How to determine if a dataset is sensitive or not?

I came across this excellent article titled, Best practices for sharing sensitive environmental geospatial data, by AMEC Earth and Environmental. It is a relevant article considering that at present there is no consistent and reliable mechanisms to access the political, social, economic sensitivity of the datasets. And secondly, the topic becomes even more important with the evolving discourse on the best practices an organisation may apply to identify sensitive datasets before sharing them with the community.

I had briefly touched on this topic in one of my previous blog (Circles of Open Government Data) wherein I attempted to visualise the complete OGD landscape to propose that out of all the datasets we must open datasets which have the least security or privacy concerns.

The article by AMEC Earth and Environmental takes this concept a step further to propose a set of factors to analyse when identifying datasets with security, privacy, legal and regulatory concerns. The article reviews international best practices and guides (such as 2006 Environics Survey) to develop the “generic” framework specifically focussing on the geospatial data. With this article, the author aims to provide “practical guidance to individuals and organisations interested in developing their own sensitive [environmental geospatial] data sharing policies and protocols.”

As per the article, the set of factors on which the geo-spatial datasets must be analysed to access its sensitivity are as follows:

  1. Legislation/Policies/Permits – Will opening up the dataset lead to any legal, regulatory, or policy issues?
  2. Confidentiality – Is the dataset considered confidential? Will opening this up be detrimental to interests of the stakeholders?
  3. Natural Resource Protection – Will the use of the information from the dataset lead to degradation of an environmentally significant site or resource?
  4. Cultural Protection – Is the dataset culturally sensitive? Will it lead to degradation of culturally significant site or resource?
  5. Safety and Security – Will opening up the dataset lead to safety and security concerns? Can the information be used to endanger public health and safety?

This report was commissioned by the Government of Canada, specifically GeoConnections, the program responsible for delivering Canada’s geospatial data to Canadians via the Canadian Geospatial Data Infrastructure.  The fact that the government is doing this kind of research and asking these questions is really encouraging.

To me article brings out two very important points:

Firstly, that the result of this decision making process to identify the sensitive datasets is highly subjective. As the perspectives on the sensitivity of the datasets vary may widely based on “context (time and recent events), an organization’s regulatory environment (legislation, policy, competition, etc.), jurisdictions and the personal views of Data Contributors/Owners/Custodians.”

Secondly, that the long term sharing of datasets must be based on the principles of “trust, risk management, the credibility of the participating organizations and their overriding desire to disseminate information”, which I believe are the key ingredients of any successful open government data initiative.

Although the factors outlined in the model are more relevant to the environmental geo-spatial datasets, with minor modifications these factors can be used for other datasets as well. After reading this article I think that we must aim to take this framework with its set of factors and try to formulate a much more generic framework based on which we can analyse the sensitiveness of the datasets. Just wondering what the readers think about this? It will be interesting to read comments on whether we can use some of these factors to analyse other datasets (such as crime data, governance data etc.)? What changes do we need to make to these factors to make this decision making process more suitable to these datasets?

Advertisements
Posted in Geo-spatial data, OGD Participation, Open Government Data | Leave a comment

DADOS – Launch of the Brazilian Open Government Data Portal

Last week I became aware of the wonderful initiative of the Ministry of Planning, Brazil to start the open government data portal in their country – dados.gov.br. The launch of the portal aims to

  1. Open up government data available using a central access point
  2. Improve transparency
  3. Contribute towards better policy making

It is the part of a larger government initiative called National Infrastructure of Open Data – INDA, where the government aims to set up the “standards to open data, promote training, and support public in the task of publishing open data.”

Currently, the open data portal has 78 datasets and 849 resources available for reuse. The portal also allows a medium to enter user feedback and publish news postings. An interesting aspect of the development of the initiative has been, a) use of free and open source tools; b) extensive social participation at all levels of development of this initiative.

The open data platform is based on Open Knowledge Foundation’s open-source data portal software CKAN. As a platform CKAN is extensively used by host of governments, for example, UK GovtEuropean CommissionCanadian Government etc. Even a few community driven websites, like French Open Data Catalogue is have been developed using the CKAN platform.

Moreover, the complete development process of open data portal did encourage the participation and collaboration of citizens, civil society, and bureaucracy. Right from the outset – planning meetings and development forums were open to any interested citizens. The open data portal did take the concept of social participation to the next level by using this synergy between government and citizens.

Now that the seeds for the start of the initiative have already been sown, what’s next? Of course in the long run, The Ministry of Planning in Brazil must aim to:

  1. Open more datasets
  2. Setup standards to open data
  3. Train government bodies on how to collect and share datasets
  4. Create awareness in civil society to encourage datasets reuse and create data visualisations

At the same time it is important that the open data catalogue must be made more feature-rich so that the datasets are easily accessible, analysable and accurate. Some of the features that can be added are –

The Ministry of Planning may look at incorporating a few of these features in their open data catalogue. I am curious to see how the initiative progresses.  To conclude – Tim Berners Lee, the founding father of internet, once famously said, “data underpins our society and our economy. “ I question – can we use this data to strengthen of the relationship between citizens, civil society, and the government?  Is it possible to achieve this by making government data open, accessible and analyzable? Only time may provide us with the answer!

Posted in ICT4D, Open Government Data | Leave a comment

Crowdfunding Critical Success Factor Model: The Case of Singapore

My colleague, Nigel Tan at the University of London recently sent me his dissertation titled “Crowdfunding Critical Success Factor Model: A case study in Singapore“. I find the research very insightful and a good resource to understand the relatively new domain of crowdfunding. In the dissertation Nigel attempts to identify the key success factors for a crowdfunding initiative by looking specifically at the case of Singapore and he does a great job.

In the research, Nigel described crowdfunding as a novel initiative where “small amounts combined with a large market of supporters can compare to funding provided by large, single organisations.” Nigel divided crowdfunding initiatives into two categories namely, “basic crowdfunding model with intermediary” (see figure 1) and “crowdsourcing without intermediary” (see figure 2).

Figure 1: Crowdfunding with Intermediary

 Although Nigel looks at crowdfunding initiatives in Singapore, for example, Angel’s Gate and Preparazzi, his research also points out initiatives like FundaGeekSellaband, and Kickstarter. Do have a look at these websites in case you are keen to know about how crowdfunding actually works.

Figure 2: Crowdfunding without Intermediary

Using the Crowdsourcing Critical Success Factor Model, Nigel brings out the similar characteristics of crowdsourcing and crowdfunding to identify the success factors for crowdfunding initiatives in Singapore. Nigel also conducted a survey to find out the relative importance of different factors towards the success of a crowdfunding initiative.

Nigel identifies Vision and Strategy, Human Capital, Infrastructure, Linkages & Trust and External Environment as the main factors affecting the success of a crowdfunding initiatives (Sharma, 2010). The results of the survey which has been based on responses from people all of whom were either active investors or self-managing their crowdfunding investment are as follows:

 Figure 3: Survey Results

Profitability is the main concern and that “shouldn’t come as a surprise as with any investment option profitability is the key determinant.” However, it is important to note that trust-building activities like periodic feedback through a social-medium, a portfolio of past crowdfunded projects, etc. is the second most important factor.

Based on the survey results and the literature review, Nigel concludes that similar factors determine success of a crowdsourcing and crowdfunding initiative. In his research, Nigel proposes that the critical success factors model may also be applied to Singapore but it needs to be “applied with a larger focus on Linkages and Trust.” Linkages and Trust has been found as the most important factor affecting the success of a crowdfunding initiative. From the perspective of Singapore it becomes even more important as Singapore boasts of high internet penetration high literacy rate so human capital and infrastructure and other factors are not bottlenecks.

The question I look forward to discuss with Nigel and readers of the blogs are –

  1. How applicable are the results of the research to crowdfunding initiatives in developing countries?
  2. If there are any examples of successful crowdfunding initiatives in the developing countries?
  3. What can practitioners do to increase the trust of the crowd in the crowdfunding initiative?
Posted in Crowdfunding, Crowdsourcing | Leave a comment

Circles of Open Government Data

Choosing what data sets to make available using Open Government Data is one of the main questions to consider when initiating an OGD initiative. The government, civil institutions, and social organizations all have limited resources so focusing efforts in the right direction from the outset becomes important. 

It is a relatively established fact that the best way to chart the future path of a project is to gather thoughts and opinions of its stakeholders which in case of OGD are social organizations, civil society, and government institutions. However, at the beginning of an OGD initiative especially in developing countries it may not be the best of ideas as:

a. OGD will still be in its infancy in those countries and social organizations, civil society, and government institutions may not have the right understanding of potential value and benefits OGD can bring

b. Owing to the vested interests of the area they work in, the social organizations maybe inclined to offer a partial opinion when asked about which data sets to open up

c. OGD aims to develop transient demands and satisfy unrealized needs so in most cases being demand driven or trying to achieve a definite end by use of a survey is not very ideal

 So key question is where to start? What data set to make openly accessible? Is there a simple way to visualize the entire government data landscape to then select the initial data set to open up?

 The circles of OGD (see figure 1) may help us find an answer to these questions. 

                                                               Figure 1: Circles of OGD


The biggest circle maybe visualized as consisting of all the machine readable datasets which the government of a specific nation has. Of all datasets, few may have security and privacy concerns (2nd circle). In an OGD initiative, it is important to take into account security and privacy issues and ensure that data is published in compliance with applicable laws and regulations. Therefore, it is advised not to open up the dataset residing in the 2nd circle within an OGD initiative.

In the figure, the remaining datasets in the 3rd circle have least security and privacy concerns hence, sharing them with the wider OGD community will be useful. However, I argue that the most appropriate idea will be to share a part of the datasets available within the 3rd circle.

Start with a small dataset and let it roll – simply open up a small data set to encourage participation. Just as a stone rolling down a hill gathers mass, small data sets will invite participation, encourage engagement and help us realize true benefits of an OGD implementation. 

It is critical to ensure that the data available must be easily accessible, analyzable, and accurate. Using the OGD portal, the civil society and social activists must get a medium to access the datasets, connect, discuss, share information. Once we reach the point where OGD portal has been setup and sufficient participation is present, then conducting a survey and gathering opinion from the wider OGD community on which datasets will be most useful to share can be useful. Hope we get to that stage in different countries with regards to OGD implementations sooner rather than later.

Posted in ICT4D, Open Government Data | 1 Comment

Improving Participation in an Open Government Data Initiative?

Ensuring significant public participation is one of the important success factors for an open government data (OGD) initiative. Metcalfe’s law states that the value of a network is proportional to the square of the number of users, therefore, the greater the number of users greater is the value. OGD is no different. The more people participate the more likely ideas/innovations will come to the forefront. In this article I try to ask these questions

– why people participate
– how to increase participation

in an OGD initiative with an aim to aid pracitioners create conditions for innovation OGD initiatives.

Of course, in most cases there aren’t direct benefits to participate in an OGD initiative which makes understanding this issue complex. However, similar domains do exist and their analysis can provide useful insights and lessons in understanding causes of participation. In fact, political participation models are one of those potential domains. Based on general incentives model, civic voluntarism model and the mobilization model I will attempt to structure this issue and bring to light key questions practitioners & policy makers must ask themselves and prepare for in order to understand and enhance public participation in OGD initiatives.

I took the approach of the civic voluntarism model to understand open government participation. Rather than looking at why people participate I (analysed) why won’t people participate and came up with three major reasons, 1) they aren’t aware of the initiative because of being outside the initiative’s information network, 2) they won’t due to lack of psychological engagement with the initiative and 3) they can’t owing to lack of resources and external stimuli. All these factors can contribute in one way or the other towards non-participation in an open government data initiative. It is important for government officials, bureaucracy, and OGD practitioners to understand these factors and identify steps to overcome them.

Creating awareness  by spreading the word about the initiative is a crucial step that determines the extent of participation. Awareness may be increased by the intelligent use of social media and promoting dialogue & information sharing through offline channels. Clear communication of the initiative’s aim and objectives will also improve the psychological engagement of the participants thereby leading to increase in participation. Taking steps for capacity building of the intended participants and ensuring an easy & less time consuming participation process is also an important factor that might improve participation.

Below is a table of the analysis summarising the reasons why people don’t participate and what can be done to improve participation.

The useful aspect of this analysis is that it can be extended to other “open call” or non-monetary compensation based participation models. Hope similar analysis see light of the day. I will like to encourage OGD practitioners to share their experiences and opinion on participation in an OGD initiative. A few question to get the ball rolling:

1. What steps can be taken to improve participation in an OGD initiative?
2. Is there a specific target group which is most active in OGD Initiatives? What are their characteristics? (eg – age group, education, interest, geographical location, political situation in the country etc.)
3. Is generating maximum participation always better?
4. Are there any other domains, eg – crowdsourcing, open source software etc, from which you have learnt lessons for implementation in an OGD initiative?

Posted in OGD Participation, Open Government Data | Tagged , | 2 Comments

Open Government Data – Separating “Data” from “Government”

Open Government Data (OGD) is the new hope. Many nations have developed, or are on the verge of developing open government data platforms with an aim to improve and potentially transform the manner in which their government is run. US, UK and even developing countries like Chile, Ghana have joined the open government data bandwagon. Obama’s Open Government Directive designed to implement the principles of “transparency, participation, collaboration” within the operations of US administration is a true testament of the value governments see in this initiative.

In spite of this focus unfortunately the terrain of open government data is still relatively unfamiliar. To achieve this goal, Yale ISP Working Paper titled, “The New Ambiguity of Open Government“, by Harlan Yu and David G. Robinson attempts to dissect the aspect of data from government to help us understand the domain of open government data better.

As per the paper, data and government are visualized as two dimensions of the open government data, and the key features of the open government data are:

  • Data Dimension – Easily accessible and reusable public sector data.
  • Government Dimension – Promote government transparency and enhance public delivery.

The data and the government dimension along with some OGD initiatives examples are shown in the matrix below.

Each of these initiatives has different levels of data adaptability and political/social nature, and hence, different open government data initiatives fall within different quadrants in the matrix. The governments must aspire to share “adaptable data” with an aim to enhance governmental transparency and “service delivery” to its citizens. Therefore, the more the initiatives fall in the upper half of the matrix the better is the open government data platform.

The paper gives the example of Hungarian cities of Budapest and Szeged which provide bus transit schedules and argues that such data is both open and governmental but this “has no bearing on Hungarian government’s troubling lack of transparency.” In light of the above example, the paper argues an important point that simply providing “platform independent, machine readable, and reusable data” does not make a government “open and transparent.” What matters is the nature of public sector data the governments make available?

The paper also separates the technological dimension of OGD from its political dimension and argues that “separating the ideal of adaptable data from that of transparent politics” is an important aspect to help us understand the nature of OGD and the benefits it can bring. Hence it’s critical that OGD practitioners and policy makers consider technology as an enabler and make room for human concerns as it’s not technology, but people, organizations and governments that will determine what benefits OGD brings to all. Certainly interesting times lie ahead on how OGD evolves in various nations!

Interested to view comments on how people feel

  • the domain of open government will evolve
  • what aspects of public sector data various government will share with its citizens and with what aims
  • will there be a difference in the nature of data made available OGD initiatives in developing and developed countries

*The matrix and all the quotes in the article are taken verbatim from the paper titled, “The New Ambiguity of Open Government.”

Posted in Open Government Data | Leave a comment

e-Dictatorship – Is it a possibility?

Right after the departure of Hosni Mubarak, Wael Ghonim, a facebook activist, famously said, “if you want to liberate a society just give them the internet, and if you want to free a society, again just give them internet!”

With its “rapidity of awareness“, internet and social media amplify and often legitimize movements to a large number of people than what was possible in the pre-internet and mobile era. As a result, internet and social media have been acknowledged to have played an important role in Arab Spring. In fact such has been the optimism about social media that during 2009 Iranian Green Revolution the world started believing that an internet microblogging service that didn’t exist five year prior to the revolution had the power to transform the history of an ancient Islamic nation.

Of course the role of internet in promoting free flow of information and ideas cannot be undermined, but on the other hand internet has the same power to abet institutions in constraining and manipulating the information flow. Numerous examples are there wherein the government or other central institutions have embraced “networked authoritarianism” to censor online conversations, deploy surveillance technologies, and communicate false information with an aim to monitor online conversations, identify opposition elements and steer public opinion. So, what if government starts using internet to strengthen its power? Let us look at this aspect of internet to bring out a few examples to illustrate that e-dictatorship or e-authoritarianism is as much possible as e-democracy.

Undemocratic countries, and in some cases, even democratic countries have been seen to use the power of the internet to filter information, perform surveillance and spread propaganda to track populace for sensitive information, monitor public discussion space and manage public opinion. China, Iran, North Korea, Russia, Syria, Cuba and Saudi Arabia are the most popular examples of states with internet censorships. However, recently a few democratic countries are implementing similar filtering, and surveillance mechanisms in their internet infrastructure. Even terrorist organizations like Al Shabab and Taliban have now started using social media to spread their propaganda.

Table below shows some examples of how different states are using internet censorship. It will be useful to point out that this table is an initial attempt to consolidate a list of internet censorship measures and if the readers can give more examples for this purpose, it will be highly appreciated.

Country

Filtering

Surveillance

Outreach

China
  • Filter search results
  • Bar individual websites with .cn domain
  • Block international news websites, anti-government websites, popular websites (like facebook, youtube, twitter etc), and human rights activist sites
  • Indigenous replicas of popular sites to ease tracking & censorship.
  • ISPs to monitor all web activity
  • Web Police monitors all web interactions
  • Pay and train bloggers to spread misinformation
Iran
  • Prior approval needed from Ministry of Culture and Islamic Culture for all websites
  • Block international news media, reformist political sites. In total 5 million websites are blocked.
  • Keep the cost of internet high to reduce wide penetration
  • Development of a domestic parallel “halal” internet. Halal internet aims at Islam at an ethical, & moral level.
  • State owned Telecommunication Company of Iran owns all internet infrastructure and controls & monitors all internet traffic
  • Identification details for internet access in internet café is mandatory
  • Crowd-source protestor identification after the green revolution
  • Recruit cyber army to spread false propaganda
  • Mass emails or texts to warn people from protesting
North Korea
  • Only allow access for websites on the domestic internet, Kwangmyong
  • Prior approval needed from the government for all websites.
  • Government control over all news media.
  • Keep internet and mobile phones prohibitively priced to reduce widespread penetration
  • Bar international phone calls
  • Monitoring of all web and telephonic communications
  • Use websites and other electronic media to widely promote government’s ideology, agenda and extol the virtues of the leader.
  • Fuel xenophobia myths by hiring propaganda artists
Russia
  • Working towards the creation of a domestic internet in Cyrillic
  • All ISPs by law provide cable link to the secret service
  • Security agencies access and monitor all internet transactions
  • Deploy internet brigades to spread government propaganda
Syria
  • Censor pro-Israel, hyper-Islamic, and anti government sites
  • Keep internet prohibitively priced to prevent widespread penetration
  • State owned Syrian Telecommunications Establishment owns all telecom infrastructure and controls & monitors all internet traffic
  • Identification details for internet access in internet café is mandatory
  • Unblock youtube access during protests to allegedly monitor opponents, and trace their location
  • Deploy internet army to spread pro-government opinion internationally
Cuba
  • Block US based websites run by Cuban dissidents
  • Low bandwidth connections to reduce internet usage
  • Pre-install computers with web-tracking tools
  • Special approval for computer purchase, which is given out selectively
  • Personal details are recorded for people accessing internet in an internet café
  • Government websites spread xenophobic propaganda and extol virtues of the leader
Saudi Arabia
  • Censor pro-Israel, anti-government, anti-Islamic, and websites criticizing ruling families
  • Prior approval needed from Ministry of Culture for all news websites.
India
  • Routing of all emails originating in India through servers in India for better monitoring

*All these facts are based on secondary research. Get in touch if you need more information on the facts cited in the article.

Posted in ICT4D, Internet Censorship, Iran | Leave a comment