A note concerning Primary Source Knowledge

TL;DR

This paper explores how automated filters on arXiv can enhance primary source knowledge by allowing non-experts to access scientific literature directly, with minimal cultural immersion in specialized fields.

Contribution

It introduces a novel perspective on primary source knowledge through the lens of automated filtering and moderation processes on scientific preprint servers.

Findings

Automated filters can identify relevant scientific papers for non-experts.

Human moderation on arXiv is based on interest and relevance rather than technical review.

This approach broadens access to primary scientific literature for the general public.

Abstract

We add a small increment to understanding the notion of Primary Source Knowledge, knowledge that the non-expert and the citizen can acquire by assiduously reading the primary scientific journal literature without being embedded in the cultural life of the corresponding technical specialty. This comes from exposing four papers to the automated computer filters used by the physics preprint server arXiv. These filters are used to flag papers in need of further review by human assessors before being promulgated on the server; papers not flagged by the algorithm are generally posted on arXiv without further review. After the filtering, human moderators decide whether papers should be posted based on a relatively low bar of whether they are of interest, relevance and value to the research communities that populate arXiv.