Doctoral Dissertations

Date of Award

8-2020

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Computer Science

Major Professor

Audris Mockus

Committee Members

Hairong Qi, Bruce J. MacLennan, Russell Zaretzki

Abstract

Motivation: Free Libre Open Source Software (FLOSS) has become a critical componentin numerous devices and applications. Despite its importance, it is not clear why FLOSS ecosystem works so well or if it may cease to function. Majority of existing research is focusedon studying a specific software project or a portion of an ecosystem, but FLOSS has not been investigated in its entirety. Such view is necessary because of the deep and complex technical and social dependencies that go beyond the core of an individual ecosystem and tight inter-dependencies among ecosystems within FLOSS.Aim: We, therefore, aim to discover underlying relations within and across FLOSS projects and developers in open source community, mitigate potential risks induced by the lack of such knowledge and enable systematic analysis over entire open source community through the lens of supply chain (SC).Method: We utilize concepts from an area of supply chains to model risks of FLOSS ecosystem. FLOSS, due to the distributed decision making of software developers, technical dependencies, and copying of the code, has similarities to traditional supply chain. Unlike in traditional supply chain, where data is proprietary and distributed among players, we aim to measure open-source software supply chain (OSSC) by operationalizing supply chain concept in software domain using traces reconstructed from version control data.Results: We create a very large and frequently updated collection of version control data in the entire FLOSS ecosystems named World of Code (WoC), that can completely cross-reference authors, projects, commits, blobs, dependencies, and history of the FLOSS ecosystems, and provide capabilities to efficiently correct, augment, query, and analyze that data. Various researches and applications (e.g., software technology adoption investigation) have been successfully implemented by leveraging the combination of WoC and OSSC.Implications: With a SC perspective in FLOSS development and the increased visibility and transparency in OSSC, our work provides potential opportunities for researchers to conduct wider and deeper studies on OSS over entire FLOSS community, for developers to build more robust software and for students to learn technologies more efficiently and improve programming skills.

Comments

Portions of this document were previously published in conference MSR 2019 and journal TSE 2020, and was mentioned in Appendices

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS