OpenCyber Academy
Doctoral research · Chapter 8 · English
Course code: OCA-196Chapter 08: OCA-196-C008

Chapter 8 — Defense and publication plan

Open, free learning for anyone who wants to protect people and technology. No account is required to read this text or its worked answers.

Lesson map

Lesson 08 published. Map, example, EX045.

flowchart TD
  N0["Learning outcomes"]
  N1["8.1 Defense"]
  N2["8.2 Questions"]
  N3["8.3 Firmware result"]
  N4["8.4 Publication plan"]
  N5["8.5 Practice impact"]
  N6["8.6 Ethics reflection"]
  N7["Answering the cost question"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6 --> N7
  N7 --> EX["Worked example"]
  EX --> Q["Exercise EX045"]

Logical drawing

flowchart LR
  N0["Learning outcomes"]
  N1["8.1 Defense"]
  N2["8.2 Questions"]
  N3["8.3 Firmware result"]
  N4["8.4 Publication plan"]
  N5["8.5 Practice impact"]
  N6["8.6 Ethics reflection"]
  N7["Answering the cost question"]
  N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6 --> N7 --> Q["EX045"]

[Learning outcomes]
    |
    v
[8.1 Defense]
    |
    v
[8.2 Questions]
    |
    v
[8.3 Firmware result]
    |
    v
[8.4 Publication plan]
    |
    v
[8.5 Practice impact]
    |
    v
[8.6 Ethics reflection]
    |
    v
[Answering the cost question]
    |
    v
[Worked example] --> [Exercise EX045]

Learning outcomes

You will (1) defend the thesis; (2) answer methodological questions; (3) plan publication and practice impact.

8.1 Defense

Yusuf presents the problem, design, results, critiques and limits in forty minutes. The forty minutes follow the thesis's own order, with the design and the Hawthorne response given the most time because examiners were expected to press on them. Yusuf rehearsed with Prof. Solberg and kept one slide that converts the failure-rate interval into failed updates per year, the same conversion that answered the examiner's question about practical meaning. He ends with the limits, so the examiners' last view of the claim includes its boundary, and leaves the remaining time for questions.

8.2 Questions

Examiners ask about generalisation, firmware updates and cost. He answers with evidence and admits where the study cannot answer. On generalisation, he explained that two sites at one trust cannot show the effect elsewhere, and proposed a multi-trust study with randomised start dates. On cost, he said the study measured outcomes, not staff hours, so he could not quantify it. Admitting these limits was more convincing than stretching the data to cover them. Firmware questions he answered from the separate firmware category, with the caution that its counts were small.

8.3 Firmware result

Firmware updates had a higher failure rate in both arms, supporting SP 800-193's emphasis on recovery mechanisms. Yusuf presents this as an observation within the study rather than a main finding, because firmware was a separate category with fewer updates. It supports the recovery principle in SP 800-193 but does not test any particular recovery mechanism, and he says so. The trust's response, adding recovery checks for firmware, follows from the observation rather than from a measured effect. Testing it would need a larger firmware sample than the trust's, which Yusuf names as future work.

8.4 Publication plan

One journal paper on the main effect, one practitioner report for health IT teams, and the open archive. Each output serves a different reader. The journal paper reports method, effects and intervals for researchers; the practitioner report explains rings, the codebook and the limits in plain terms for health IT teams; and the archive lets anyone rerun the analysis. All three use the same wording for the main claim, so no version says more than the others. Each output cites the archive's persistent identifier, so a reader of the practitioner report can reach the same data as a reader of the paper.

8.5 Practice impact

Harrowgate keeps ring deployment and adds firmware-specific recovery checks. The trust's decision came from its own managers after reading the practitioner report, not from Yusuf. Recovery checks for firmware respond to the higher failure rate seen in both arms. The trust also plans to keep collecting the same outcome measures, so it can tell whether the effect lasts beyond the study period. That continued measurement is the practical test the thesis could not provide. Recovery checks are tested before each firmware cycle, and failures feed back into the codebook.

8.6 Ethics reflection

He reflects on balancing research with clinical service: no clinical system was changed solely for research. Every change during the study was one the trust had already planned, with clinical leads consulted on ring membership and timing. Yusuf's role was to measure, not to direct clinical IT. His reflection notes that this limited his design choices, such as randomising wards, but protected patients and staff from research-driven disruption. The ethics committee's conditions were met throughout, and the reflection is filed with the archive.

Answering the cost question

Rings require more planning: defining rings, reviewing failure signals and handling exceptions. Yusuf estimates the trust's added effort at about two staff days per month, compared with the hours saved by fewer failed updates and fewer help-desk tickets. He presents this as a rough estimate, not a full economic evaluation, and suggests a cost study as future work.

Generalisation

The findings apply most directly to organisations with similar device fleets, central management and staff who share devices. Organisations without central management or with very different fleets would need their own evidence.

After the defense

Examiners request minor corrections to the limitations section, which Yusuf completes within a month.

Closing reflection

Yusuf notes that the most valuable habit was pre-registration: it kept him honest when early pilot numbers looked tempting.

Thanks

He thanks Prof. Solberg and the trust's desktop team.

Worked example

An examiner asks what he would do with more time. Yusuf proposes a multi-trust study with randomised start dates, which would address site differences.

Synthetic case. Synthetic.

Errors and limits

Overstating impact in a defense invites hard questions; stay with evidence.

Exercises

Synthesis

Defend with evidence and plan impact honestly.

Continue studying — chapter 08 of 08

← Chapter 07 (draft) · Course chapter index · Worked results ↓ · Next course via chapter index →

Worked answers

EX045. Generalisation, firmware updates and cost.

EX046. Higher failure rates in both arms.

EX047. A journal paper, a practitioner report and the open archive.

EX048. Kept rings and added firmware recovery checks.

EX049. A multi-trust study with randomised start dates.

EX050. No clinical system was changed solely for research.