Question: What exactly is the phenomenon you would like to study?
Answer: "Agnotology in times of Virality"
Question: How can we observe/measure it?
Answer: Agnotology1 as mentioned and used by Proctor's book, talks about campaigns from tobacco industry to keep their consumers in doubt and not loose them. Those were times when still media(print/digital) was the source of the news. With Internet, source of information has shifted and media keep exchanging its place with the audience(active citizen reporters). Some of approaches which could be tried:
- A very first step around which we can narrow down the hypothesis could be using news which is shared by people with mass following on social platforms like FaceBook and twitter, their source, propagation and response.
- Following advertorial articles in media, looking for declarative sentences and finding references of equivalent/derivative of same sentences in subsequent news, tweets, posts and response to them.
- Following election campaigns, people contesting, their personal platforms(apart from social media profiles) and articles being shared or written there.
Question: What research exists that potentially already has an answer?
In computer science work is being done to classify such a scenario as filter bubble2, 3. There are some political science groups working in related field4. Also online advertising uses these user behaviors(links visited, clicked, liked) for targeted ads, similarly, such traits of users could be used to re-assert "uncertainty".
Question: What could be possible results
- A derivative work in lines of recent article in EPW5 titled "Twitter and the Rebranding of Narendra Modi" where we map relationship between "hot-spots" on twitter/FB and trending topics.
-
Long-shot: A app/service on top of socialmedia, for start:
- given a image
- given a context and claim/fact
Could verify and assert or debunk the claim. Results should always be backed with reasons, citations. A most preliminary work could be done using information stored with image(GPS, type of camera/handset, date and timings) or doing DeepLearning with image and trying to locate similar images in sprawl of internet. Something like Straight Dope.
Context
Not everything on Internet is True.*
*And if people are not aware of this fact, is it getting misused.
As big bulk of population have jumped onto bandwagon of own-a-smartphone-frenzy, are we prepared to handle "news" being circulated via channels(watsapp, twitter, FB, more) which comes embedded6. Do they realize that not everything on Internet is True7, 8. For many this is their first experience of using internet, what kind of checks do we have in place for this medium. Please note, check doesn't mean regulation/censorship but AWARENESS.
Or is it the case where, we are being overtly cautious and not able to scale actual village, small town psyche, which has survived, evolved and flourished through all these generations.
Background:
As being said in this article9 "A rumour led to an accident." and how young generation has an access to these extra information10 apart from news and TV. Do we have the critical feedback loop in place which takes in every bit of news with a pinch of salt, and everything being shared on "groups"(watsapp, facebook) with rather fistful of salt11. While it is a possibility that now charged with the crime, they use this excuse to play innocence12, but is it actually the case.
Moreover, if it(understanding of how internet works) is missing and If social media is getting misused to pass things as news and facts, could we try to get such an evidence, if there is any, or else, publish and state the findings anyway. And there are many reports of misuse of this medium across countries and religions13, 14.
Following section is about various things which could be tried.
Solutions
Public officials.
Official representations of departments on such channels to authenticate credibility of statement.
- cons: there could be overwhelming demand.
- but that shouldn't stop from trying.
- have a system to rate user based on his/her previous raised requests.
- Public awareness camps
Community based, karma system.
Additional to the things getting circulated, having a community based group where a reported incident or fact can be verified. Reddit, wikipedia, stackoverflow, hacker news and other online communities(most of them are online). Based on your actions and engagement you get certain rights to moderate things in more balanced and transparent way.
Bursting the FilterBubble15
Troll busters16
I am not so sure about this "vigilante" approach, targeting users based on their political-party alignment(that too just one particular party), to me it is lopsided.
How to go about getting Data/Evidence
Collect data from Twitter, FaceBook, watsapp etc
Challenges:
- Will we be able to access data using simple APIs?
- Would it be still available, publicly? Often times, posts related to investigations are taken down.
- Can we reach out to these organizations for support, in terms of APIs and data?
Build a kind of graph of connected components.
In terms of:
- Who is following whom, is there a kind of "structure" to it.
- How does a news propagate through this mesh(in twitter terminology, organic impression and engagements).
- How does action of central-node ripple across branches.
- Can we map them with timeline, election polls, events, holidays
Challenges/Roadblocks:
- Data is very noisy, difficult to normalize.
- It would have different language and dialects.
- images/memes being shared.
Can we fit a mathematical model(non-linear dynamics) on such data?
I am not at all sure about what I mean by this. Just have read a few things about Ludwig's insect outbreak model and how it could fit a model on ecosystem of a forest to understand the "cycle"17.
On second thought, while the above mentioned ecosystem has dependency between worms and tree, I am not sure we will have similar variables(ghar-wapsi18/award-wapsi19??) at play with these incidents.
Footnotes:
http://www.newsweek.com/2015/04/03/manipulating-wikipedia-promote-bogus-business-school-316133.html Kolbe believes the site's flaws are felt more keenly outside the West. "People in the developing world don't always understand how Wikipedia is created. It's such a credible website, it comes so high up the search rankings-people think it's just another encyclopedia."
It's all over the internet." http://www.theguardian.com/world/2015/jan/29/love-jihad-india-one-man-quest-prevent-it