A study of the evolution and transformation of liquid software over time : AI-boosted systematic literature review
Eeronheimo, Otto (2023)
2023
Tietotekniikan DI-ohjelma - Master's Programme in Information Technology
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-12-13
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-2023112810324
https://urn.fi/URN:NBN:fi:tuni-2023112810324
Tiivistelmä
In modern era, humans are constantly interacting with multiple devices on their everyday lives. From the adoption of early personal, mass marketed computers in 1970’s, the markets are now filled with different computational devices, such as different mobile devices, modern computers and different embedded systems and devices. This has led towards multiple device ownership, where devices operate on different operating systems, which support different spectrums of software. Multiple device ownership has introduced some problems: devices are usually incompatible with each other affecting the seamless usage of different devices, installation and upgrading applications to various devices can be time consuming and impractical, and software as well as user settings must be configured on each platform individually. A paradigm called liquid software addresses these issues by defining a seamless usage of multiple devices.
Technical requirements for seamless multi-device usage were addressed in 2014 as paper “Liquid Software Manifesto: The Era of Multiple Device Ownership and Its Implications for Software Architecture” was published. This thesis investigates the development of liquid software paradigm described in liquid software manifesto. The goals are to identify the key trends and directions in academic literature since liquid software manifesto was published.
The approach for studying liquid software over time is AI-Boosted systematic literature review, which acts as a case study on how traditional systematic literature review methods can be enhanced with generative AI (Artificial Intelligence). Case study contains two phases where the use of AI is applied in systematic literature review: source selection and article classification. Source selection included a method of AI-Boosted snowballing, which was used to automatically identify academic papers discussing the development of liquid software. Article classification utilized AI generated text embeddings and machine learning to provide a data-driven approach to identify relationships and structures in selected articles. The findings from this AI-Boosted systematic literature review were three categories describing different aspects of liquid software. These categories contributed to understanding the evolution of liquid software from theoretical concepts to practical applications.
Technical requirements for seamless multi-device usage were addressed in 2014 as paper “Liquid Software Manifesto: The Era of Multiple Device Ownership and Its Implications for Software Architecture” was published. This thesis investigates the development of liquid software paradigm described in liquid software manifesto. The goals are to identify the key trends and directions in academic literature since liquid software manifesto was published.
The approach for studying liquid software over time is AI-Boosted systematic literature review, which acts as a case study on how traditional systematic literature review methods can be enhanced with generative AI (Artificial Intelligence). Case study contains two phases where the use of AI is applied in systematic literature review: source selection and article classification. Source selection included a method of AI-Boosted snowballing, which was used to automatically identify academic papers discussing the development of liquid software. Article classification utilized AI generated text embeddings and machine learning to provide a data-driven approach to identify relationships and structures in selected articles. The findings from this AI-Boosted systematic literature review were three categories describing different aspects of liquid software. These categories contributed to understanding the evolution of liquid software from theoretical concepts to practical applications.