Главные вкладки

Порталы открытых данных в мире

Тезисы выступления Игоря Кандыбы на первой лаборатории открытых данных в Минске 27 сентября. Доклад посвящен ситуации с раскрытием данных в разных странах мира и основан на материалах выступлений участников июльского Open Knowledge Festival в Берлине и других источниках.



Argentina is focused on interaction design as a way to identify the most appropriate interface to connect people with data.

Burkina Faso


Burkina Faso has just launched an open data platform, with data sets on health and education.
Developers are starting to build applications via an API based on the open data, such as a school selector based on the education data.




Canada launched an open data portal a year ago, connecting both French and English resources.
More than 190,000 data sets are included, many with APIs.
The government is particularly impressed by the winner of a recent hackathon, NewRoots (by Electric Sheep), which has built a startup around its app, which uses APIs to connect a range of data sets in order to help users identify where they might want to live. While attuned to Canadian open data sources, the idea is that using similar open data sets from around the globe, the app could be scaled to help residents make similar decisions around the globe.


Chile uses the Junar open data platform, which can also convert open data sets into APIs automatically.
The Chilean national government is working with cities to come up with a local government open data strategy, as Chilean cities rarely publish open data.
More than 300 people have registered for an upcoming marathon of inequality visualizations being planned by the government.
The government is also investing in a creating a Civic Innovation Lab to encourage use of open data.


City-level open data portals are emerging, with 300 data sets available, but these are mostly for data sets like places of interest.
China is ranked 61 out of 77 countries for openness of data. The government is still researching what portal to use to make open data available.
China has a contradiction in that data, when it is released, is often machine-readable, but the licensing rights for using open data are unclear.


Egypt has two main open data issues: Data formats are not very open, and the devil is in the details. For example, if you look at transport data, open data indexes show that it is available, but it doesn’t include rail or bus networks. It's the same with government spending data, which doesn't include military spending data.
There is no freedom-of-information legislation in Egypt.


France recently launched a new version of its open data platform.
It has appointed at the state level a chief data officer, who is focused on promoting data sharing across government agencies.
One of the most advanced projects is the French open API model, which provides the complete tax system in machine readable form, the OpenFisca project. Using the API or web service, it is possible to simulate various policy models and proposals to see the impact on French households, for example.



2014 has been a productive year.
Most data sets are available on the web (by searching Google, for example) but not via an open data platform.
None of the data is available in an API, machine-readable format yet.
“So therefore we have data that is open, but without knowledge.”


A national data-sharing policy was passed in 2012.
An open data platform was established in 2013.


The Indonesian government’s open data portal is in beta stage, with 631 data sets, and will be officially launched later this year.
Following recently held elections, general election data has been opened up to allow monitoring and counting of votes in a marked move to increase transparency in the archipelago.


While more data is being opened, volunteers must often push for access to the specific data they need.
A move to open up the budget via open data has revealed that there are a lot of budget transfers, some with changes of 400 to 500 percent during the year. So while the data is open, the reality is that the budget can change dramatically during the fiscal year.
Open data advocates launched a campaign to explain this and to ensure that data on budget transfers is similarly opened. They have since used APIs to create an app that alerts when a budget transfer is going to occur. Among the most interested app users are parliamentarians themselves, who want to stay informed of expected budget transfers ahead of voting on such decisions.


While 10,511 data sets have been released, there is a big divide between north and south in policy implementation (the bulk of open data occurs in the north).
Official guidelines on how government agencies should manage open data have been released.
Community groups have scraped data for 10,000 previously Mafia-owned buildings to enable these properties to be used by communities.


It launched an open data portal in late 2013.
More than 10,000 data sets are available.
Civic sector participation in opening data is growing rapidly, with 32 local governments participating in the recent Open Data Day.
The Japanese government has created standard terms of use for open data, but this isn’t compatible with existing Creative Commons licenses. One concern is that the standards say open data use is not authorized where it is “against public policy,” but this is not defined.


Last week, Mexico launched the beta site Data Squad.
The government’s open data policy was crowdsourced by the community, with polls conducted to identify which data was most important to open up.
So far, 10 strategic agencies are publishing more than 100 data sets on energy, agriculture, etc.
The government is now focused on community collaboration to spark innovation.


In the North-West Frontier Province, government officials are leading the launch of Code for Pakistan, which will establish a fellowship program that will bring coders to work with four government departments to build applications on top of open data.
Governments are making applications open source. Citizens are participating in a civic right-a-thon to document policy responses for frequently asked questions.


A big challenge is to get public institutions to publish data.
What is most effective in the country is that citizens are really active in suing the government in order to access information. These cases tend to be highly reported by the Polish media, showing the need and community support for open data. As a result, more action is coming from citizens releasing their data on a community-run open data portal.
The Polish government has just published an open data portal, but most of the data published is in PDF format.


Romanian data advocates have focused on creating three advocacy tools:
A coalition of programmers, activists and startups are in dialogue with public sector agencies to provide solutions to help open up data.
Advocates tend to approach embassies to encourage ambassadors to speak with the Romanian prime minister about the value of opening up data, as this is often a more effective advocacy route than community lobbyists approaching the government directly.
Public policy on open data is being introduced to make it less dependent on political will.
More than 100 data sets have been published nationally, and there is a department in charge of coordinating open data initiatives in Romania.


Switzerland launched an open data government portal last year and has had an open data government strategy in place since April.
There is strong government commitment to publishing data on a national platform and to collaborating with the open knowledge community to establish an open data culture in Switzerland.
An open transport API, which sees several hundred thousand calls made per day, has been developed. The API is provided for free, with long-term resourcing issues remaining unresolved.


An open data portal has been launched.
The nonprofit Social Boost is actively involved in using open data as a source of information for socially meaningful projects

United States

While there are good intentions for the opening up of data, progress has been slow. For example, any organization with funding over a certain level is required to open up its research data, but this hasn’t occurred yet, and there is no regulating of the open science agenda. However, the National Institutes of Health has recently stopped providing grants to projects until their research data is opened in accordance with the open science funding requirements.
The national Project Open Data is run on GitHub and provides a central resource to help government agencies develop open data strategies and implement common standards, including standards for open data licenses and for the creation of APIs.
Data.gov has more than 104,000 data sets uploaded, but a question remains as far as how useful many of them are.
Across the U.S., cities are increasingly opening up their data.


Cities sharing their data account for half of all new data sets added to the open data portal in the country.
The goal is for more structured release of data and to engage communities in using the open data—for example, using open data to measure the accountability of education and health services.