I went and got the real data, and it split into two different true things — kept honestly separate, not averaged together. Signal A: each domain's real subcategory count on Wikipedia (Society 52 down to Philosophy 14) — the same seed as before. Signal B, new: the documented "Getting to Philosophy" phenomenon — click each article's first link, repeat, and 94.5%+ of all English Wikipedia articles eventually converge on Philosophy, median chain 23 hops. That is not a metaphor — it is a measured property of the entire link graph. So Philosophy is rendered here as what it actually is: small by subcategory count, but the sink the whole field slowly drains toward. The other seven keep their real weight and a random, reproducible scatter — I could not honestly fabricate true distances between them, so I didn't.
Bridge-Burners LLC · Fiddler · signal A: subcategory count · signal B: Getting-to-Philosophy convergence (94.5%, median 23 hops) · no fabricated distances · self-testing
seed — · click a cloud to inspect
the eight
the sink, documented
94.5%+ of English Wikipedia articles converge to Philosophy by following each article's first link, repeated. Median chain: 23 hops. Mechanism: lead sentences define a topic via its broader category, climbing a classification ladder that empties here.
field spec — runs live
—
Status discipline
LiteralBoth signals are real and citable: subcategory counts from Category:Main_topic_classifications; the 94.5%/23-hop convergence from documented research on Wikipedia:Getting_to_Philosophy.
BridgeRendering convergence as a "gravity sink" and subcategory count as "weight" are visual metaphors for two different real graph properties — kept distinct, not merged into one number.
SpeculativeTrue pairwise distances between the other 7 domains were NOT available without fabrication, so their positions remain a random, reproducible seed — flagged, not pretended otherwise.