πFS
Points and comments are a snapshot, not live.
πFS stores file data as digit locations in pi rather than on disk, achieving theoretical infinite compression.
πFS is a filesystem joke based on the mathematical conjecture that pi is a normal number, meaning all possible finite digit sequences appear in its decimal expansion. The system stores actual file metadata (names, byte locations) on disk but claims to retrieve file content by looking up individual bytes as sequences within pi using the Bailey-Borwein-Plouffe formula. The repository notes it is a prototype with severe performance limitations and redirects to a newer project called InferenceFS.
The humor rests on encoding file locations in pi as metadata while storing the locations themselves in pi, creating infinite recursion. The author frames this as solving storage scarcity through "data-free" filesystems, proposing future improvements like parallelized pi lookup and cloud-based pi searching.
What commenters are saying
Commenters recognized this as a deliberately absurd but mathematically grounded joke. The thread focused on philosophical implications: if pi contains all possible sequences, it also contains all falsehoods indistinguishably from truths, making the data useless without external information. Multiple users noted the metaproblem that storing file locations requires storage equal to or larger than the original data, rendering compression impossible. Discussion also surfaced comparisons to Tom7's Harder Drive video and the Sloot Digital Coding System, a real scheme using block deduplication. One commenter verified that pi's longest known consecutive digit sequence is 13 eights, raising questions about realistic data discovery times.