By Cliff Potts, CSO, and Editor-in-Chief of WPS News

Baybay City, Leyte, Philippines — May 21, 2026, 17:35 PHST

One of the most consequential design decisions in the modern internet was the quiet fusion of discovery and ranking into a single, opaque process. What began as a way to help users find information evolved into a system that determines which information survives.

This essay advances a single claim: we can fix the internet, and this is how—by separating discovery from ranking and restoring user- and community-level control over how information is ordered.

Discovery Is Not Judgment

Discovery answers a basic question: what information exists that might be relevant? Ranking answers a different question: what should be shown first?

Early internet systems treated these as distinct steps. Indexes cataloged what was available. Users or intermediaries decided what mattered. Over time, these functions collapsed into a single process controlled by centralized platforms.

The result is not neutral discovery, but delegated judgment.

How Fusion Creates Power

When discovery and ranking are fused, the entity that controls ranking effectively controls reality. Information that is indexed but never surfaced is functionally invisible. Information that is surfaced repeatedly acquires authority through repetition.

This fusion concentrates power because it removes alternatives. Users are not choosing among rankings; they are receiving the ranking. Institutions cannot apply their own priorities. Communities cannot emphasize local relevance.

One ranking becomes destiny.

Popularity Is a Poor Substitute for Relevance

Most large discovery systems rely heavily on popularity signals: clicks, links, engagement, and traffic volume. These signals are easy to measure at scale, but they conflate attention with value.

Popularity favors incumbents, scale, and amplification. It disadvantages local knowledge, specialized expertise, and emerging voices. Over time, it flattens the information landscape into a small number of dominant sources.

Efficiency replaces judgment.

What Separation Looks Like in Practice

Separating discovery from ranking does not require abandoning search engines or indexes. It requires modularity.

In a separated system:

  • discovery produces a broad, plural set of relevant sources
  • ranking is applied afterward, using transparent and adjustable criteria
  • multiple ranking profiles can coexist

A user might choose a local-first ranking, an academic-first ranking, a chronological ranking, or an institutional ranking. No single ordering becomes mandatory.

Choice replaces imposition.

Restoring Local and Regional Context

Once ranking is modular, locality can be restored as a primary signal rather than an afterthought. Information can be weighted outward in concentric layers: local, regional, time-zone, national, and global.

This does not exclude global information. It delays it until closer, more contextually relevant sources have been surfaced. The effect is an electronic version of a town hall rather than a global billboard.

Relevance becomes situated again.

Corrective Measures at the Discovery Layer

Practical steps toward separation include:

  • open or federated indexes usable by multiple ranking systems
  • user-selectable or institution-defined ranking profiles
  • clear separation between indexing and ordering logic
  • constraints that prevent popularity from dominating ranking

These measures do not dictate outcomes. They restore plurality.

Why This Weakens Platform Capture

Centralized platforms depend on controlling both what is found and how it is ordered. Once those functions are separated, capture becomes harder. No single actor controls the full pipeline.

Ranking becomes a service, not a mandate.

The internet becomes navigable without being governed.

A Fix, Not a Fantasy

Separating discovery from ranking does not solve every problem. It does, however, strike directly at the mechanism that turns scale into authority.

The internet does not need fewer sources.
It needs more ways to order them.

This essay will be archived in the WPS News Monthly Archive, available through Amazon.

This work may be cited freely. Licensing or implementation for commercial or institutional use requires prior arrangement.

References

Benkler, Y. (2006). The wealth of networks: How social production transforms markets and freedom. Yale University Press.

Gillespie, T. (2018). Custodians of the Internet. Yale University Press.

Rieder, B., Matamoros-Fernández, A., & Coromina, Ò. (2018). From ranking algorithms to “ranking cultures.” Convergence, 24(1), 50–68.

Sandvig, C., Hamilton, K., Karahalios, K., & Langbort, C. (2014). Auditing algorithms. Data and Discrimination Conference Proceedings.

Discover more from WPS News

Subscribe to get the latest posts sent to your email.