Every search on The Blog Bureau returns results shaped by three layers: discovery, judgment, and synthesis. Here's what each one actually does.
We use Firecrawl to crawl the open web for blog-shaped content, posts with a byline, a publication date, and substantive prose. Aggregators, marketing pages, listicles under 400 words, and machine-generated content are filtered out at ingestion. We currently index roughly 3,200 active independent publications.
A source enters the index when it meets three tests: it has published original writing in the last 12 months, it discloses authorship, and it isn't primarily promotional. Sources that stop meeting these criteria are quietly removed. We keep a log of removals with reasons.
When you search, we retrieve the most topically relevant recent articles from the index and pass them to a language model with one instruction: describe what these writers are collectively saying, quote them accurately, and name every source. The model never invents claims, never merges quotes, and never contributes an opinion of its own. Every sentence you read in a synthesis is traceable to a specific numbered source in the sidebar.
If you're a publisher and want your site removed from the index, email remove@blogjournal.com with the domain. Removal is processed within 24 hours.
If a synthesis misrepresents a piece of writing, email corrections@blogjournal.com with the article URL and the search query. Corrections are logged publicly.