Many good ideas today are born in tweets, threads, blog posts, and personal essays: hypotheses, methods, conceptual models, critiques, partial results. But they live in feeds—personality-driven, ephemeral, hard to search by topic, and almost impossible to cite. If an idea doesn’t become a full paper, it usually just disappears.

The proposal is a non-profit, community knowledge infrastructure that publishes idea-posts the way journals publish papers:

One archived entry = one original post (blog post, thread, essay, etc.)

Only the original author can submit

The post becomes a lightly edited canonical version, approved by the author

Once published, it is frozen forever

Each entry gets a unique, persistent ID (DOI-like) and stable URL

Entries are organized into subject-wise “Idea Journals” across all fields

The result is a permanent, citable record of ideas that would otherwise be lost in feeds. Follow-ups or corrections are new entries, so the system preserves a chronology of ideas rather than endlessly edited pages.

This isn’t a social network, forum, wiki, or preprint server. It’s a publisher + library for ideas: a subject-indexed archive and memory layer for the internet’s best thinking. Discovery is subject-first, author-second.

There’s no heavy peer review—just light editorial curation, with AI assisting filtering and classification. Think library standards, not journal gatekeeping.

I’m not building this myself—sharing it because it feels like missing infrastructure. Does this solve a real problem for you? Does something like this already exist in a form I’ve missed?