A Brief Colonial History Of Ceylon(SriLanka)
Sri Lanka: One Island Two Nations
A Brief Colonial History Of Ceylon(SriLanka)
Sri Lanka: One Island Two Nations
(Full Story)
Search This Blog
Back to 500BC.
==========================
Thiranjala Weerasinghe sj.- One Island Two Nations
?????????????????????????????????????????????????Thursday, April 11, 2019
As India votes, social media faces big test
10 April 2019
“This work is done across dozens of teams, both in India and around the globe, and began more than 18 months ago with a detailed planning and risk assessment across our platforms,” this is not the Indian government or a political party talking about the upcoming elections. No, not the Elections Commission either.
These are the words of Ajit Mohan, Managing Director and Vice President of Facebook in India talking about the platform’s preparations just three days before 900 million Indians were getting ready to vote during an election that would take over a month.
He added, “the findings allowed us to concentrate our work on key areas, including blocking and removing fake accounts; fighting the spread of misinformation; stopping abuse by domestic actors; spotting attempts at foreign meddling; and taking action against inauthentic coordinated campaigns.”
The Indian elections are going to be a real tester for Facebook (with over 400 million Indian users on Facebook and Whatsapp combined) in particular and other social media platforms in general on their role in forming public opinionWorld over every day, Facebook alone blocks or removes over a million accounts. There is no verified data on how many accounts, real and fake, are being created per day.
The Indian elections are going to be a real tester for Facebook (with over 400 million Indian users on Facebook and Whatsapp combined) in particular and other social media platforms in general on their role in forming public opinion. Facebook to a large extent and others like Twitter and Google-owned Youtube have been heavily scrutinized on how they tackle misinformation and manipulation efforts.
Ironically, the last time Indian PM Narendra Modi contested for the post, Facebook was almost part of the campaign. The 2014 election was dubbed India’s First Social Media Election and by extension Modi became India’s first social media PM. With 12 million plus followers on Twitter then, Modi was only second to US President Barak Obama in popularity. Modi is even more popular on Facebook now with 49.5 million followers. But the situation is poles apart from what it was five years back.
After what happened at the last US elections, Facebook has tried hard make sure that it does not in any way appear as close as it was to Modi is 2014. It has taken down thousands of Indian accounts, with millions of followers for misbehaviour meaning, these accounts together were sharing posts in such a way to create more attention on them. The biggest losses in these sweeps were suffered by Modi supporters.
Facebook has also been more transparent in who pays for its ads. But as with every patch-up Facebook has come up with, serious flaws remain. For one the money trail stops at the account who paid for the ad, Facebook cannot seek more details. On a more serious note, recent research has shown that Facebook’s ad placement algorithm itself is skewed.
The 2014 election was dubbed India’s First Social Media Election and by extension Modi became India’s first social media PM. With 12 million plus followers on Twitter then, Modi was only second to US President Barak Obama popularitynow with 49.5 million followers“To accomplish this (ad placements), the platforms build extensive user interest profiles and track ad performance to understand how different users interact with different ads. This historical data is then used to steer future ads towards those users who are most likely to be interested in them, and to users like them,” a research paper titled Discrimination through optimization: How Facebook’s ad delivery can lead to skewed outcomes revealed.
“By running experiments where we swap different ad headlines, text, and images, we demonstrate that the differences in ad delivery can be significantly affected by the image alone. For example, an ad whose headline and text would stereo-typically be of the most interest to men with the image that would stereo-typically be of the most interest to women delivers primarily to women at the same rate as when all three ad creative components are stereo-typically of the most interest to women.”
The author is the Asia-Pacific Coordinator for the DART Centre for Journalism and Trauma, a project of the Columbia Journalism SchoolTwitter - @amanthap