty
on the extent and nature of the return on investment represents a clear risk for the sustainability of Open Data
initiatives. Very optimistic calculations advanced for example by the organizers of the Apps for democracy
13
contest only take into consideration a limited number of parameters : the contest itself deemed the value
drawn by the applications worth $2.000.000. The return on investment should take into consideration such
heterogeneous benefits as an increased service quality, transparency and trust by citizens, active citizenship
through a higher participation in political and public debates, as well as actual cost savings, and the generation
of economic activities. In order to match the economic and service related expectations when open datasets,
the development of applications and services based on Open Data sets is critical. Application contests help
data providers identify new ideas to reuse the data. However, as Janssen (2012) noted « little is known about
the conversion of public data into services of public value ». The way in which sustainable services can be
created from these initiatives would require further studies. Dedicated actions, such as challenges (e.g.,
hackathons), applications showcase and the availability of an API are among the key actions which can support
the development of applications and services based on Open Datasets (Chan, 2013). However a major issue
remains tracing reuse in order to measure benefits. Access through APIs allows tracing reuse more easily than
with publication of data dumps.

Finally, the return on investment should take into consideration the benefits to the administrations and public
institutions themselves, including benefits related to semantic interoperability (Alonso, 2011) and the change
of behaviours of public employees by ingesting a data sharing culture which benefits first of all to
eGovernment itself.

4.3 Sustainable business models for the production of data

Data creators are indirectly funded through taxation. They are sometimes also funded by the sale of data or
services created from these data. This has led to much debate in the UK about the opening of geographic data
from the Ordnance Survey. Releasing data for free on the Internet entails a risk to weaken the data production
process and jeopardize the data quality (Uhlir, 2009). Moreover, this might create a distortion in the
competition between companies, since certain companies have already established business models based on
data they have paid for. The onerous access to this data represents a barrier to entry for new competitors.
Opening data in this case entails a risk for the business model of companies that already use public data.

11
Retrieved from http://www.w3.org/2012/06/pmod/pmod2012_submission_19.pdf
12
Retrieved from http://www.journaldunet.com/ebusiness/le-net/budget-etalab-1111.shtml
13
http://www.govtech.com/e-government/Do-Apps-for-Democracy-and-Other.html

www.ejeg.com 352 ©Academic Publishing International Ltd

 Sébastien Martin et al

Identified risk Mitigation Contingency actions

The cost of opening data Assessing the costs of not opening Share part of the costs with other
Open Data platforms

Benefits and return on Use APIs which allow tracing reuse Adopt a realistic approach to costs
investment and benefits; Encourage stakeholders
who use Open Data to indicate that
use.
Take into consideration and measure
benefits to eGovernment and data
producers themselves

Sustainable business models for Promote networking between

the production of data stakeholders; participate in clusters
that sustain incubation of companies
grounding their business model on
Open Data

Table 3: Summary of risks related to the economic issues

5 Risks related to licenses and legal frameworks

5.1 Heterogeneous licenses across datasets

Legal constraints raise mainly the risk of fragmentation of Open Data at all levels, both within countries and
internationally, if the licenses and conditions for reuse are mutually incompatible. Indeed, a number of
services are based on multiple datasets (e.g., mashups) for which managing heterogeneous conditions of reuse
is very challenging. This risk is linked to an incomplete openness of data if re-use or commercial use of the
data is limited by the licenses. This was the case in Rennes and the United Kingdom with the first licenses. It
has been taken into account by the authorities. Thus, in 2011 the British changed their license. The former
license, Click-Use Licence designed by the Office of public sector information did not allow modifying data. The
new Open Government License overcomes this limitation by explicitly allowing changes in the data, which is
necessary for any complex form of reuse. All the same in Rennes in 2011, this has resulted in the publication of
the second version of the « licence de réutilisation des informations publiques Rennes Métropole en accès
libre ». In both cases, the modifications in the reuse conditions led to a greater openness. The example of the
United Kingdom shows the potential value of defining an Open Data initiative at the national level with a
partially top-down approach where the national catalogue not only aggregates data, but also defines a real
editorial policy. The UK catalog opens over 9000 datasets released with a single license. Coherent conditions
of reuse are expected to facilitate the reuse of data.

In Rennes, the whole datasets recorded were covered by the license Rennes Métropole V2 at the time of this
study. In Berlin, the datasets follow a slightly less homogeneous licensing model. Despite differences, in both
cases, licenses applied to datasets in Berlin are open, except for 4 datasets described as "keine Freie Lizenz"
(Table 4).
License Occurrences

Creative Commons Namensnennung – Creative commons attribution 54

Creative Commons Namensnennung - nicht – kommerziell – Creative commons attribution- 1

NonCommercial

Creative Commons Weitergabe unter gleichen Bedingungen - Share Alike 1

GNU - Lizenz für freie Dokumentation - GNU Free Documentation License 1

Keine Freie Lizenz – No free license 4

Table 4: Licenses in Berlin

www.ejeg.com 353 ISSN 1479-439X

Electronic Journal of e-Government Volume 11 Issue 1 2013

5.2 The stacking of rights over individual datasets

Certain datasets may also include data owned by multiple stakeholders with different policies. The stacking of
rights on a dataset happens when several organizations claim ownership or control over a dataset and the
conditions of its opening. Some may contest the opening and delay it. In Rennes, this has been prevented by
14
the early involvement of the local transporter, Keolis , a subsidiary of the French national train company
SNCF, in the data opening process.
15
Certain initiatives, such as the Europeana digital library attempt to enforce a single licence (CC0) so that
reusers do not have to work with multiple licenses and terms of use for datasets. On the opposite, the
16
Singapore Open Data platform focuses on the publication of clear licensing conditions and terms of use.

While most efforts currently focus on public data, Deloitte in its analysis (Deloitte, 2012) suggests that the
main value of Open Data will result from the combination of public data, business data, and personal data. The
creation of most services in the coming years would be based on combinations of datasets. This will likely lead
to even more complex situations regarding rights and licenses over datasets.

Identified risk Mitigation Contingency actions

Licence is not open enough Release data complying with the Collect the concerns of re-users and
definition of openness modify licenses if the barriers are too
constraining.

Heterogeneous licences across Awareness raising among Strengthen the role of the agency that
datasets stakeholders organizes Open Data
Enforce a clear policy to publish terms
of use

Stacking of rights Governance choices

Privacy Data anonymisation

Table 5: Summary of legal risks

6 Risks related to data

The data also represent risks related in particular to their reliability, their quality, and their format.

6.1 Data accuracy and bias

The dependence of data producers on State funding can raise suspicions on the accuracy of the data. Some
data can be sensitive to political pressure (e.g., unemployment figures) and the context in which they were
created may raise concerns regarding potential manipulations by the State.

The choice of the datasets to open is also critical to ensure a high return on investment. An initial phase of
data analysis is necessary in an Open Data project in order to ensure that the data which will be open both are
of high quality and have a high value for reusers (see Martin, 2013). For instance the Bluenove study in France
showed that the data most expected by companies are economic and commercial data as well as geographical
data (Bluenove, 2011).

6.2 Data quality as a result of a high quality production process

The discontinued funding of certain activities represents significant risks for the quality of the data. The case of
the Netherlands cadaster, even though it is not available as Open Data, shows the sensitivity of the datasets to

14
http://www.keolis.com/
15
http://www.europeana.eu/
16
http://data.gov.sg/

www.ejeg.com 354 ©Academic Publishing International Ltd

 Sébastien Martin et al

financial aspects. Entirely dependent on funds provided by the State, these were cut repeatedly over 1990’s,
which has led to a sharp deterioration in its quality (Uhlir, 2009).

Open Data advocates discard the risks regarding data quality by showing the opportunities of involving users in
the process of data improvement. By identifying errors and warning the data curators, re-users as well as any
citizen can contribute to maintain high quality datasets through crowdsourcing mechanisms.

6.3 Data available in heterogeneous formats

In order to efficiently access datasets, users must identify the appropriate software to read the data and work
with them, then to choose the best format according to their needs.
Data are made available in a variety of formats (Figure 2The Klessmann report (2012) indicates that
approximately 90% of the datasets in Germany are in PDF format, a format which presents the greatest
problems for reuse, but a large part (up to 56% depending on the organization) contains structured
information that could be made reusable by converting it for instance to the CSV format.
Some formats are proprietary and the combination of Open Data in proprietary formats incompatible with
each other already raises conversion difficulties. This also represents an indirect entry barrier for re-users who
wish to access the data but could not acquire the required software.

Figure 2 - Formats of datasets in Rennes (blue) and Berlin (red)

Access and reuse can therefore be facilitated if they are produced by software whose code is open (open
source software) and published in an open well documented format. In order to ensure that the format in
which data are made available is not an obstacle, Rennes and Berlin make many datasets available in multiple
formats (Table 67).
Datasets Minimum Maximum Average

Rennes 137 1 8 5.4

Berlin 61 1 9 2.3

Table 6: Formats by dataset in Rennes and Berlin

The policy of opening a dataset in multiple formats is however not systematic and depends on data
17
creators.Tim Berners-Lee proposes to evaluate Open Data according to criteria that give each dataset a rank
based on its openness and its reuse abilities. One limitation of the current Open Data is that most of the data
obtained at most three out of five stars, which in the frame of Tim Berners Lee limits the success of the
releases and restrains the value of the data.

17
http://5stardata.info

www.ejeg.com 355 ISSN 1479-439X

Electronic Journal of e-Government Volume 11 Issue 1 2013

Identified risk Mitigation Contingency actions

Data accuracy and bias Clarify the context of the data creation
process

Data quality as a result of a high Stabilize funding for the creation of

quality production process data and promote crowdsourcing

Data available in heterogeneous Publish datasets in various formats Develop guidelines and
formats but early to create incentive for later encourage standard formats
standardization across government bodies
Training in semantic
technologies

Table 7: Summary of data issues

7 Risks related to metadata

Metadata are assigned to describe datasets. They are very important for the retrieval and reuse of datasets.

7.1 Lack of single standard to describe datasets

18
For the description of datasets, metadata are most often formatted according to the Dublin Core and DCAT
19
vocabularies . However, there is no single standard to describe Open Datasets. Re-users have to deal with
multiple vocabularies. Coordination efforts are then necessary to overcome the difficulties raised by the
heterogeneity of metadata models used to describe Open Datasets. In France an initiative has been launched
20
to harmonize metadata from practices identified at the local level .

7.2 Incomplete metadata

The lack of metadata, the lack of mechanisms to ensure the quality of metadata, and the lack of information
on the objectives and means that have led to their creation or their aggregation also represent risks for the
efficient reuse of Open Datasets. For example there is often no information on how data were used in the first
21
instance. Generally, the documentation of data provenance and context which would allow interpreting the
data is critical.

In Rennes, the catalogue is being enriched with the recent addition of metadata properties, for instance the
precision of geographic data and their original reference system.

In Berlin, Both (2012) demonstrates the key role of metadata for the future of Open Data and even suggests
tracing reuses through metadata.
Identified risk Mitigation Contingency actions

Lack of single standard to Start with standards as early as Participate in the harmonization of
describe datasets possible metadata between Open Data
catalogues

Incomplete metadata Gather metadata needs from re-users;

implement mechanisms to trace the
provenance and use of datasets.

Table 8: Summary of metadata related risks

8 Risks related to access
Open Data should be accessed by both humans (end-users) and machines (through re-users). When setting up
an access interface, some platforms request users to register and log in to access the data. This can discourage

18
http://dublincore.org/documents/dces/
19
http://www.w3.org/TR/vocab-dcat/
20
http://opendata.montpelliernumerique.fr/Vers-une-harmonisation-des
21
http://www.w3.org/2011/prov/wiki/Main_Page

www.ejeg.com 356 ©Academic Publishing International Ltd

 Sébastien Martin et al

potential re-users by establishing tedious procedures. On the opposite if the platform does not impose any
identification, it becomes very difficult to know who is accessing what data and reusing it.

More and more, platforms enable access through APIs (e.g., data.gov in the United States) for re-users who
can then automatically access and update the datasets, instead of maintaining their own copy of the data.
They relieve service creators of the task of updating data. By ensuring that data used by service creators is up-
to-date, the data providers increase the quality of services. They also control better their reputation through
the accurate representation of their datasets. Nevertheless, the proportion of data accessible through APIs is
still low. For example, in Rennes, there are only five datasets opened through an API. These datasets are also
among the most used by applications created from public data. Although it is unclear that this is due to the
presence of an API (as they also happen to belong to the domain of mobility, highly popular among data re-
users), it suggests that APIs can indeed support the reusability of data.

Identified risk Mitigation Contingency actions

Balance between free access Provide all the data through an API
and the need to know the use capable of reporting access and use
of data

Table 9: Summary of access related risks

9 Risks related to user and re-user skills

While Alonso (Alonso, 2011) emphasizes the importance of training data producers in semantic technologies in
order to ensure the publication of high quality data, reuse also depends on the skills of potential reusers.
Indeed, the risks entailed by the implementation of an Open Data initiative also relate to the potential users
and re-users identified for the data.

In particular, analysing the skills of re-users can help understand how to facilitate the reuse and the type of
services that can be developed on top of the datasets.

9.1 The language barrier

Whereas for Berlin and Rennes, the question of multilingualism did not arise, other countries such as
Luxembourg or Belgium are multilingual. Moreover the creation of services at European level requires that the
data published from different countries be understood sufficiently by re-users to be retrieved and used
without any risk of misinterpretation. In Luxembourg, the vast majority of the datasets held by public
administrations are in French. While there are only few cases where data published in German are not also
available in a French version, the creation of transnational services requires implementing mechanisms to
guarantee the linguistic interoperability of datasets.

9.2 Skills related to information literacy and domain knowledge

Concerns have been raised regarding the ability of Open Data to equally benefit all social categories.
Benjamin, Bhuvaneswari and Rajan (2007) suggest that in some cases opening data can lead to a deterioration
of living conditions for a part of the population, while benefiting to a minority who had the necessary skills to
make use of the newly released information. In India, an open government project aimed to put online the
digitized cadastral data. The better-educated classes, largely those wealthiest, have seized this opportunity to
enhance their properties and challenge the rights of other owners. However, this type of risks is inherent to
any innovation and only reflects the extent of the digital divide.

The issue of skills is also related to the ability of stakeholders to generate profits from Open Datasets. It is also
represented in the concerns about the privatization of public data, with a few people grabbing what should be
a common wealth. In this regard, Chignard mentions genealogical data, which represent a very important
market (Chignard, 2012).

These risks are to a large extent beyond the scope of Open Data, in particular risks related to the level of
education and information literacy. However, they can be addressed through the development of data
visualisations, which can ease the understanding and interpretation of the phenomena described in datasets.

www.ejeg.com 357 ISSN 1479-439X

Electronic Journal of e-Government Volume 11 Issue 1 2013

In addition, education can help improve the skills of users and re-users through initiatives led by re-user groups
22
which organize training sessions to present available data and the tools and methods to work with them .

Identified risk Mitigation Contingency actions

The language barrier Publish data in multiple languages and

/ or fix the issue through metadata

Skills related to information Mention the data provenance and

literacy and domain their first use through metadata;
knowledge provide training to re-users during the
events around Open Data

Re-users are unfamiliar with Assessing metadata formats known by

metadata users

Table 10: Summary of skills related risks

10 Conclusion
Many reports have analysed the benefits of Open Data and report on Open Data initiatives. Janssen,
Charalabidis and Zuiderwijk (2012) insist on the myths that have accompanied the development of Open Data.
From a strategic perspective, Yu and Robinson explore the particular risks related to open government (Yu,
23
2012), while Lessig has early expressed reservations on the benefits that one can expect from transparency .

This calls for a more pragmatic approach grounded in demonstrated benefits and a clear assessment of the
risks associated with the implementation of an Open Data strategy. By analyzing the barriers and potential
benefits of Open Data, without the ambition of being exhaustive, we propose prevention measures and
contingency actions which can be taken. As illustrated by Alonso (Alonso, 2011), barriers to Open Data are not
technical. They are rather 1) cultural, 2) economic, 3) legal, and 4) semantic.

However, while Open Data is often considered at the level of general public policies, we note that not all types
of data raise the same risks and opportunities. The sale of certain types of datasets is potentially very
profitable, whereas others do not have existing markets. Rennes has to a large extent focused on geographic
data, while Berlin has opened many economic datasets. The services developed based on the datasets can
therefore be of very different nature, making all analyses on costs and benefits very difficult to apply across
cases. As noted by Martin (Martin, 2013) who led a survey on barriers to Open Data, the current focus is on
Open Data supply. This only represents one aspect to be tackled to match the promise of Open Data. It is
necessary to also investigate the creation of services based on Open Data so as to maximize the return on
investment of data producers and publishers.

The analysis in terms of return on investment is very different according to the type of data. However, specific
actions, such as the definition of complete and standardized metadata can enhance the potential for reuse of
datasets and therefore increase the return on investment, whichever the type of data that is considered.

All the same, different types of actors may perceive risks in a different way, due in particular to their local
context. Engaging in a risk management framework tailored to the specific context of data providers can help
considering Open Data beyond the traditional barriers highlighted by opponents. Most importantly, it
demonstrates the need to consider the deployment of an Open Data initiative as a long term process whose
sustainability can be improved through the evolution of all stakeholders: users and re-users through the
enhancement of skills and the creation of efficient associations; data creators through the prediction and
selection of formats necessary to enhance the reuse of data and the release of multiple data formats; finally
intermediary platforms such as national aggregators which can help overcome risks related to the
fragmentation of datasets, in technical, semantic, as well as legal terms.

22
Retrieved from http://lemag.lacantine-rennes.net/2012/10/atelier-infolab-a-la-chasse-aux-donnees-rennaises-de-
mobilite-1752
23
Retrieved from http://www.tnr.com/article/books-and-arts/against-transparency?page=0,0#

www.ejeg.com 358 ©Academic Publishing International Ltd

 Sébastien Martin et al

Future work will be dedicated to the study of the different types of datasets and services developed and the
way in which it is possible to optimize the return on investment of Open Data initiatives by selecting relevant
datasets and understanding the process by which successful services can be built on top of those datasets.
References
Alonso J.-M., (2011) “Open Government Data (approaches, concerns and barriers, lessons learned)”
Share-PSI workshop, Brussels.
Benjamin, S., Bhuvaneswari, R., Rajan, P. (2007) “Bhoomi:'E–governance', or, an anti–politics machine necessary to
globalize Bangalore?”, CASUM–m Working Paper.
Bluenove (2011). “Open Data : quels enjeux et opportunites pour l’entreprise ? “
Both, W. & Schieferdecker, I. (2012) Berliner Open Data-Strategie, Fraunhofer Verlag.
Chignard, S. ( 2012) Open Data: comprendre l’ouverture des données publiques. FYP.
Davies, T. (2010) “Open Data, democracy and public sector reform”. Available at:
http://www.opendataimpacts.net/report/wp-content/uploads/2010/08/How-is-open-government-data-being-used-
in-practice.pdf
Davies, T. & Bawa, Z. (2012) “The Promises and Perils of Open Government Data (OGD)”. The Journal of Community
Informatics. Available at: http://ci-journal.net/index.php/ciej/article/view/929/926
Deloitte. (2 12) Open Data driving growth, ingenuity and innovation. Deloitte analytics paper. Available at:
http://www.deloitte.com/assets/Dcom-
UnitedKingdom/Local%20Assets/Documents/Market%20insights/Deloitte%20Analytics/uk-insights-deloitte-analytics-
open-data-june-2012.pdf
Gilbert, D., Balestrini, P. & Littleboy, D., 2004. “Barriers and benefits in the adoption of e-government”. International
Journal of Public Sector Management Available at:
http://www.emeraldinsight.com/journals.htm?articleid=868029&show=abstract
Houghton, J., 2011. Costs and benefits of data provision. Melborne: Centre for Strategic Economic Studies (Victoria
University) Available at: https://www.oerknowledgecloud.com/sites/oerknowledgecloud.com/files/houghton-cost-
benefit-study.pdf
Janssen, M., 2012. “Benefits, Adoption Barriers and Myths of Open Data and Open Government”. Information Systems
Management.
Klessmann, J. ; Denker, P. ; Schieferdecker, I. ; Schulz, S., 2012. Open government data Deutschland. Eine Studie zu Open
Government in Deutschland im Auftrag des Bundesministerium des Innern. Berlin: Bundesministerium des Innern
(Germany).
Martin, Ch. (2013). “Understanding Barriers to Open Government Data”. Open Knowledge Foundation Blog.
Pasquier, M. & Villeneuve, J., 2007. “Organizational barriers to transparency a typology and analysis of organizational
behaviour tending to prevent or restrict access to information”. International Review of Administrative Science.
Uhlir, P.F., 2009. The Socioeconomic Effects of Public Sector Information on Digital Networks: Toward a Better
Understanding of Different Access and Reuse Policies. OECD.
Yu, H. & Robinson, D., 2012. “The New Ambiguity of Open Government”. Princeton CITP/Yale ISP Working Paper. Available
at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2012489&

www.ejeg.com 359 ISSN 1479-439X