Comments on Telangana Open Data Policy 2016


Arul George Scaria and Rishika Rangarajan


The Telangana State Open Data Policy 2016 is a positive step towards better governance through more open data. We would like to share herewith some of our comments and suggestions to make the policy more holistic and more inclusive. We need to emphasise in particular the need for a broader definition of open data and the role it can play in promoting social and technological innovations.


Preamble and Introduction

  1. The preamble as it stands now lacks vision in terms of the outcomes expected from a policy document. A preamble should be highlighting the broader goals and potential benefits of a policy, rather than the specific steps which are necessary for reaching those outcomes.
  2. As this is not a code (and as it is only a policy document), we recommend merging preamble and introduction to create a more comprehensive, concise and focused introduction.
  3. The proposed introduction section should highlight the context of the policy, including the potential benefits from open data (which are currently being discussed in Sec. 3).
  4. The proposed introduction section should clearly highlight the importance of open data and citizen participation in governance through open data.
  5. The proposed introduction section should also focus on the importance of citizen science and how open data can be used for identifying socially relevant problems.
  6. The discussion on benefits should also include the role open data can play in increasing collaborations as well as its role in promotion of social/ technological innovations.
  7. While the preamble and introduction parts of the policy document are giving emphasis on accessibility related aspects, the policy hasn’t given due importance to usability related aspects of data. While clause 1.6 highlights some of the important principles that open data is based upon, the importance of reusability, particularly the need for use without restrictions, needs to be emphasised.
  8. While advocating for transparency and open data, the proposed introduction section should also highlight the importance of protecting the privacy rights of citizens.
  9. The proposed introduction section should also highlight the importance of limiting the exceptions for the open approach.




The objectives section should be redrafted in the light of the suggestions mentioned above. The objectives in a policy document like this should be clearly highlighting the specific outputs expected from the policy.




  1. Definition of Data: The definition given for ‘data’ in this policy document is very narrow, particularly due to the existence of several conditions such as the requirement of preparation in a formalised manner. We would also like to point out that there is a general notion that open data is limited to data produced by ministries, government departments and subordinate offices. However it is important to explicitly include within the ambit of ‘data’, research data and research outputs produced by public funded institutions/ public funded projects also. This can play an extremely important role in promoting social and technological innovations. In this regard, we would like to highlight the recent step taken by the Government of Tamil Nadu which mandates all publications from public funded research/ educational institutions to be released under Creative Commons Licenses.[1]
  2. The present classification of data into multiple categories such as ‘Shareable Data’, ‘Negative List’, ‘Restricted Data’, ‘Non-Shareable Data’, etc. will lead to confusion among relevant stakeholders. To implement this policy effectively, we suggest having a simpler, two-tier classification of data – ‘Open Data’ and ‘Restricted Data’.
  3. Definition of ‘Restricted Data’: The term ‘Restricted Data’ can be defined as “any data exempted under Sections 8 and 9 of the Right to Information Act 2005.”
  4. Definition of ‘Open Data’: ‘Open Data’ should be defined as “any data which does not fall under the definition of ‘Restricted Data’. The term ‘open’ in this context signifies accessibility, reusability and non-existence/ minimal existence of IP restrictions. Accessibility means disabled friendly access and availability in online digital formats free of cost and in print formats at marginal costs. Reusability indicates the requirement to release the data in usable open formats.”
  5. Add definition of ‘Public Funded Institutions’: “Any institution which directly or indirectly receives funding from the government of Telangana for education or research activities.”
  6. Add definition of ‘Public Funded Projects’: “Any project which is funded directly or indirectly by the government of Telangana.”



Section 5: Non- Shareable Data

In the light of the recommendations we have made with regard to the definition of ‘open data’, we recommend deleting Section 5.


Section 6: Shareable Data

  1. This section can be simplified and a more appropriate title should be given for this section. We recommend “Data Access Infrastructure”.
  2. This section can start by highlighting that will be the official gateway for different datasets released under this policy.
  3. The common data sharing policy applicable for all the funding agencies under the Telangana government shall also be shared through this data aggregation portal.
  4. We appreciate the position taken in the policy that funding agencies should provide appropriate support and incentive for effective data management. We recommend that information about the support mechanisms and incentives made available in this regard should be disseminated through the Telangana data access portal.
  5. This section may also discuss the ‘data warehouse’ related aspects discussed currently under Section 8, which may now be deleted.


Section 7: Data Classification

  1. We suggest removing most parts of Section 7 in the light of our simplified two tier definition of data.
  2. However there are two important aspects that need to retained-
    1. Timely release of data: Data should released as early as practical.
    2. Appropriate attribution should be given to the authors/departments during any use of the data.
  1. In view of the changes suggested, we recommend removing the current heading and these two issues may be discussed under two separate headings – “Timely Release of Data” and “Attribution Requirements”.


Section 9: Current Legal Framework

In view of all the recommendations above, this section is redundant.


Section 10: Implementation Schedule

  1. Section 10 should be specifically limited to the implementation schedule.
  2. We recommend avoiding the usage of any arbitrary number (5 in the present document) for datasets that need to be released by all the departments within 3 months. We instead recommend mandating all departments and publicly funded institutions to release all their open datasets within a stipulated time period.
  3. Para. 2 of Section 10 can be moved to Section 6.


Section 11: Responsibilities of Database Owners/Data Generators/Controllers:

  1. We recommend redrafting of Section 11 in light of the recommendations given above, particularly the reusability requirement.
  2. We consider it inappropriate to specifically target students, in the context of the security related issues highlighted in Sec. 11(e).


  Section 12: Infrastructure provision:

Most of the aspects discussed under this Section can be covered under Section 11, through proper redrafting.


General remarks

  1. Constitution of a data management team:  To effectively implement any open data policy like this, data officers with adequate professional capabilities in the area of data management will be necessary. Those data officers should constantly monitor the implementation of this policy. The policy is currently silent on this aspect.
  2. As one may observe from the recommendations above, suitable modifications in the structure and organisation of the sections are necessary to make this policy more useful for relevant stakeholders.
  3. As already indicated in some of the above comments, reusability and accessibility are extremely important aspects of open data and the policy should give more emphasis on those aspects.
  4. We recommend adopting a uniform and liberal open license for all the datasets released under this policy.
  5. The policy should also go beyond the traditional notions of ‘open data’ and bring within its ambit data produced by public funded institutions/ public funded projects, to fuel more social/ technological innovations for socially relevant problems.