Read on Twitter
Today @Facebook published executive summaries of 3 #HRIAs done in #Cambodia, #SriLanka and #Indonesia. The modicum of transparency is welcome. But they leave a lot of questions unanswered. Here are a few (THREAD)
Was there no actual fieldwork in Indonesia? A country of 17,000 islands and 300 ethnic groups doesn’t experience technological change as a monolith, and ethnic/island disaggregation is urgently important to understand Facebook’s various impacts. (2/n)
35 people were interviewed. Picking the right 35 people to holistically understand Indonesian complexities would be near-impossible, even with great background country knowledge. https://about.fb.com/wp-content/uploads/2020/05/Indonesia-HRIA-Executive-Summary-v82.pdf (3/n)
Why was fieldwork conducted in 2018 but summaries are only coming out now? Much has changed in these countries over two years, as the authors all point out in their summaries. https://about.fb.com/news/2020/05/human-rights-work-in-asia/ (4/n)
In media-restricted landscapes, governments use facebook to control narratives, as the UN found in Myanmar. https://www.nytimes.com/2018/10/15/technology/myanmar-facebook-genocide.html (5/n)
Maybe the full versions of these HRIAs acknowledge that, but summaries don’t, so we can’t know if Facebook is proactive about the risks in these countries that it ignored in Myanmar. (6/n)
Facebook has partnered with mobile operators in Indonesia and Cambodia through “Free Basics” – an internet project that gives privileged access to certain sites for free/cheap. This has been called a violation of "net neutrality" by critics (e.g. the Indian government). (7/n)
Since local news never makes the cut for Free Basics sites, is Facebook restricting local voices and dissent in these countries? https://www.theguardian.com/technology/2017/jul/27/facebook-free-basics-developing-markets (8/n)
Also, how did assessors identify 'vulnerable populations'? Women, children & LGBTQ populations are often vulnerable. Religious minorities (noted in Sri Lanka but not elsewhere), too. But what about the poor? Those w/o internet? Rural residents? People on the wrong networks? (9/n)
Again, it is good that FB is publishing HRIA summaries and that it is screening the globe for risk areas. But I fear the delayed publication, narrow definition of rightsholders, lack of fieldwork are shortsighted. FB had a chance to learn so much more about its impacts. (10/n)
It seems like these assessments were scoped to the technology, and not scoped to the human rights and righstholders touched by it. How could they be, when fieldwork was so limited? Operators across industries struggle with this, and in 2020 I think they could do better. (11/11)
It would be great if @FarisNatour @findchloe @dunstanhope and others would/could weigh in (NDAs notwithstanding). The more that assessors share about methodology, the stronger the practice becomes. To that end, @TulikaBansal2, @ergonsteve and others may have reflections, too.
