FST Graduate Research Topic

Research Topic: Improving email management
Suggested By: Jon Dron
Last Update: May 16th, 2025

Overview Most of us are familiar with the problem of managing large quantities of email - keeping track of and responding to important messages, managing threads, organising messages, dealing with flags, tags, filters and so on. Some systems, such as MS Outlook, provide additional tools that allow (for example) the scheduling of messages, pinning them for follow-up, and integration with calendars, though this mostly relies on proprietary extensions or protocols that are not portable between different email system. For all but the supremely well-organised this is, at least occasionally, a big problem. This project is to determine and explore approaches to the management of personal email accounts, including the development of tools to assist the process. The project will have two main phases, an investigative phase and a design & construction phase. Investigative phase This will involve a combination of literature review and some primary research to establish effective methods, technologies and other factors leading to effective management of email. Primary research might include interviews and (possibly) questionnaires, observations and usability studies. Note that conventional usability evaluation methods may require some modification to deal with the fact that email management takes place over a long period. The output of this phase will be a set of patterns/models of behaviour and, if possible, a qualitative and/or quantitative evaluation of the effectiveness of different approaches. Design & construction phase Based on the identified patterns/models, to develop a standards-based open source computer program and documentation (probably as a plugin for Thunderbird, or Apple Mail, or as a standalone program to directly manipulate standard mbox files) which embodies and automates good practice, and/or that provides support for or enhances different email management approaches. Of course, the system should be properly tested, evaluated and documented. Note that is is unlikely and nor is it expected that the program will cater for every pattern or model that is identified. An illustrative example I tend to deal with some messages immediately but often flag and tag messages for later follow-up. I also auto-redirect some messages to different folders and manually move messages to folders on a regular basis. This works up to a point but, when the volume of messages gets too great, important flagged messages get lost and unanswered and, contributing to this problem, when I am very busy I tend to neglect to unflag messages that have been dealt with. Furthermore, having flagged mail in different folders makes some messages less obvious. I know - there are other and probably better ways to manage email! However, a system to support this style of email management might (for instance) allow me to bring up the list of tagged messages at the touch of a button, including the first line or two of the message, with simple options to untag them, archive them, send an auto apology or deal with them directly. Such a system might alternatively/as well pre-emptively warn me of such messages after a pre-specified time (perhaps integrated with existing message filters for greater relevance) and might even be configurable to send a response to the sender, based on pre-specified rules, with (perhaps) an appropriate apology and suggestion to resend. This is just an example, of course, and there may be many far better ways to handle this situation, including switching to a different style of management. Indeed, a program that helped me to adopt a different style that has been identified as more successful, or that maybe even enforced that different style or that translated my actions to conform to that style, might be a good solution.