[Event Report] India, Lets build the list

The Bachchao Project in partnership with OONI hosted an online event on 9th and 10th October 2021 to update the Citizen Lab test list for India. The event, which was called “India, Lets build the list”, was organised to help strengthen community based monitoring of internet censorship in India. The event allowed experts from different fields to contribute to a curated list of websites that are relevant to India and which are regularly tested for censorship by volunteers in India.

Censorship in India, specifically online, has been evolving steadily since the notification of the Information Technology Act of of 2000 and its associated rules. Though the Act itself offers multiple ways in which the Government can remove content and/or block access to content (including shutting down internet services), very little data is available to confirm if due process is regularly followed in these matters. This  raises serious concerns about its impact on Indian citizens’ right to freedom of expression and access to information.

While many such blocked sites may fall in the expected categories of illegal streaming, adult content, file sharing etc., research has also shown that internet censorship in India also impacts a wide variety of other sites, such as news media and human rights sites.This list building and monitoring activity is therefore crucial for us as citizens and as a community of digtal rights practioners to safeguard the essence of a free internet and uphold the rule of law.

One open software project that aims to increase transparency of internet censorship (and other forms of network interference) around the world is Open Observatory of Network Interface (OONI). To this end, the project builds free and open source software – called OONI Probe – designed to measure various forms of network interference.

A recent study used the OONI Probe testing software to measure the blocking of websites in various states in India (such as Manipur and Bangalore) from January 2019 to January 2020. It found that while 136 sites from the Citizen Lab test list for India were confirmed to be blocked, the major decrepancies in access were between ISPs rather than between regions. A large number of media outlets seemed to be targeted for blocking as well.

As of now, a relatively small community in India reviews and contributes to the Citizen Lab test list for India, which means that it’s entirely possible that we are not looking at all the possible thematic areas in which website censorship may be happening.

It therefore becomes essential that more people from varied backgrounds and fields of interest support such open source testing for censorship. By reviewing and contributing to the the Citizen Lab test list for India, you can help ensure that a broad range of relevant websites are tested, and that the censorship measurement data collected from the testing of these websites is more comprehensive, robust, and timely. This will enable citizens to ask important questions to lawmakers and even mount legal challenges when necessary.

To this end, on Day 1 of our 2-day workshop, our OONI partners facilitated a session (“Introduction to Internet censorship”) which introduced participants to key concepts around internet censorship and how website censorship is implemented, with the goal of ultimately highlighting the importance of contributing to the Citizen Lab lists of websites that are measured for internet censorship. For the purposes of this workshop, the following forms of censorship were kept out of our scope:

  • Censorship on social media platforms
  • Internet outages/blackouts/shutdowns
  • Takedown requests
  • Online trolling
  • Self-censorship

We used these two days to specifically look at websites that may have been or could be at risk of being blocked by Internet Service Providers (ISPs) . The group discussed the recent history of internet censorship specifically related to blocking of sites under Section 69 (A) and Section 79 of the IT Act .  We also reviewed existing research and public advocacy efforts with regards to internet censorship in India.

The concept of the Citizen Lab Global Test List and India Test List, both hosted on Github, was introduced to the group. These lists are compiled and maintained as a voluntary global effort to monitor website censorship. The India test list has over 600 URLs  which fall under many of the Citizen Lab’s 30 standardized categories.

A review of this list showed that the list was not balanced in terms of URLs in each category. The list also needed an update based on recent events in the country. Our workshop was specifically aimed at rectifying this and making the list more comprehensive & inclusive of the myriad concerns of citizens of our country.

A few of the participants shared their own experience with state censorship and their work on these issues. One of them presented a list that they had compiled by testing for DNS hijacking of sites specifically on the ACT Fibernet. Another participant found that many official government websites are not accessible to people outside the nation and shared their own work on creating a proxy to allow researchers and others to access Indian government websites from other countries.Geo-blocking prevents archival by the Internet Archive, which many researchers depend on. Participants also shared their experience of studying the issue of internet access in conflict zones in India and that even though access to the internet is recognized as a human right, it is often on the very bottom of the priority list for communities who are facing very intense threats on the ground. They also shared that being able to help these communities understand that the role the internet can play in responding to some of the other threats they face (and the tools to enable this, while foregrounding their safety needs) had been a very positive, empowering experience for all involved.

To end Day 1, we dove into the methodology of list building and list pruning which was developed and presented to the group by our friends at Netallitica. This session was specifically aimed to prepare us for Day 2 during which we (the organizers and the attendees) split into groups and co-worked on updating the India test list.

We started Day 2 with practical inputs on how to make changes to this list, important points to remember so that anyone who looks at this list later to test or to clean it understands what changes have been made and why. Our partners from OONI also showcased their beta tool which will make updating the Citizen Lab test lists much easier (through a web platform, without requiring GitHub accounts), once it’s launched.

A total of 10 participants split into two online co-working groups and selected a single theme to work on for 30-minute hands-on sessions each. The participants selected themes based on their area of knowledge and interest and also on how much information the list for that theme already contained. The focus was to make each theme list cover a wider base making it representative of platforms/ sources of information/ interaction that are currently important in our country.

In each group there were discussions to decide which sites need to be added and/or removed, and how websites should be categorized . An important part of this exercise was to ensure that we are including sites that cater to various schools of thought so that the list is not skewed in its representation. This is important to do so that we can measure censorship across the board and not only of target sites that may be important to the world view of the people building and testing these lists.

Day 2 of the workshop resulted in the follow changes to the India test list :

Category Code (Name) New URLs added Updated to

HTTPS

Moved to Global list Recommended for deletion Domain Updated Category Updated
ECommerce 7 1 0 3 1 0
LGBT 15 0 0 1 0 0
Human Rights 8 0 0 0 0 0
Environment 31 1 0 0 0 1
Public Health 26 1 0 1 0 0
News Media 11 0 0 0 0 0
Terrorism & Militancy 0 0 0 1 0 0
Culture 19 1 0 0 1 1
Hate Speech 0 0 0 0 0 0
Political Criticism 4 0 0 1 0 1
Government 1 0 0 0 0 0
Pornography 5 0 0 0 0 0
Total 127
4
0
7
2
3

The participants were able to significantly add to the categories of LGBT, Environment , Culture and Public Health which were very sparsely polluted earlier.

Accomplishing this took time and effort to ensure no sites were repeated, URLs were added correctly, and that existing URLs in the list were still relevant. Our workshop focused specifically on contributing new URLs and we did not specifically set out to prune the existing list (though some of us took the initiative to look at this aspect too). Here is the pull request for this update: https://github.com/citizenlab/test-lists/pull/864

At the end of workshop, participants and us as organizers were enthused by the amount of understanding built about the importance of community based monitoring of internet censorship and a huge role that people from all walks of like can (and in our opinion, should) play to help technologists and digital rights advocates around the world to stand guard over a free Internet.

We hope that this effort will give impetus to more people to engage in these sort of open source list building and testing activities that will enable the generation of in-depth and representative data on the true nature of the Internet that citizens in India get to experience.

“Anatomy of Internet shutdowns”: Panel discussion at Nullcon 2020

Prateek Waghre, Research Analyst, The Takshashila Institution was a speaker in a panel discussion at Nullcon on March 8, 2020 about a study carried out jointly with Rohini Lakshané of The Bachchao Project. In the discussion entitled “Anatomy of Internet Shutdowns”, Waghre spoke about the study on usability testing of the whitelist issued for Internet access in Jammu and Kashmir.

Details about the study may be accessed here: http://thebachchaoproject.org/even-the-301-whitelisted-sites-in-jammu-and-kashmir-are-not-entirely-accessible-an-analysis

Details about the session here: https://nullcon.net/website/media-track.php. Nullcon is an annual conference held in India on the topic of cybersecurity.

[Podcast] All Things Policy: The Dark Side Of The Kashmir Whitelist

Rohini Lakshané (of The Bachchao Project) and Prateek Waghre (of The Takshashila Institution) spoke with Anirudh Kanisetti on this podcast about their analysis of the 301 entries whitelisted for Internet access in Jammu and Kashmir in January 2020. This episode is a part of the All Things Policy series of The Takshashila Institution. The detailed analysis and test results were published on Medianama — Even the 301 whitelisted sites in Jammu and Kashmir are not entirely accessible: An analysis.

Podcast URL: https://ivmpodcasts.com/all-things-policy-episode-list/2020/1/29/ep-250-the-dark-side-of-the-kashmir-whitelist

Even the 301 whitelisted sites in Jammu and Kashmir are not entirely accessible: An analysis

The article “Even the 301 whitelisted sites in Jammu and Kashmir are not entirely accessible: An analysis” written by Rohini Lakshané (The Bachchao Project) and Prateek Waghre (The Takshashila Institution) was published on Medianama on January 28, 2020.

https://www.medianama.com/2020/01/223-analysis-of-whitelisted-urls-in-jammu-and-kashmir-how-usable-are-they

Archive URL: https://web.archive.org/web/20200128191547/https://www.medianama.com/2020/01/223-analysis-of-whitelisted-urls-in-jammu-and-kashmir-how-usable-are-they

Excerpt

The Supreme Court made a judgement on January 10, 2020 directing the Central government to review the total suspension of Internet services in Jammu and Kashmir imposed since August 5, 2019 and to restore essential services. In response, the government of Jammu and Kashmir issued a whitelist comprising 153 entries on January 18, and increased the number of entries to 301 on January 24. What would the experience of an ordinary resident of Jammu and Kashmir be like under the whitelist arrangement? We conducted a preliminary analysis to empirically determine whether the 301 whitelisted websites and services would be practically usable and found that only 126 were usable to some degree. Before we delve further into the questions the list raises, the role of ISPs, and analyse the list itself, it is pertinent to understand the background and context in which an ordinary resident of Jammu and Kashmir may access the Internet…

Tweet thread: Preliminary analysis of second whitelist for Internet access in Jammu and Kashmir

Dataset: https://zenodo.org/record/3629633

In continuation of the tweet thread from: Preliminary analysis of first whitelist for Internet access in Jammu and Kashmir

Detailed analysis published on Medianama — Even the 301 whitelisted sites in Jammu and Kashmir are not entirely accessible: An analysis, January 28, 2020

Tweet thread

Tweet thread: Preliminary analysis of first whitelist for Internet access in Jammu and Kashmir

Rohini Lakshané (The Bachchao Project) and Prateek Waghre (The Takshashila Institution) conducted a preliminary analysis of whitelist comprising 153 entries issued by the Home Department, Government of Jammu and Kashmir on 18 January 2020, to empirically determine whether the whitelisted websites and services would be practically usable for an ordinary resident. The Twitter thread published here shines a light on the method of testing and the signficant findings. A detailed write-up will be published soon. The dataset comprising test results is licensed and can be accessed here: https://zenodo.org/record/3627665

Dataset: Analysis of whitelisted URLs in Jammu and Kashmir (order dated 24 January 2020)

This preliminary analysis was conducted by Rohini Lakshané (The Bachchao Project) and Prateek Waghre (The Takshashila Institution) from 22 and 26 January 2020 IST, to empirically determine whether the whitelisted websites and services would be practically usable for an ordinary resident of Jammu and Kashmir at the time of writing.

This dataset contains an analysis of a whitelist comprising 301 entries issued by the Home Department, Government of Jammu and Kashmir on 18 January 2020 [Order number: Home-05 (TSTS) of 2020]. The department issued an order with the accompanying whitelist in response to a Supreme Court judgement dated 10 January 2020 (Anuradha Bhasin vs. Union of Indian and Ors.) that directed the Government of India to review the blanket suspension of Internet services in Jammu and Kashmir since 5 August 2019. The first version of the whitelist (dated 18 January 2020), and this dataset by extension, comprised 153 entries. The Home Department states in its orders that this whitelist will be continually updated; the next update may be issued on 31 January or earlier.

A Chrome browser extension was used to simulate access to only those URLs that are mentioned in the government order. A detailed description of the method, its limitations, and the full analysis of the findings has been published on Medianama (Even the 301 whitelisted sites in Jammu and Kashmir are not entirely accessible: An analysis).

For information on how to read this dataset, refer to the tab entitled “About this sheet”. A numerical summary of the findings of this analysis is present in the tab entitled “Summary of findings”.

Data is provided AS-IS, without warranty as to accuracy or completeness.

This dataset has been released under the Creative Commons-Attribution-Share Alike (CC-BY-SA) 4.0 International License. All uses of the accompanying data and modifications and derivatives thereof must contain the following attribution: “By Rohini Lakshané and Prateek Waghre (2020)”.

This dataset was first published here on Zenodo.
DOI

Version 2

Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir, 27 January 2020

 

XLSX format: Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir.XLSX,_27 January 2020_CC-BY-SA-4.0

ODS format: Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir.ODS,_27 January 2020_CC-BY-SA-4.0.

Version 1

Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir, _24_January_2020

 

XLSX format: Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir.XLSX, 24_January_2020_CC-BY-SA-4.0

ODS format: Lakshane_Waghre_Analysis of whitelisted URLs in Jammu and Kashmir.ODS, 24_January_2020_CC-BY-SA-4.0

Last edited: 12:58 a.m., January 29, 2020, IST to add Version 2 of dataset.