This is a preliminary effort to further explore the concept of modular journalism, picking up from the work started in 2021 at the JournalismAI collaborative by a group of newsrooms and individuals that included Il Sole 24 Ore, Shirish Kulkarni, the teams at Deutsche Welle and Maharat Foundation, and the invaluable contributions of Gary Rogers and Robert Dale. I was at the time with Il Sole 24 Ore, participating as head of the Content Innovation Lab. I am now with The Trust Project.

By modular journalism, we refer to a form of storytelling that transcends the traditional long-form article by breaking content down into self-contained modules. These modules are designed to respond directly to specific user information needs and can be arranged in a variety of ways, regardless of established journalistic norms, to promote engagement.

What's new in 2.0? The modules API has been refactored and updated. User needs have been renamed information needs for more clarity, and a new user effects entity has been added alongside them.

User effects, which I have only started to compile, help define the 'bad habits' of modularity. They highlight portions of the text that not only provide no value to users but also appear deliberately crafted to misinform and manipulate or do so unintentionally due to a lack of journalistic standards and malpractice. User effects should not typically be found in ethical journalism, but some of them are quite frequently used by mainstream sources in many countries, particularly when it comes to sourcing transparency, mixing opinion with facts, neglecting context and verification, and relying on false narratives.

User effects sit outside of information-needs-based modularity and can serve as guardrails, for both humans and machines. For information needs and user effects, I will attempt to define prompting.

Finally, I am testing a new simplified modular-first approach within the content management system to improve the journalist user experience of writing in chunks.

• The original modular journalism project started with the realization that we were reaching only a small portion of our audience, and we wanted to change that. Often, what we consider priorities as journalists are not priorities for our users, especially when we look at long-form articles that are seldom consumed entirely. So we asked ourselves if we could do something differently in terms of storytelling to foster positive engagement and reach perhaps underserved and neglected audiences.
• An approach to modular journalism that is fully based on information needs leads to an entirely new experimental form of news artifacts, with little in common with traditional long-form articles. It is so different, in fact, that the easiest way to leverage modular journalism is to use a modular-first approach, where journalists do not build their stories starting from a blank page but instead use novel tools that guide them to answer specific questions hypothetically posed by users. Questions like, "What is the big picture?" or "What is the impact on my community?" or "How can someone affect what happens next?"
• This approach has a very immediate consequence: any portion of a journalistic text that does not respond to a question posed by users is discarded.
• What this means, in practical terms, is that if we were to take a journalistic report and pass it through the filter of information-needs-based modular journalism, we would end up with only crumbs of the original story — most of the original text would be tossed. The reason is that traditional long-form articles rarely respond to questions frequently posed by users, which in itself is not at all a bad thing, but it is one of the reasons why most users don't read them.
• We are left with a series of inevitable questions. What should we do with the parts that we have discarded? Do they really not have a purpose or a function for users? And if they don't respond to user information needs, how can they be possibly marketed to users?
• In these notes on modular journalism, I attempt to introduce the new concept of 'user effects' as a placeholder while we continue investigating the discard pile. User effects are, in a way, the opposite of information needs. If a information-need module is an answer to a question posed by users, a user effect can be expressed as an intention that the author has to engage with the user. My working list of user effects is somewhat nefarious so as to illustrate the process more clearly. Some examples are "I will present the events in a way that highlights our faction's successes" or "I will subtly incorporate my opinions into my reporting through careful word choice and by how I present facts and events." Or "I will knowingly omit, distort, or misrepresent details in my reporting to support a narrative."
• This exercise in modular journalism will examine existing news artifacts and attempt to analyze them, assigning user effects if present, and rewrite them relying on information needs and using the principles of information-needs-based modular journalism. The first example is an article about a scientist who denies climate change, published by an Italian newspaper in January 2023.

The low-hanging fruit is giving newsrooms new ways to foster positive engagement and reach underserved audiences by presenting users with news artifacts that more closely reflect their needs. This is done through a data-driven process that first determines what the information needs are (the original modular journalism project did so through user research conducted by participating publishers) and then imagines how modularity can help build flexible stories where portions are added or rearranged based on reliable templates for specific audiences or even personalized for specific users, each with their particular preferences.

The bulk of information needs, as emerged from user research, relate to context, background, and factual analysis. The result is that modular stories are often upside down compared to tradition, giving prominence to aspects that would be relegated to the bottom or entirely omitted.

Modularity breaks the mold of the article in other ways too. Not every portion needs to be updated at the same time, or by the same author. Context, background, and factual analysis can be invoked on multiple pages and be reusable across multiple stories. Articles cease to be unique and unequivocal documents dated in time and become algorithmic.

In a structured approach that uses modules for storytelling, automation and AI can play a role. From detecting information needs to flagging content that does not reflect information needs effectively to (with immense caution and a lot of caveats) generation and story building.