View profile

Q&A with Cortico, the nonprofit trying to improve Twitter

This week, Twitter asked for help as it tries to promote healthier conversations on the platform. Yes
March 2 · Issue #95 · View online
The Interface
This week, Twitter asked for help as it tries to promote healthier conversations on the platform. Yesterday, we talked with Time Well Spent movement cofounder Joe Edelman about his concerns that a focus on optimizing for any metric, however well intentioned, turns people into Tamagotchis. 
I also reached out to Eugene Yi, president and co-founder of the nonprofit research group Cortico. Cortico, which uses tech tools to public conversations, has been in touch with Twitter about its effort to improve the platform, and was cited in the company’s request for proposals on how to gauge healthy conversations. 
Yi agreed to answer a few questions via email about what’s happening.
CASEY NEWTON: Tell me about what Cortico does. What are some projects you’ve worked on that speak to your main interests?
EUGENE YI: Cortico has roots in the MIT Media Lab. Several projects that point to our main interests stem from research conducted at the Laboratory for Social Machines (LSM) in the MIT Media Lab. 
LSM’s Electome project–funded by the Knight Foundation–focused on automated semantic and network analysis applied to map social media conversation and news media coverage of the 2016 U.S. presidential election. This approach to mapping the issues people are talking about and the news they’re sharing within multiple domains (e.g., politics, religion, etc.) and across various tribes (e.g., political, cultural, ideological, geographic) will help inform LSM and Cortico’s development of the public sphere health indicators.
To date, interest in our unique approach to mapping and analyzing public sphere conversation has been strong. A December VICE News piece on the Electome’s issue/supporter networks was among the most-viewed articles in the publication’s history. The New York Times featured our filter bubble work in a front-page story in March.
How did you come to work with Twitter on this?
Twitter became a member of the Media Lab in 2014, supporting LSM media analytics research “to understand the role Twitter and other platforms play in the way people communicate.” During this time, LSM has completed multiple projects and published several papers based on Twitter data.
Cortico was launched as a non-profit in 2016 to scale up professional deployment of LSM media analytics R&D, including this public sphere health indicator work.
Finally, Deb Roy and Eugene Yi both formerly worked at Twitter. Deb’s role was the Chief Media Scientist and Eugene’s role was the Head of Asia Public Policy. Both Eugene and Deb have stepped down from their roles at Twitter and now are, alongside co-founder Russell Stevens, building Cortico.
What sort of impact does the Cortico team hope to have with Twitter?
The first step to improve our public spheres is to diagnose and measure where it is currently healthy or unhealthy. We hope that our health indicators will provide Twitter and other media platforms with a useful benchmark of how their product affects public discourse both positively and negatively. Twitter’s commitment to increasing the collective health, openness, and civility of public conversation is an important step towards this end.
I understand that Twitter is looking for a set of metrics around which to optimize for better conversations. Do you think such a metric exists? And how to you guard against the unintended consequences that have come from optimizing from earlier, innocuous-seeming metrics, such as daily active users or time spent on site?
We thought this was a terrific, nuanced question that represents one of the key challenges that we will need to address. Our hope is that our health indicators will reflect aspects of healthy public discourse. While the principles of shared attention, shared reality, variety, and receptivity might be enduring, these metrics will require continuous evaluation and refinement as the public sphere evolves. We think putting forth a first version of health indicators is a positive step forward, and recognize that what we measure and why we measure will need to progress over time. 

How Facebook Misread America’s Mood on Russia
Should satire be flagged on Facebook? A Snopes debunk sparks controversy
Facebook’s algorithm change hasn’t obliterated fact-checkers’ engagement — yet
New data shows just how much social sharing has decreased since 2015 (and News Feed tweaks are just one factor)
When selling ads, Facebook avoids the topic of Russia's interference campaign
As D.C. sits on the sidelines, these states are looking to regulate Facebook, Google and Twitter
Russians Used Reddit and Tumblr to Troll the 2016 Election
Why did USA Today lend its pages to a conspiracy theorist who just got banned by YouTube?
News Corp. CEO Slams Facebook, Google for Not Sharing Enough Ad Revenue
Twitter claims more diversity in 2017 but that’s not what data shows
Snap to Release New Spectacles This Year on Cheddar
Instagram code reveals unreleased voice and video calling
How fast is Facebook slated to grow in Menlo Park?
Kim Kardashian West’s Celebrity Kid-Prank Series Ordered by Facebook
Micropodcasting? Facebook tries Voice Clip status updates
Twitter is public, but it could offer a lot more privacy
Twitter is rethinking everything. At last.
And finally ...
Take the Vine AP Test
Talk to me
Questions? Comments? Weekend plans? 
Did you enjoy this issue?
If you don't want these updates anymore, please unsubscribe here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue