March 14 2024 GM

From TCU Wiki
Glitter Meetups

Open Data and Civic Tech to Circumvent Government Censorship and Disinformation in Latin America

We will talk about how open data and civic tech improves people's access to information, and institutions and authorities accountability. We will also learn about, a new Latin American organization based in México, focused on generating knowledge from open data. They are working with access to public information, political candidates, open contracting and beneficial ownership information, supporting the strategies of other regional an international organizations and publishing information in the open web to overcome the usability challenges of government data.

  • Featured guest bio: Martín Szyszlican is a programmer since childhood and has been developing civic tech in Latin America (mostly in Argentina and México) for the past 12 years. He studied design (UBA) and specialized in UX (UTN). He is a co-founder of and a member of

What is Glitter Meetup?

Glitter Meetup is the weekly town hall of the Internet Freedom community at the IF Square on the TCU Mattermost, at 9am EST / 2pm UTC. It is a text-based chat where digital rights defenders can share regional and projects updates, expertise, ask questions, and connect with others from all over the world! Do you need an invite? Learn how to get one here.


Could you introduce yourself for people who are entering the chat please?
  • I'm Martín Szyszlican, software developer and user experience designer, originally from Argentina, now living in México.
  • My first civic tech project was in 2010 and I have been working on it for the past 15 years.
Currently you are in Abrimos, could you talk about it and what is it working on currently?
  • We are a new organization in México, with lots of combined experience helping news organizations and NGOs.
  • Our slogan is "creating strategic knowledge from data". But we actually create websites, dashboards, reports and send daily alerts.
  • We already have processed some data sources for Latin America, we can also work with any dataset an organization needs.
  • Our main projects now are freedom of information, public contracting, candidate data and our data transformation software.
  • We have processed contracting data for seven different countries, and counting... In this case we have an algorithm to detect risks of corruption.
What do you mean with contracting data? public contracts with private companies, for example? And could you explain a bit more about this algorithm to detect risks of corruption please?
  • Yes, when state money (tax money by every citizen) is spent, the government needs to publish information about the contract, and ideally, the full contract itself.
  • Different countries do better or worse at this public procurement transparency, but we scrape the data or use what they have and send it through are flagfetti software.
  • Flagfetti is configurable for each country and if produces an output evaluation for each of a countries contracts (often several million documents).
  • So we then rank the offices with more red flags and the suppliers with most red flags, and give them to journalists or organizations to continue the research and improve public policy or straight up denounce them.
  • Hopefully this will make states more efficient in spending, so more able to comply with the rights of their citizens.
Who did this flagfetti software?
  • We coded it, it's written in JavaScript and customizable via YAML, it has a GPL license.
You said that it ranks the offices and suppliers with more red flags related to corruption, and then the team gives them to journalists or organizations. What are the main cases that you are taking care of in Mexico and Latin America?
  • We have three main lines:
    • I've already mentioned our public procurement information, we have the full Guatemalan data published at and we have also crossed OFAC sanctions with national Mexican sanctions at
    • Then we are helping improve Mexican access to public information... we have 8.5 million FOI requests at for this we do daily email alerts, dashboards, research training and a website.
    • Finally, our candidate information for Argentina, Colombia and México at, is about helping researchers and voters have more information on democracy.
  • We can also do this for any data source that is interesting to an organization, thanks to our TeseoETL system that we are offering as consultants.
Could you speak a bit more about how you collaborate with journalists?
  • We have two journalists in our organization, so they usually find other journalists to collaborate on projects.
  • We process the data sources and give them access to our dashboards.
  • Then they find the story in the data and write.
  • There's usually heavy fact checking at some stage, because data can be deceiving or have errors at any stage, and you don't want to be wrong by a comma when you're publishing a big story.
  • Also we help them make FOI (Freedom Of Information) requests to get the information they need that is not published, and we now send them every new answer that is relevant to their line of work as a daily email alert.
  • The alerts work with keywords, they give us a list of keywords and every day we check for new answers and send it to them.
  • In Mexico we have around 1000 new answers every day, 8.5 million historical answers.
You have a full infrastructure for this, did your team found out how to make this a workflow with journalists?
  • Yes, we have been refining it over the years, we have our own servers and everything runs on free software.
Could you describe what is happening in Argentina with the digital public information and how this is attempting against digital rights?
  • Right, there are very strong budget cuts across all of Argentinean government, because of the ultra-liberal "anarcho-capitalist" president that recently took office.
  • There are some websites that are being stripped or shut down too, like the one for sexual education school materials and the national news agency.
  • A few months back, as a community project not related to, we started backing up all govt websites just in case... we actually saw what was coming.
  • In the past few weeks we have been releasing those archived websites to the internet archive so they are now available in perpetuity.
  • There's a communiqué there, with a strong call to action to preserve common digital goods (aka websites).
How did you know the new government will shut down these websites and how did you back them up?
  • They were doing campaign with a chainsaw and saying that they were going to cut everything, and you know, digital rights are human rights... so we started making a full list of govt sites to backup.
  • There are many techniques for the backup, I have not been involved in the technical side of it, because I'm outside the country.
  • At the beginning they were just using wget to crawl the websites, but they started using the warc format, this has many advantages.
  • Warc is used by the internet archive, so, together with a group called Archive Team, we send the archived websites to them, and they publish it directly on the Internet Archive's WayBack Machine.
  • This means that they are "easy" to find, at least for anyone who knows what the internet archive is, and also we can just publish a link to it, and we know they Archive's got our backs with copyright claims or whatnot.
Could you give us your opinion about how promoting transparency and preserving digital public goods is related to digital rights?
  • Privacy is actually relevant to everything we do... such as making sure data published by the state is accurate and does not contain private information that shouldn't be public, and being able to access websites that are sometimes very limited. Also, we have detected and reported privacy leaks in state platforms.
  • But more widely, as we believe that digital rights are human rights, and we use government data (published on the internet) to advance human rights, we are doing digital rights.
  • For instance, our project is based on data produced by the freedom of information requests, this is a right Mexicans exercise though the internet and have a right to privacy if they desire it...
  • We have helped them stay more private by securing the endpoints that were leaking their unpublished requests on the platform to other users.
  • Also, since we are publishing the entirety of the requests, we are working with those whose names are there (an ilegal privacy leak by the govt) to help them remove it.
How can we contact you?
  • You are welcome to contact me at
  • I am @martinszy on Mattermost