Minimum Viable PhD

if you are a software engineer, or developer, you may be familiar with the term Minimum Viable Product (MVP). This is the term used to describe the absolute minimums a developed product needs to accomplish to be useful.

In the terms of your PhD studies we also have a term Minimum Viable PhD (MVPhD). This is something that you should discuss in detail with your supervisory team at the earliest possible juncture, and indeed, we would expect that the supervisory team would provide an MVPhD pathway, as part of their project description for the advertising of the project, and also in discussions with you. Which, of course, might change this MVPhD at some point.

So what does a minimum viable PhD look like? Well the University regulations are reasonably clear.

“The Degree of Doctor of Philosophy (PhD) is awarded by the University in recognition of the successful completion of a period of supervised research and training, the results of which show convincing evidence of the capacity of the candidate to pursue research and scholarship and make an original contribution and substantial addition to knowledge. The results of this research shall then be embodied in a thesis or other appropriate form and must contain material of a standard appropriate for peer-reviewed publication.”

This is the yardstick by which we will measure your progress. So in this case the PhD may not have the outcomes that you expect, or wanted. And it might not be the large piece of research work which you envisaged when you started or thought about PhD studies when you appied.

Most PhDs, are incremental, very few are groundbreaking to the extent that new knowledge is created by the PhD alone, especially in Computer Science. We see more that large research breakthroughs are part of research projects, and those research projects have PGR, who are undertaking incremental but important research work.

So, we would expect that in your first year Research Progress Review, and in the End of Year (EOY), progressions for year one and year two, you should also include a description of what the minimum viable PhD is; whether you have achieved this MVPhD, and what additional aspects/extensions to the MVPhD you’re going to undertake; ie once you’ve completed the minimum, what will be the maximum.

You should be realistic about exactly what the minimum viable PhD is, appreciating that you need to complete your studies in three years. The first year is typically given over to background studies, and a small prototypical experiment, or slice, through a particular problem to gauge the effectiveness of possible solutions. Year two – and some of year three – is where most of the research work happens. And you must have your research complete by the end of year three.

A MVPhD Specification

So what might a minimum viable PhD description look like? A possible roadmap for this PhD may look something like:

  • Year 1: A literature review that establishes the different techniques used to evaluate human memory augmentation systems. Further literature review and/or study that establishes the everyday practices by which individuals determine if their memory is performing better or worse than expected.

  • Year 2: Design and development of a platform that operationalises the everyday practices identified in year 1.

  • Year 3: Use of the platform to evaluate one or more human memory augmentation systems. Writing of thesis (or alternate format submission) for examination.

The student should expect to publish their work in the form of academic papers at venues such as ACM CHI, PACM IMWUT and the Computers in Human Behavior journal.

Thesis by Journal Format

Attached to the concept of the MV|PhD is the way that work will be written up. Traditionally we have used the form of the thesis, in which the last three to six months of your PhD studies are taken up with writing a large thesis text. And forms the basis of your oral examination (your viva).

However, the university has introduced a new format – which is becoming common in the sciences – and this is the Journal Format thesis, in which you assemble published papers relating to your PhD studies throughout the period of your PhD; at the end these papers are amalgamated into a single text.

This has to be done very carefully to follow the narrative of the PhD and so this is why the minimum viable PhD is critically important. Simply you must know what you’re going to investigate out the outset, even if you don’t yet know the results. And you must beable to describe the narrative of this work so that it can be easily understood by a reader – as opposed to being a disjoint collection of your published papers.

Even though each paper is included as a chapter within the PhD thesis, you would still normally need to write an Introduction (of possibly 10 to 15 pages) and Conclusions and Future Work (again, probably five to 10 pages). You would then need to bracket each paper with a couple of pages discussing how this paper fits with the overall research aims, and the trajectory of the overall research you are presenting (and defending). If you manage to publish background work, such as a narrative survey, in, say, ACM Computer Surveys (or a systematic review), then you may have less work to do when it comes to the background chapter. This background chapter typically involves a primer on your specialist domain, followed by related work. The related work section can be narrative, but, in some cases, PGRs include a systematic review to solidify the gap analysis; ie the analysis of related work, such that the gaps in the domain are self-evident.

PGRs should declare their intended thesis format on the ‘Notice of Submission (Student)’ form (eProg). PGRs are also asked to confirm the format as part of the thesis submission process on the eThesis repository.