Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Search Plurality

Shiran Dudy

TL;DR

The paper addresses the problem that traditional search engines enforce a hierarchical ranking that can reflect non-epistemic biases. It proposes a category-first approach called Search Plurality, featuring three elements: categories, explicit relevance, and plural design, to reveal the breadth of a topic rather than a single ranked item. Through case studies, it demonstrates how this design can enhance comprehension of the information landscape and support exploratory learning while mitigating bias. It further argues for participatory governance in determining relevance to promote inclusive, democratic access to information.

Abstract

In light of Phillips' contention regarding the impracticality of Search Neutrality, asserting that non-epistemic factors presently dictate result prioritization, our objective in this study is to confront this constraint by questioning prevailing design practices in search engines. We posit that the concept of prioritization warrants scrutiny, along with the consistent hierarchical ordering that underlies this lack of neutrality. We introduce the term Search Plurality to encapsulate the idea of emphasizing the various means a query can be approached. This is demonstrated in a design that prioritizes the display of categories over specific search items, helping users grasp the breadth of their search. Whether a query allows for multiple interpretations or invites diverse opinions, the presentation of categories highlights the significance of organizing data based on relevance, importance, and relative significance, akin to traditional methods. However, unlike previous approaches, this method enriches our comprehension of the overall information landscape, countering the potential bias introduced by ranked lists.

Search Plurality

TL;DR

The paper addresses the problem that traditional search engines enforce a hierarchical ranking that can reflect non-epistemic biases. It proposes a category-first approach called Search Plurality, featuring three elements: categories, explicit relevance, and plural design, to reveal the breadth of a topic rather than a single ranked item. Through case studies, it demonstrates how this design can enhance comprehension of the information landscape and support exploratory learning while mitigating bias. It further argues for participatory governance in determining relevance to promote inclusive, democratic access to information.

Abstract

In light of Phillips' contention regarding the impracticality of Search Neutrality, asserting that non-epistemic factors presently dictate result prioritization, our objective in this study is to confront this constraint by questioning prevailing design practices in search engines. We posit that the concept of prioritization warrants scrutiny, along with the consistent hierarchical ordering that underlies this lack of neutrality. We introduce the term Search Plurality to encapsulate the idea of emphasizing the various means a query can be approached. This is demonstrated in a design that prioritizes the display of categories over specific search items, helping users grasp the breadth of their search. Whether a query allows for multiple interpretations or invites diverse opinions, the presentation of categories highlights the significance of organizing data based on relevance, importance, and relative significance, akin to traditional methods. However, unlike previous approaches, this method enriches our comprehension of the overall information landscape, countering the potential bias introduced by ranked lists.
Paper Structure (5 sections, 1 figure)

This paper contains 5 sections, 1 figure.

Table of Contents

  1. Introduction
  2. Search Plurality
  3. Design elements
  4. Case studies
  5. Who gets to decide

Figures (1)

  • Figure 1: Search Plurality is illustrated through three scenarios where the size of the blurb reflects its relative importance to the query (i.e. number of items within a category), and the categories demonstrate the breadth of the query. (a) Presenting a diverse array of perspectives to tackle a query like "how do I avoid being evacuated," encompassing categories pertinent to law, healthcare, media and news, as well as law enforcement. (b) Offering a variety of options to clarify the term and its meaning in a query such as "what is banking it." (c) Providing a spectrum of opinions surrounding a query like "what should I pursue a degree in?"