Media

Category: Media

Read the original document

1. Ideology in media organizations

Problems
1. See “How Does Ideology Permeate a Story?”
2. Lack of multi-perspective presentation
3. Non-profits with political motivations are biased (e.g. Koch Brothers strategy for long-term permeation of libertarian ideology started with non-profits, Heritage Foundation, etc.)
Solutions
1. Create meta-media organizations that aggregates the best side of each issue
  1. Create a strong-man argument site
2. Talk explicitly about both the issues and the way that the issues are being discussed by major sides of the conversation
3. Present multiple perspectives
  1. Hire more journalists on right
  2. Hire more academics on right (source)
  3. Create non-partisan political non-profits
4. General Perspective Points
  1. Always present the strong-man/steel man of the other side’s arguments
  2. Beware binaries, embrace greater complexity: e.g. Republican / Democrat; actually there are multiple voting blocs within the Republican and Democratic parties that have different demographics
Problems with Solutions
1. Getting adoption of a meta-media aggregator, and technical cost in setting it up
2. People aren’t as interested in nuanced discussion that requires thoughtfulness as they are in more emotive content
News optimizes for attention
1. Problems
  1. Media with high emotional valence outperforms other news
  2. Media as emotional valence + content rather than just true content or just an opinionated valence on the content
  3. Entertainment and news are conflated
  4. Sensational stories are overrepresented, leading to an absence of coverage of important but less emotional issues.
2. Solutions
  1. Individual guidelines for political info diet:
    1. Read long-form, in detail
      1. As an individual, convert to books and long form articles exclusively
    2. Seek unemotional reporting
      1. Select from media sources explicitly focused on unemotional / fact based reporting
    3. Take info and perspectives from at least two major sides, if not more, and sub-“sides” (and don’t assume there are only two sides)
  2. Focus on orienting to true information instead of consuming more information
3. Problems with Solutions
  1. Problems with political info dieting:
    1. Time Costs
      1. Takes approx. double time to actually watch both sides to an issue (assuming there are only two sides to the issue, which is a usually a false binary)
      2. Primary materials take too long to absorb. That is why people pay news sources they trust to summarize and present it to them.
    2. Access
      1. Institutions like media companies can develop access to truth, contacts, institutionally develop tools for other information-gathering. Similar to how an individual $10 donor cannot get a charity’s details in the way B&M Foundation or Givewell can.
Incentive Misalignment of Primary Sources
1. Problems
  1. Ex. Syria
2. Solutions
  1. Layer fact checking on top of the content, say crowdsourced highlighting of content with sources / intelligent debate over the facts [Wikipedia overlay for the internet]
  2. Multiple angles of primary sources in the same piece (common solution)
    1. Use historical sources of similar arguments to get past current tribalism (Professor David Moss explicitly is trying to do this, starting in an HBS case-based course Jesse took on American history)
  3. 1. Problems with Solutions
  4. Getting adoption of a layered crowdsourced content site, getting people to contribute
  5. Often multiple angles of primary sources don’t exist / hard to find
Inaccuracy
1. Problems
  1. Sources:
    1. Primary Source misalignment
    2. Bias in journalism through data collection, confirmation bias, availability, etc.
    3. Lack of information filled with opinion of Journalist
    4. Facts are difficult to find + check
      1. Data unavailable
      2. Data requires transformation (controls, corrections) which takes time + resources
  2. Are generally left uncorrected
2. Solutions
  1. Bias inside organizations can be attenuated by having a team explicitly focused on rooting out ideological and biased claims and affect in the organization's pieces
  2. Better training for journalists in how to read and interpret academics’ research
    1. Shorenstein Center’s Journalist’s Resource attempts to offer tips
  3. Research data could be aggregated and made public in an accessible way through easy to use APIs and downloads (certainly all gov. Funded research)
    1. Modernization of many government entities that generate data, making that data painfully easy to access
3. Problems with Solutions
  1. Having a team eliminate ideological claims isn’t in the benefit of the organization, whose consumers demand ideological / emotion captivating content
Short / Fast turnover of content and attention
1. Problems
  1. Shortening of attention spans due to internet / availability of content
  2. Training of consumers of media to expect short bursts of emotional content
2. Solutions
  1. Focus on orienting to true information instead of consuming more information
    1. There’s a distinction between gathering new information and creating a model of a system that adjusts in response to information. The fast turnover of information leads to a situation where the consumer is disoriented, surrounded by information that they haven’t come to grips with and attempting to drink from a firehose of new information that makes it hard to orient to the old information. We have to tradeoff updating our model for systems against gathering new information with which to update, and this environment leads to choosing new information much more often than improving a model. And so solutions that promote creating and updating a model (here is how the system works, rather than here’s something that happened to / within the system) lead to better outcomes
  2. Organize information based on issue, not current events / occurrences within the current news cycle. Updates and expands issues as new information appears. However, current news on, e.g., oil is always presented within context of greater oil context (e.g. history, arguments, stakeholders, dilemmas, current popularity of different positions).
    1. Create a resource which collects information in this way and incentivizes people to learn about an issue holistically instead of Ie. Anton’s Chair/Conscious
    2. Have a module that has all major news sources, books, academic papers, videos, documentaries, audio interviews, etc, non-profits,
    3. Have tree diagrams (of whichever type works best) that break down the problem, each of which can go into their own breakdown (like in a Workflowy document).
  3. Better educate the society. Education provides at least some model and context for “elaborating” (memory literature term) the new information into a representative structure.
    1. 1. High Quality doesn’t pass a cost / benefit test
3. Problems
  1. As the public is misinformed, low quality & inacurate content can pass unnoticed
  2. Increasingly a problem as media organizations (“legacy” media) seek to maintain profits, having lost attention share to the internet
  3. Stakeholders, currently the GOP and Trump administration, can use a strategy that defames the mainstream media and seeks to convince the public that only its preferred sources.
    1. 1. Solutions
  4. Reduce the cost of access to primary sources through direct channels to primary source and by crowdsourcing information
  5. Attempts to solve this from academics:
    1. “Journalist’s Resource from Harvard’s Shorenstein Center”
      1. https://journalistsresource.org/
      2. Attempts to offer non-partisan, fact-based, unemotional multi-perspectival primers with the best arguments from multiple interests on given issues
      3. Attempts to give tips and tools to journalists for how to use and interpret academics’ work
  6. Influence the consumers to demand b
Conflation of reporting and what our reaction should be to the reporting
1. Problems
  1. Stronger version of this problem - news sources know what the consequences of their report of the news will be, and so the modify the news to have the ‘correct consequences’ instead of presenting the news as it appears to have happened.
2. Solutions
  1. This is in part a result of ideology driven news organizations being in demand from tribal consumers.

Plan for attacking media

First principles
1. What would we want the media landscape to look like?
  1. What would we want from the suppliers?
  2. What would we want from the consumers?
  3. What would we want from the other stakeholders? (E.g. Government’s, commerce's relationship)
2. What would we
What do we currently feel and know (not from first principles)?
What are our 80/20 most important questions? Policies? Positions?
Continue

Outline

Actors in Media
Types of Media
Problems in Media
1. Ideology in media organizations
2. News optimizes for attention
3. Incentive Misalignment of Primary Sources
4. Inaccuracy
5. Short / Fast turnover of content and attention
6. High Quality doesn’t pass a cost / benefit test
7. Conflation of reporting and what our reaction should be to the reporting
Solutions to Problems in Media
Consumers
1. Utility Function
  1. Entertainment
  2. Information / Learning
  3. Confirmation of position / status
  4. Tribal engagement with ‘other’
2. Attention being monetized
  1. Purchase media directly
  2. Advertising
Producers
1. Medium
  1. Internet
    1. New Internet Media, ex. Buzzfeed, Vox, Youtube
    2. Article / Essay
    3. Blogs
    4. Videos
  2. Print
    1. Newspaper
      1. ex. NYTimes, WSJ, Washington Post, Gaurdian
    2. Magazines
      1. Ex. Economist
  3. Television
    1. Talk Shows
    2. 24h news networks
      1. ex. Fox, MSNBC, CNN
  4. Audio
    1. Radio
      1. Conservative Radio
    2. Podcasts
2. Reputation
  1. Branding
    1. Historical quality media, ex. NYTimes, WSJ, Washington Post, Economist, Guardian
    2. Television Media, ex. Fox, MSNBC, CNN
    3. New Internet Media, ex. Buzzfeed, Vox, Youtube
    4. Blogging, ex. Wordpress sites, Medium, personal blogs
3. Utility Function (all elements inter-related)
  1. Money
    1. Advertising
    2. Subscription
  2. Readership
  3. Respect / Reputation / Honor
  4. Truth
4. Components
  1. Content Side
    1. Journalists/Writers
    2. Editors
    3. Producers
  2. Business Side
    1. Standard business divisions (e.g. Marketing, finance, legal)
Subjects
1. Main subjects:
  1. Politics
    1. World
    2. National
    3. Law
  2. Business
    1. Companies
    2. Economics
    3. Global
    4. By Industry
  3. Local / Municipal (e.g. NYTimes or New Yorker on New Yorker)
  4. Celebrities
  5. Arts / Culture / Fashion
  6. Travel
  7. Real Estate
  8. Cars
  9. Books
2. Types of media:
  1. Historical quality media, ex. NYTimes, WSJ, Washington Post, Economist, Gaurdian
  2. Television Media, ex. Fox, MSNBC, CNN
  3. New Internet Media, ex. Buzzfeed, Vox, Youtube
  4. Blogging, ex. Wordpress sites, Medium, personal blogs
  5. Dailies
  6. Weekly magazines
    1. General
    2. Specific subject
  7. 1. 1. Utility function:
  8. 1. How does ideology permeate a story?
3. (E.g. If NYTimes is seen as left-wing, why is it seen as such?)
4. Theories:
  1. Bottom-up
    1. Individual journalists have their own biases; consciously or unconsciously, they write their own biases into the story, explicitly in op-eds or argumentative books.
      1. The journalists for the mainstream, high-quality papers
      2. Geographical argument: New York, D.C., Baltimore
        
        Famous false attribution to New Yorker film critic Pauline Kael: “
        
        Closer to real quote, also expressing the bubble of being in the New York literary elite: “On Friday, on the New Yorker’s website, the magazine’s film editor Richard Brody offers what may be the first accurate version of the quote I’ve ever seen (I’m assuming it’s accurate because it comes from the New Yorker itself): “Pauline Kael famously commented, after the 1972 Presidential election, ‘I live in a rather special world. I only know one person who voted for Nixon. Where they are I don’t know. They’re outside my ken. But sometimes when I’m in a theater I can feel them.'” https://www.commentarymagazine.com/culture-civilization/the-actual-pauline-kael-quote%E2%80%94not-as-bad-and-worse/
      3. Evidence: Most journalists give to left and center-left candidates
      4. Cultures of different papers are more attractive to a writers of a certain political ideology:
        
        E.g. The Harvard Crimson asks new inductees during Grand Elections, “What are your politics? Right or left?” Plays Darth Vader music if new members say “right” and cheer if new members say “left”. (Source: Jesse’s experience Freshman year. Jesse’s answer: Up.)
        
        In NYTimes, nothing as unprofessional as Harvard Crimson’s Grand Elections ritual but:
        
        Possible social network effects:
        
        There will be more left-wing journalists who are friends with each other.
        
        2. Once an outlet has an ideological reputation, it will attract less writing from people of different ideologies.
      5. Either:
        
        Editorial decision to publish like-minded opinions
        
        Selection effect of who is submitting them opinions (i.e. more submissions or would-be writers from left, less from right)
        
        It is not the content of the contrary opinions that passes the gatekeepers but that the people making the contrary opinions argue their values in different ways. These different ways can, e.g. appeal to religious argument, can appeal to nationalistic values, etc. that are not recognized as valid axioms or premises of argument. Therefore, the contrary opinions are not of enough “quality” in reasoning to pass the quality gate-test; the gatekeepers define acceptable values or means of reasoning/arguing that constitute “quality” writing. The NYTimes does not sympathetically publish articles for why gays should not be married based on religious grounds.
        
        It could also be that there are far fewer highly educated, good writers in terms of style on the right than left. There is a stylistic gate-test for acceptance into the mainstream media (MSM).
        
        Legitimately, what happens when the right retracts so far into its anti-intellectualism as to lose the best writers (in stylistic measures) for its case: David Brooks, David Frum, Ross Douthat, etc. all remain conservative but have been strongly critical of Trump and the current turns in the current conservative party under Trump. William Buckley, Jr., might not be for Trump today.
        
        4. 3. Cycle of readership bias rewarding journalistic bias:
      6. Journalists are valued according to the readership (attention) that they bring to the publication
      7. Readers want to read content that aligns with their ideology
      8. Journalists that write content that align with the readership’s ideology garner more attention
      9. Those journalists are rewarded and kept on, journalists who write misaligned content shift their alignment or are replaced over time
    2. Groupthink
      1. Once many journalists in the organization have a belief, it becomes true in the social environment. Ingroup-outgroup treatment of people based on their beliefs leads to conformity.
    3. Availability Heuristic
      1. Once an organization is slightly biased, the best arguments for one side become more avaliable than the best arguments for the other side
  2. Common conspiracy: Top down
    1. Because the owners of the media sources have some form of explicit or implicit editorial influence (e.g. Carlos Slim has stake in NYTimes, neocons claim NYTimes is pro-immigration because it helps Slim’s businesses)
    2. Negative evidence:
      1. The bottom-up journalist driven explanation for the ideology of a story already explains the phenomenon. We don’t need to overdetermine it with this extraordinary claim.
      2. Most journalists write the stories and don’t care know or understand much of what their bos
5. Further notes:
  1. Self
Media optimizes for attention
1. Content that has high emotional valence outperforms neutral content
2. Leads to Sensationalism
Incentive Misalignment of Primary Sources

Business models don’t work as well as they used to… went from being a monopoly to have many competitors when internet leveled the playing field

Types of media:
1. Historical quality media, ex. NYTimes, WSJ, Washington Post, Economist, Gaurdian
  1. Left-center left:
    1. NYTimes
    2. 1. Television Media, ex. Fox, MSNBC, CNN
2. New Internet Media, ex. Buzzfeed, Vox, Youtubege
3. Blogging, ex. Wordpress sites, Medium, personal blogs
4. Dailies
5. Weekly magazines
  1. General
  2. Specific subject
What do we want out of this aspect of the project?
1. have predictive opinions/models about media
  1. If we change X, Y happens
2. Understand how information flows / stories propagate
  1. Practically, how can one generate a story that gets coverage?
    1. What framing of the story gets the coverage?
    2. If you have a particular intellectual agenda, how do you get those ideas to propagate?

Potential Solutions for Problems

Ideology in media organizations
1. Problems
  1. See “How Does Ideology Permeate a Story?”
  2. Lack of multi-perspective presentation
  3. Non-profits with political motivations are biased (e.g. Koch Brothers strategy for long-term permeation of libertarian ideology started with non-profits, Heritage Foundation, etc.)
2. Solutions
  1. Create meta-media organizations that aggregates the best side of each issue
    1. Create a strong-man argument site
  2. Talk explicitly about both the issues and the way that the issues are being discussed by major sides of the conversation
  3. Present multiple perspectives
    1. Hire more journalists on right
    2. Hire more academics on right (source)
    3. Create non-partisan political non-profits
  4. General Perspective Points
    1. Always present the strong-man/steel man of the other side’s arguments
    2. Beware binaries, embrace greater complexity: e.g. Republican / Democrat; actually there are multiple voting blocs within the Republican and Democratic parties that have different demographics
3. Problems with Solutions
  1. Getting adoption of a meta-media aggregator, and technical cost in setting it up
  2. People aren’t as interested in nuanced discussion that requires thoughtfulness as they are in more emotive content
News optimizes for attention
1. Problems
  1. Media with high emotional valence outperforms other news
  2. Media as emotional valence + content rather than just true content or just an opinionated valence on the content
  3. Entertainment and news are conflated
  4. Sensational stories are overrepresented, leading to an absence of coverage of important but less emotional issues.
2. Solutions
  1. Individual guidelines for political info diet:
    1. Read long-form, in detail
      1. As an individual, convert to books and long form articles exclusively
    2. Seek unemotional reporting
      1. Select from media sources explicitly focused on unemotional / fact based reporting
    3. Take info and perspectives from at least two major sides, if not more, and sub-“sides” (and don’t assume there are only two sides)
  2. Focus on orienting to true information instead of consuming more information
3. Problems with Solutions
  1. Problems with political info dieting:
    1. Time Costs
      1. Takes approx. double time to actually watch both sides to an issue (assuming there are only two sides to the issue, which is a usually a false binary)
      2. Primary materials take too long to absorb. That is why people pay news sources they trust to summarize and present it to them.
    2. Access
      1. Institutions like media companies can develop access to truth, contacts, institutionally develop tools for other information-gathering. Similar to how an individual $10 donor cannot get a charity’s details in the way B&M Foundation or Givewell can.
Incentive Misalignment of Primary Sources
Problems
1. Ex. Syria
Solutions
1. Layer fact checking on top of the content, say crowdsourced highlighting of content with sources / intelligent debate over the facts [Wikipedia overlay for the internet]
2. Multiple angles of primary sources in the same piece (common solution)
  1. Use historical sources of similar arguments to get past current tribalism (Professor David Moss explicitly is trying to do this, starting in an HBS case-based course Jesse took on American history)
3. 1. Problems with Solutions
4. Getting adoption of a layered crowdsourced content site, getting people to contribute
5. Often multiple angles of primary sources don’t exist / hard to find
Inaccuracy
Problems
1. Sources:
  1. Primary Source misalignment
  2. Bias in journalism through data collection, confirmation bias, availability, etc.
  3. Lack of information filled with opinion of Journalist
  4. Facts are difficult to find + check
    1. Data unavailable
    2. Data requires transformation (controls, corrections) which takes time + resources
2. Are generally left uncorrected
Solutions
1. Bias inside organizations can be attenuated by having a team explicitly focused on rooting out ideological and biased claims and affect in the organization's pieces
2. Better training for journalists in how to read and interpret academics’ research
  1. Shorenstein Center’s Journalist’s Resource attempts to offer tips
3. Research data could be aggregated and made public in an accessible way through easy to use APIs and downloads (certainly all gov. Funded research)
  1. Modernization of many government entities that generate data, making that data painfully easy to access
Problems with Solutions
1. Having a team eliminate ideological claims isn’t in the benefit of the organization, whose consumers demand ideological / emotion captivating content
Short / Fast turnover of content and attention
Problems
1. Shortening of attention spans due to internet / availability of content
2. Training of consumers of media to expect short bursts of emotional content
Solutions
1. Focus on orienting to true information instead of consuming more information
  1. There’s a distinction between gathering new information and creating a model of a system that adjusts in response to information. The fast turnover of information leads to a situation where the consumer is disoriented, surrounded by information that they haven’t come to grips with and attempting to drink from a firehose of new information that makes it hard to orient to the old information. We have to tradeoff updating our model for systems against gathering new information with which to update, and this environment leads to choosing new information much more often than improving a model. And so solutions that promote creating and updating a model (here is how the system works, rather than here’s something that happened to / within the system) lead to better outcomes
2. Organize information based on issue, not current events / occurrences within the current news cycle. Updates and expands issues as new information appears. However, current news on, e.g., oil is always presented within context of greater oil context (e.g. history, arguments, stakeholders, dilemmas, current popularity of different positions).
  1. Create a resource which collects information in this way and incentivizes people to learn about an issue holistically instead of Ie. Anton’s Chair/Conscious
  2. Have a module that has all major news sources, books, academic papers, videos, documentaries, audio interviews, etc, non-profits,
  3. Have tree diagrams (of whichever type works best) that break down the problem, each of which can go into their own breakdown (like in a Workflowy document).
3. Better educate the society. Education provides at least some model and context for “elaborating” (memory literature term) the new information into a representative structure.
  1. 1. High Quality doesn’t pass a cost / benefit test
Problems
1. As the public is misinformed, low quality & inacurate content can pass unnoticed
2. Increasingly a problem as media organizations (“legacy” media) seek to maintain profits, having lost attention share to the internet
3. Stakeholders, currently the GOP and Trump administration, can use a strategy that defames the mainstream media and seeks to convince the public that only its preferred sources.
  1. 1. Solutions
4. Reduce the cost of access to primary sources through direct channels to primary source and by crowdsourcing information
5. Attempts to solve this from academics:
  1. “Journalist’s Resource from Harvard’s Shorenstein Center”
    1. https://journalistsresource.org/
    2. Attempts to offer non-partisan, fact-based, unemotional multi-perspectival primers with the best arguments from multiple interests on given issues
    3. Attempts to give tips and tools to journalists for how to use and interpret academics’ work
6. Influence the consumers to demand b
Conflation of reporting and what our reaction should be to the reporting
Problems
1. Stronger version of this problem - news sources know what the consequences of their report of the news will be, and so the modify the news to have the ‘correct consequences’ instead of presenting the news as it appears to have happened.
Solutions
1. This is in part a result of ideology driven news organizations being in demand from tribal consumers.
Privacy
1. Personal
  1. E.g. Gawker outing people
2. Public military
  1. Post 9/11 overuse of classification, often to protect US military from looking bad after doing something bad
    1. Looking bad
  2. See “Secrecy” documentary

Foundational vs. Derivative Problems

Foundational problems:

Human mental ability: e.g. limited IQ, memory, storage, ability to organize knowledge
Human Nature as survival and replication oriented - Survival: - Tribal: Human Nature as tribal, human attention span, desire for emotional valence - Replication: -Status:
-Desire for sex, mammalian brain; direction towards advertising, celebrity culture, entertainment value (largely attractiveness) versus intelligent selection decision rule e.g. TV broadcasters,

Intermediate problems:

Difficulty of accumulating, storing, remembering and using true facts.
‘’ ‘’ ‘’ ‘’ ‘’ ‘’ ability to take these facts and turn them into coherent opinions -Ability to update (away from one’s current tribe
Misalignment of primary sources
Accuracy / truth is costly and difficult to discover

Solutions that target foundational problems are much more difficult to generate. Solutions that target derivative problems are more tractable but are less effective.

Derivative problems:

Ideology in media organizations
Short / Fast turnover of content / attention
High Quality Doesn’t pass cost / benefit test
News optimizes for attention
Conflation of reporting and what reader’s reaction ‘should’ be

There’s this fear of a dystopian future which looks like fitting media more closely to human nature. And “Technology” is not improvement in general, but greater alignment with what humans want (and in general these serve short term needs). And so what we’ll get from or media is titillation, emotional stimulation, instead of factually true content that leads to well reasoned positions on important issues. And as technology improves, it’ll fit ever closer to human emotion.

What is the underlying ‘real’ value of the space? What distortions are there between ‘real’ value and the actual product provided? Why do those distortions exist? How can this gap be bridged

“Gawker, Business Insider, Politico, BuzzFeed, Huffington Post, Drudge Report”

Problems

People to Talk to:

George Ko, Visitas host who tried to start a website

-Major journalists at Harvard: -David Sanger -Nieman Fellows -(Nieman Foundation for Journalism) -Esp. Jesse might be able to use Safra Center as a connection for the Nieman Fellows in the Safra Center -Shorenstein center faculty

Sources:

General media problems / solutions
Political media coverage
1. Coverage of Trump
  1. Interesting Shorenstein Center (HKS) study on media coverage of Trump in the first 100 days
    1. Findings include:
      1. President Trump dominated media coverage in the outlets and programs analyzed, with Trump being the topic of 41 percent of all news stories—three times the amount of coverage received by previous presidents. He was also the featured speaker in nearly two-thirds of his coverage.
      2. Republican voices accounted for 80 percent of what newsmakers said about the Trump presidency, compared to only 6 percent for Democrats and 3 percent for those involved in anti-Trump protests.
      3. European reporters were more likely than American journalists to directly question Trump’s fitness for office.
      4. Trump has received unsparing coverage for most weeks of his presidency, without a single major topic where Trump’s coverage, on balance, was more positive than negative, setting a new standard for unfavorable press coverage of a president.
      5. Fox was the only news outlet in the study that came close to giving Trump positive coverage overall, however, there was variation in the tone of Fox’s coverage depending on the topic.
    2. https://shorensteincenter.org/news-coverage-donald-trumps-first-100-days/

Source: Original Google Doc