[IS/MDIA 590]Yohta's Workspace-Community Data

Week13(4/17)

"Data Voids: Where Missing Data Can Easily Be Exploited," Data & Society 2018 Report. 


The potential risk of data void leads to polarization. Luck of multi-aspect information could misleadingly manipulate people.
"A rising trend in online misinformation is to encourage users to search for a topic for which the motivated manipulator knows that only one point of view will be represented."(p.4)

With the following system, search engines seem self-reinforcing by nature. If a greater amount of contents were generated by mass people with low literacy, that would take up a significant amount of portion online information.
"The more frequently a particular phrase is used, the more likely it will be suggested in auto-suggest. For example, at the time of writing, Bing’s auto-suggest for “subway” includes the following phrases"(p.6)


"Likewise, implicit bias and the amplification of problematic societal attitudes often contribute to the production of data voids even when there’s no maliciously intended actor. For example, image searches for terms like “CEO” tend to return dozens of pictures of white men."(p.7)

[Data voids type]
• Data voids that are actively weaponized by adversarial actors immediately following a breaking news event (e.g., “Sutherland Springs” or “Parkland.”)

• Data voids that are actively weaponized by adversarial actors around problematic search terms, (e.g., “black on white crime” )

• Data voids that passively reflect bias or prejudice in society but are not actively being weaponized (e.g., “CEO.”)

[Key Term]

Data voids:
"There are many search terms for which the available relevant data is limited, non-existent, or deeply problematic. We call these “data voids.”(p.2)
------------------------------------------------------------------------------------------------------------------------

"Searching for Alternative Facts: Analyzing Scriptural Inference in Conservative News Practices," Data & Society 2018 Report

"Through my interviews and observations a clear pattern emerged that comprised conservatism as “faith, family, the Constitution, and national security.”(p.8)

"The perspective that separation of church and state is about protecting the church from government encroachment not exclusive to my respondents; it is a central belief for Senator Cruz,19 Vice President Pence, and newly appointed Supreme Court Justice Gorsuch (Grossman 2015). "(p.12)

"For those I observed, Protestantism and patriotism are inextricably linked to protectionism.~For them, there is historical evidence that the US is fundamentally, rather than incidentally,  Christian, and this belief spills over into how conservatives conceptualize the need to “protect” their nation."(p.14)

"Commitment to protecting the nation was discussed in terms of physical safety, but revolved around the protection of religious (i.e., Christian) and racial (i.e., white) identity."(p.14)

"The influence of Christian beliefs on conservative rituals does not end with Prayer & Pledge. In fact, one of the most important and consistent findings of my research was how conservatives utilize theological teachings to unpack texts like the Constitution or other forms of media. I label these media literacy practices scriptural inference, and argue that understanding this conservative approach to Biblical interpretation can shed light on contemporary solutions to “fake news.”(p.18)

"Google obviously has a vested interest in users spending more time on their platforms, so is likely to recommend stories or websites to users based on their previous browsing (Castells 2013)."(p.33)

[Key Term]

Scriptural Inference:

Drawing from a conservative Christian worldview, these groups critically interrogate media messages in the same way they approach the Bible.


Reverse racism:
The study argues that white people are feeling resentful of political correctness because they are not allowed to say what they are thinking or are not sure how to behave in interracial situations. (p.45)
------------------------------------------------------------------------------------------

"Content of Context Moderation: Artisanal, Community-Reliant, and Industrial Approaches," Data & Society 2019 Report.

1) Artisanal, for platforms such as Vimeo, Medium, Patreon, or Discord; 
2) Community-Reliant, for platforms such as Wikimedia and Reddit; and 
3) Industrial approaches, for platforms such as Facebook or Google.

"the greatest tension identified was how to be consistent in developing rules geared toward hate speech and disinformation, which tends to be highly contingent on local contexts and power dynamics."(p.14)

"The need to balance between consistency of rules, with being sensitive to local contexts, particularly for issues like hate speech and disinformation, is of particular concern when considering both the design of platforms and issues of scale."(p.14)

------------------------------------------------------------------------------------------------------------------------

Question

These three Data & Society White Paper Reports seemed to me analyzed three key roles of digital communication:1. As mentioned in "Searching for Alternative Facts", our assumption is that showing contents based on a user's browsing history would lead to more usage of their platform. But is this always the case?

I'm skeptical that a majority of users have a specific viewpoint so that they prefer certain contents inclined to their perspective. Among the silent majority, there would be a significant amount of users who don't have a specific opinion or viewpoint, so that takes the opposite/different viewpoint positively.
2. Are there any cases in the States that monetize these type of anti-echo chamber/ algorithm into a sustainable business model?
I could find some example from overseas, Romania.