A problem like the comments

Almost everyone in journalism seems to agree: internet comments are broken.

But what does it mean to fix comments? Of course, we don’t want racism, misogyny, harassment and violent and unpleasant language on news websites. Is simply cleaning up the space enough? What is the journalistic purpose of engagement and community?

The big questions

The Coral Project is a collaboration between Mozilla, The New York Times and The Washington Post, funded with a grant from the Knight Foundation. We’re building open source tools and practices for publishers of all sizes to form better communities around their journalism.

In order to do this, we’ve so far talked to people at more than 100 news organizations, to understand what community means to them. Among our questions is some version of:

What is the role of the audience in your journalistic mission?

Followed by:

What else could it be?

To the first question, we’ve heard a broad spectrum of responses.

They’re the reason we do this.

They pay for our journalism.

They give us feedback when they don’t like something.

They read and share what we do.

To share their knowledge and ideas with us and with each other.

I don’t know.

The second question is more difficult. Dreaming big is hard when you lack the technical resources to manage current problems and improve existing tools — which is the situation for pretty much everyone to whom we’ve spoken. That’s why The Coral Project exists. We want to help you to dream, and then to build open source tools for you to make it happen. And maybe the tools we build don’t all look like comment boxes.

How do we decide what to build?

We’ve interviewed many people across the news industry, using design thinking principles to understand the needs of different people in news communities.

Screengrab of our Trello board of needs: https://trello.com/b/Dhrb4D74/coral-s-needs

You can see the main user needs we’ve collected here, arranged by role. They aren’t in any particular order, and we won’t tackle all of them in the 21 months we have left.

Did we miss something? Tell us in the annotations space on this page, so we can add it to our board.

What are we going to build?

Based on the user needs gathered on that board, we created thematic collections of products we want to build. The collections of products currently look like this:

Identity — includes login, user profiles, privacy controls and more
Engage — includes tools for journalists including specific tools for user-generated content collection and management
Conversation — includes tools for discussion and exchange between community members
Curate — includes tools for engagement editors to manage, moderate and elevate contributions of all kinds
Admin — includes tools to install and manage Coral-compatible products
Analyze — includes algorithmic analysis including anomaly detection and comment quality
History / Logs — includes capture logs and analytics
Core — common libraries that allow Coral products to work seamlessly together or independently

Each publisher can choose which products they want to install on their version of Coral. We haven’t yet decided what exactly each of the products will be — that’s coming next.

Next steps

We’re bringing our team together in New York (some work remotely) to work on next steps, including a product roadmap, a relaunched website and a deliberate strategy for working in a more open and collaborative way from here.

We also need to answer the questions we’re asking of others: what is the role of the community in our mission?

If you have any thoughts or ideas about what we’re doing, please email me at andrewl@mozillafoundation.org

Coral is coming. Let’s dream together.

(This was first published in our Medium publication)

Click here to discuss this piece within The Coral Project Community.

andrew Written by: