How this works
Curius Front Page ranks public saves and highlights from the local Curius crawl. The page shows two kinds of rows: saved links and repeated highlighted passages.
What the toggles do
- Links show saved URLs with title, domain, save counts, highlight counts, and age.
- Highlights show repeated quoted text, source link, reader/repeat counts, latest user, and age.
- Popular sorts by score; newest sorts links by latest save/highlight time and leaves score as context.
Popularity score
Links
Slink = 3usave + 5umark + h
usave is distinct savers, umark is distinct readers who highlighted the link, and h is total highlights.
Highlights
Squote = 4ureader + r
ureader is distinct readers behind the same quote on the same link, and r is repeat count.
How to read a row
The title line opens the source. The small line says how many distinct readers touched it, how many marks were left, and when Curius last saw save/highlight activity for it.
Glossary
points— the current popularity score for the active row type.saver— a distinct Curius user who saved a link.reader— a distinct Curius user who made a matching highlight.repeat— another copy of the same highlighted passage on the same link.
References
- Hacker News API. Lists the compact story fields this page echoes.
- SQLite aggregate functions. Gives
countandcount(distinct ...), which produce the local evidence counts SQLite.