Algorithmic Reality

Article tag · 50 matching pages

Human-edited directories as an alternative discovery system | Dead Internet Theory

Dead Internet Theory: Human-edited web directories replace opaque ranking with visible editorial selection, but their coverage depends on volunteer attention, maintenance, and judgment rather than crawling everything automatically.

Webrings as reciprocal discovery between independent sites | Dead Internet Theory

Dead Internet Theory: Webrings let independent sites send visitors directly to one another, creating a human-scale discovery network that does not depend on a central ranking algorithm.

Blogrolls as personal recommendations with visible authorship | Dead Internet Theory

Dead Internet Theory: A blogroll is a visible recommendation list attached to a known person or site, making discovery subjective but legible in a way opaque recommendation systems are not.

RSS subscriptions and discovery outside a ranked home feed | Dead Internet Theory

Dead Internet Theory: RSS lets readers choose sources directly and receive updates without a ranked home feed, trading algorithmic selection for explicit subscriptions and reader-controlled ordering.

Social bookmarking as an archive of human selection | Dead Internet Theory

Dead Internet Theory: Shared bookmark collections preserve a visible record of what people chose to save, organize, and recommend, creating a different discovery layer from ranked search.

Independent search indexes and the tradeoffs of smaller coverage | Dead Internet Theory

Dead Internet Theory: A search engine with its own index can expose a different web from systems that reuse another provider's results, but independence does not guarantee equal coverage or relevance for every query.

Random-page tools and the recovery of accidental discovery | Dead Internet Theory

Dead Internet Theory: Random-page tools deliberately remove some ranking and preference signals, creating unexpected encounters with pages a user would never have thought to search for.

Measuring how many distinct domains a person encounters in a month | Dead Internet Theory

Dead Internet Theory: Counting distinct domains in a month of browsing can reveal how concentrated a person's practical web experience is, but the metric needs careful definitions and says little about depth or quality by itself.

Reproducing search experiments across accounts, locations, and time | Dead Internet Theory

Dead Internet Theory: Search experiments need recorded account, location, language, device, and time conditions because result pages can legitimately vary across those contexts even when the query is identical.

Distinguishing an undiscoverable web from a nonexistent web | Dead Internet Theory

Dead Internet Theory: A page can still exist while being difficult to locate through ordinary search, so claims that the web has emptied out need to separate true disappearance from discovery failure.