Reimagine search.

PeARS: Peer-to-peer Agent for Reciprocated Search

PeARS (Peer-to-peer Agent for Reciprocated Search) is a search engine that you can run locally from your browser. It allows you to ‘index’ pages (i.e. to produce a computer-readable representation of the pages’ content, essential to the search process), and to search pages that you or your friends have indexed. Search happens entirely on your machine, meaning that no one knows what you are searching and when.

Currently, the main feature of PeARS is the ability to convert a small index into a greyish, unassuming picture, called a ‘snow pod’. Snow pods are the mini-weapons of the indexing revolution. They can easily be shared with others by email, on social media, or through any other mean, so you can be your very own search engine, for yourself and for your friends.

PeARS, phase 1

There are no good pears without a good winter. And there is no search engine without a good index. The PeARS project is currently in phase 1, recruiting volunteers to build the first community-based index of the Web, and to test the basic search capabilities of the system. This PeARS version is codenamed ‘orchard’ (because it is planting time!) and is available for cloning / downloading on GitHub. It allows anybody to create searchable representations of their favourite websites and share them with others, to allow them to search those sites.

Example scenario: Alice is writing a novel based in the Czech Republic, and wants to find out more about Czech cuisine. She has no idea what the best Internet resources are and so far, has relied on the top results of traditional search engines, which heavily prioritise large recipe repositories, and tourist information. In another place and another time, Pavel, who is a real foodie, has been spending an appreciable amount of time browsing interesting culinary sites. He has decided to make that browsing knowledge available by creating a PeARS pod on Eastern-European cuisine, and posting it on Twitter. Alice now comes across Pavel’s pod and downloads it to her local PeARS installation. She can now search all of Pavel’s favourite websites, gaining access to real expert information.

PeARS, phase 2

The fully-fledged PeARS system is supposed to be completely distributed. You can imagine it as an ‘automated’ version of Phase 1, where you don’t have to go and hunt for pods yourself. Your PeARS install will automatically find them on other users’ systems and connect to them. The original blueprint for PeARS was published at the WWW conference in 2016.

Find out more about PeARS at http://pearsearch.org/.