How OSC Works
This page provides a high-level overview of the current process. The pilot cohort will receive more detailed guidance, templates, and documentation. Participants will also help shape the norms and practices that ultimately define OSC.
Overview:
Choose a Question
Inquiry begins with a question, not a topic. Questions may emerge from personal experience, academic interests, current events, or simple curiosity. The goal is not to begin with an answer, but with a puzzle you genuinely want to understand. A good question is not necessarily groundbreaking. For example:
Why do these things over here look different from those things over there?
Does this widely cited study hold up to different data or methods?
Is a popular claim circulating online actually true?
Why did a historical event unfold the way it did?
What explains a pattern in a public dataset?
Questions need not be revolutionary to be meaningful. Often the best investigations begin with a simple question that someone is willing to pursue carefully.
Investigate
Let your chosen question guide your approach to investigating it. Some inquiries involve replicating a published protocol, reviewing the literature, running experiments, analyzing data, interviewing subjects, digging through archives, or visiting field sites. Many draw on multiple approaches.
Participants are encouraged to use the methods and frameworks that best help them investigate their question, not necessarily those that are easiest, most familiar, or most comfortable. Some of the most rewarding inquiries involve learning new skills en route to an answer. OSC believes that methods are tools, not identities. Those tools may or may not include AI.
Inquiry often involves uncertainty, discomfort, surprises, and false starts. Some promising ideas fail, and evidence sometimes points in unexpected directions. These moments are not obstacles to learning but part of the process.
Document the Process
Throughout their inquiry, participants document their steps toward answering their initial question. Documentation can take traditional and nontraditional forms; OSC encourages participants to innovate creative modes of tracking their steps. These may include lab notebooks, research journals, field notes, process logs, annotated bibliographies, version histories, checklists, records of AI interactions, or formats yet to be invented. Some participants may even find that documenting an investigation becomes much more than dry paperwork, transforming into a fascinating investigation of its own.
OSC places special emphasis on process documentation because it is often where the deepest learning, scientific rigor, and intellectual humility take root. OSC is interested in unhiding—and celebrating—aspects of inquiries that might not look like typical science fair projects or get published in traditional journals, but that nevertheless offer more value than superficial success. Dead ends can help us update our assumptions about how something works, and surprising outcomes may yield new hypotheses that future inquiries go on to test.
The goal is not just to report what you discovered, but also to chart how your understanding evolved along the way. Students may learn more by learning in public, and help others learn as well.
Share and Revise
While biographies of famous scientists and inventors often depict inquiry as a solitary activity, many great ideas emerge through collaboration, conversation, and good-faith critique. Other people may notice assumptions we overlook, raise questions we forgot to ask, and propose explanations we never considered.
Participants share their work with the cohort throughout their inquiries, not just at the end. They may hash out problems, compare approaches, exchange written feedback, push back on each other's assumptions, and inspire one another through conversations only loosely connected to the work at hand. Depending on the inquiry, they may also receive feedback from mentors, educators, researchers, or other members of the broader OSC community.
Mutual learning takes priority over competition. That said, OSC also makes room for friendly competition that can complement intrinsic motivation without supplanting it.
In many educational settings, feedback serves primarily to evaluate performance. OSC treats feedback differently, as a chance to strengthen one's reasoning. Participants can gain confidence by sharing unfinished ideas in a supportive cohort of others who are also revising their assumptions, questions, methods, and conclusions.
Publish What You Learned
In OSC, publication means making your inquiry available for others to learn from. At the end of the cohort experience, participants share a snapshot of what they learned and how they learned it, recognizing that even the best investigations often leave important questions unanswered.
Depending on the inquiry, this might take the form of a research or replication report, process note, website, data visualization, video, public presentation, or something else entirely. The specific format matters less than how it helps others understand and build upon the work.
OSC aims to compile an archive of projects and supporting documentation so that future participants and the general public can learn from and apply these artifacts of inquiry. Publication is not the end goal, but rather an invitation for curious others to expand on the original question.
What Might an OSC Project Look Like?
There is no single "correct" OSC project. Participants pursue questions that genuinely interest them using methods appropriate to those questions. The examples below are meant to illustrate the wide variety of projects participants might undertake—not to serve as templates. Whatever your project, we encourage you to document your process transparently and follow the evidence wherever it leads.
Example 1: Exploring the Natural World
Question
Do native wildflowers attract more pollinators than ornamental flowers in my neighborhood?
Possible Activities
Read background literature.
Write down your initial predictions.
Develop a standardized observation protocol.
Share weekly updates with the cohort.
Discover that your measurement protocol is producing unreliable data, document the problem, and revise the protocol transparently.
Track your results, whether they match your predictions or not.
Reflect on your journey: what you learned, what went wrong, remaining questions, etc.
Possible Outputs
Fieldwork log
Public dataset
Reflection essay
Blog post
Example 2: Understanding Human Attitudes and Behavior
Question
Do students overestimate how many of their classmates use AI for homework?
Possible Activities
Review existing research.
Record your hypotheses.
Design a survey.
Pilot the survey and reword confusing questions before collecting additional responses.
Share your questionnaire and anonymized data with the cohort.
Discuss your findings, any surprises, lingering uncertainty, and alternative explanations.
Possible Outputs
Research plan
Survey instrument
Data visualizations
Summary of your project in a student magazine
Recommendations for future studies
Example 3: Interpreting the Past
Question
How did local newspapers describe immigration over the last twenty years?
Possible Activities
Read historical scholarship about immigration and journalism.
Develop transparent coding criteria.
Chart changes in interpretation over time.
Discuss with your peers where patterns challenged your prior assumptions.
Share your notes and sources publicly.
Possible Outputs
Annotated article archive
Coding framework
Essay on your methods and findings
Public presentation
Op-ed in your local newspaper
Example 4: Designing and Testing Solutions
Question
Can I build a website that helps students distinguish reliable from unreliable scientific claims?
Possible Activities
Interview classmates for background information.
Explore existing websites.
Build a prototype.
Test user experiences.
Notice aspects of the interface that confused users and simplify based on feedback.
Publish your code and documentation.
Possible Outputs
Interview summaries
Preliminary website
Archive of version histories
Code
User manuals
These examples differ in questions and methods, but they share the same commitment to curiosity, transparency, and learning from evidence. They are hypothetical and intended only to illustrate the range of projects participants might pursue. As OSC grows, we hope to replace them with real examples from student work.
Questions? hello@opensciencechallenge.org