This checklist has been developed by the Scottish Covid-19 Response Consortium and has been used to assess the software engineering aspects of "model readiness" for our six epidemiological models. This is intended to be part of a broader scheme to evaluate and describe suitability of model results to be used in decision making so this checklist covers only the software implementation and assumes that other documents cover questions about model validation, quality of science, data provenance and quality and policy readiness.
In order to use this checklist for the SCRC FAIR data pipeline components, three questions that apply only to software producing scientific results have been marked "models only" and for other software an N/A response can be given.
rDataPipeline
04/10/21
v0.52.0
Name
Sonia Mitchell
Role on project:
RSE
Person identifier:
Do we have sufficient confidence in the correctness of the software to trust the results?
This is your overall judgement on the level of confidence based on all the aspects of the checklist. There is no formulaic way to arrive at this overall assessment based on the individual checklist answers but please explain how the measures in place combine to reach this level of confidence and make clear any caveats (eg applies for certain ways of using the software and not others).
- [ ] Yes
- [x] Yes, with caveats
- [ ] No
The code is well written and includes adequate documentation. Unit testing has good coverage and is fully automated via continuous integration. The Data Pipeline is not yet being used in practice and so bugs may yet be revealed. Additionally, further work is required to implement cross validation against the other language implementations.
Please use a statement from this list: "Sufficiently addressed", "Some work remaining or caveats", or "Needs to be addressed" to begin each response.
Additionally, for each question please explain the situation and include any relevant links (eg tool dashboards, documentation). The sub bullet points are to make the scope of the question clear and should be covered if relevant but do not have to be answered individually.
- [ ] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
- [x] N/A
- [ ] Sufficiently addressed
- [x] Some work remaining or caveats
- [ ] Needs to be addressed
Tests are run via GitHub Actions CI (
) with good code coverage (
).
(We don't require bitwise identical results here, but the broad conclusions after looking at the results of the test case should be the same.)
- [ ] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
- [x] N/A
- [x] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
lintr
has identified a number of cosmetic issues that were considered low priority.
- [x] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
Though the code has survived multiple iterations of the project, it is well structured, commented and documented. Further refactoring is required to reduce duplication and improve modularity.
- [ ] Sufficiently addressed
- [x] Some work remaining or caveats
- [ ] Needs to be addressed
There is a readme that describes how to install the code and user examples that show how the API interacts with the Data Pipeline. Additional documentation exists (
), however this needs expanded upon.
- [x] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
The code is stored in a GitHub repository, which includes a licence and issue tracking.
- [x] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
Dependencies are listed in the DESCRIPTION file.
- [ ] Sufficiently addressed
- [ ] Some work remaining or caveats
- [ ] Needs to be addressed
- [x] N/A
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.