Final Thesis: Crowdsourcing License and Copyright Scanner Corrections
Abstract: In today’s software development world, open-source software is a big player, but it presents a number of challenges in managing and complying with a sea of licenses. Tools like ScanCode are used everywhere for detecting the licenses within the code, but they regularly fail at newer or modified licenses since the open-source landscape keeps on changing. The result is often missed or wrong licenses, putting
organizations at legal risk. This thesis also proposes a new approach that will make identification of licenses more accurate by adding a crowdsourcing feature to SCA Tool. Users can suggest licenses that might have been missed by the scanner, endorse the finding of an existing license, and finally provide comments to support decisions made on both suggestion and endorsement activities. Upon reaching the number of endorsements, the license is flagged as concluded but takes only full approval upon review and confirmation of the admin himself in the database. By doing so, the methodology pools collective knowledge of the community in a quest towards keeping pace with a fast-changing world of OSS licenses. Human judgment along with automated scanning achieves a more accurate process and also makes sure that organizations can confidently stay compliant.
Keywords: crowdsourcing, open source
PDF: Master Thesis
Reference: Eldrin Sanctis. Crowdsourcing License and Copyright Scanner Corrections. Master Thesis. Friedrich-Alexander-Universität Erlangen-Nürnberg: 2024.