Written and maintained by CASRAI Editorial Board
Last updated
Short answer: the Lumen database is a research repository of legal complaints and requests to remove online material, such as copyright (DMCA) takedown notices, run as a project of Harvard’s Berkman Klein Center for Internet & Society. This guide explains who runs it, where its notices come from, how the public and researchers reach it, and what limits to keep in mind when you use it as research data.
What is the Lumen database?
Lumen’s own code repository describes it as a project that “collects and analyzes legal complaints and requests for removal of online materials, helping Internet users to know their rights and understand the law.” Its documentation adds that it helps people see where removal requests come from and lets researchers study how common legal threats are. The service is hosted at lumendatabase.org.
Harvard’s Berkman Klein Center describes the Lumen Database as “a repository of content takedown requests sent to OSPs” (online service providers), and calls Lumen its longest-running project. It is not a law firm, a legal-advice service or a complaint-resolution service: it archives and analyzes notices, it does not adjudicate them.
Who owns and funds the Lumen database?
Lumen is a Berkman Klein Center for Internet & Society project at Harvard University. The source code carries a “Copyright (c) 2016 President and Fellows of Harvard College” notice and is released under the GNU General Public License version 2 (GPLv2). In a 2022 Berkman Klein Center post, the project’s funder is named as Arcadia, the charitable fund of Lisbet Rausing and Peter Baldwin, and Adam Holland is named as project manager.
Who contributes notices?
Notices arrive mainly from online service providers that choose to share the takedown requests they receive. Partners named by the Berkman Klein Center include Google, YouTube, Twitter, Wikipedia, Medium and Automattic. The Center states that sharing is voluntary and subject to each provider’s internal guidelines and shifting priorities. The practical consequence is that the database is a large, useful collection of notices but not a complete census of every takedown request sent anywhere.
For high-volume senders and recipients of notices, the project offers an API so notices can be submitted without using the web interface.
How can you access it?
There are three routes, according to the project’s documentation:
- The website. The public site at lumendatabase.org is where notices are published and where the sign-in page lives. The project documentation notes that the public can try the service but that full access needs a token.
- The API. Researchers can search programmatically. To receive a token that provides full access, you request an API key from the Lumen team ([email protected], as listed in the documentation). The documentation has an “API documentation” page and a “Troubleshooting” page.
- The code and documentation. The software is open on GitHub (berkmancenter/lumendatabase) and the maintained documentation is on GitBook (berkman-klein-center.gitbook.io/lumen-database). The GitHub wiki home page itself warns that it may be outdated and points to the GitBook.
How do you see DMCA complaints on Lumen?
DMCA takedown notices that service providers have shared with Lumen are the kind of material the database collects, and the website is the place to look for them. Public access can be more limited than the access an API token gives, so if you need systematic or high-volume searching for research, request a token from the Lumen team rather than relying on the public interface.
Is the Lumen database down?
This guide cannot report live status. If the site does not load, check lumendatabase.org directly and, for persistent problems, contact the Lumen team at the address above; the documentation’s Troubleshooting page covers API issues.
Strengths and limits as a research source
- Strength: a single, named, institutionally hosted archive. The Berkman Klein Center states it serves researchers, scholars, academics and journalists studying patterns in content takedown requests.
- Strength: open code. The GPLv2 repository lets you see how the service is built.
- Limit: voluntary, partial coverage. Because contributors choose what to share, absence of a notice in Lumen is not evidence that no notice was sent. State this when you cite findings.
- Limit: access tiers. Full access depends on a token granted by the team, which matters for reproducibility; document how and when you obtained your data.
- Limit: you are reading allegations. A notice records what a sender claimed, not a court’s finding. Treat counts and categories as claims data and see research transparency for reporting practice.
Alternatives and complements
| Source | What it is (per its own pages) | Relationship to Lumen |
|---|---|---|
| Lumen database | Repository of takedown and removal requests shared by online service providers | Cross-provider archive of individual notices |
| Wikimedia Foundation Transparency Report | Report with sections on content alteration and takedown requests and on DMCA takedown notices for the Wikimedia projects (checked 2026-11-07) | One provider’s own reporting; Wikipedia is also a named Lumen partner |
For finding scholarship that analyzes takedown data, a bibliographic index such as Web of Science or the Dimensions research database can locate papers that cite Lumen.
FAQ
What is the Lumen website?
lumendatabase.org is the public site of the Lumen Database, the Berkman Klein Center project that collects and analyzes legal complaints and removal requests for online material.
Who owns the Lumen database?
It is a project of Harvard’s Berkman Klein Center for Internet & Society; Harvard holds the copyright on the software, which is licensed under GPLv2.
Is Lumen a complete record of DMCA notices?
No. Provider sharing is voluntary, so it is a partial record.
Sources
- Lumen Database repository, GitHub (berkmancenter/lumendatabase), read 2026-11-07.
- Lumen Database documentation, Berkman Klein Center GitBook, read 2026-11-07.
- Berkman Klein Center, “Toward best practices around online content removal requests” (2022-07), cyber.harvard.edu, read 2026-11-07.
- Wikimedia Foundation, Transparency Report, wikimediafoundation.org/about/transparency/current/, read 2026-11-07.
