-
Notifications
You must be signed in to change notification settings - Fork 3
FAB Project Board Agenda Minutes October 2021
A1 – Fab team to engage with Stuart and Victoria in their GITHUB analysis.
A2 – SW to report back on progress with using a VirtualBox with ubuntu to perform more interactive testing.
A3 – RG to add a high-level statement of the project’s scope and circulate it for review to GG and DB.
A4 – SW to forward the latest UML diagram to Richard Hattersley for review, and check whether class diagrams meet the requirements.
A5 –RG to discuss with DB how best to communicate with partners to seek expressions of interest for collaboration.
Phase 4 of the project is still technically live. The main goals from the phase 4 plan have seen some progress:
- Building JULES largely works with some minor tweaks needed, eg to pick up a missing pattern
- Fortran-C capability is on Master, the main thing needed for the UM
It is proposed that we close phase 4 to make a clean break for the fresh start promised by the arrival of contractors.
In October, 2 contractors will be starting with us- David Brown on 11th October, and Byron Blay on 18th October (date TBC). This is a new departure for us, and is a calculated risk to learn about how to work with external resource. Thanks to Keir Bovis and Phil Gill for helping with recruitment.
It is expected that the first couple of weeks will involve corporate onboarding, and familiarisation with Fab and all the supporting things that implies. Line management will be overseen by Keir Bovis, and Fab task management by Richard Gilham.
The plan is for David and Byron to push Fab forward to demonstrate a Minimum Viable Product for UM and LFRic (excluding, eg, incremental builds). This will require both development of Fab's functionality and applying it to the JULES, UM and LFRic codebases. Note that JULES and UM bring in successive portions of the code compiled by LFRic, and provide many co-benefits in terms of quality and sustainability.
It is proposed that the following milestones are set up with the NGMS Programme, largely representing a retiming of existing ones:
- Mar-22 Fab UM build demonstrator
- Jun-22 Fab LFRic build demonstrator
Later phases would look at enhancing Fab such that it may take over from existing solutions. the work required would be based on an assessment of the demonstrators. Implementation by LFRic and/or UM would require discussion with the relevant projects.
David and Byron are experienced developers and will likely have views about how they would like to work. We must respect this and allow them to self-organise for efficiency. Both lean towards Agile in their approach.
At the outset, the existing Fab project team will need to take time to get them up to speed both in terms of what has been done and what development work is anticipated to meet/exceed the two milestones. Throughout, it is expected that the project team will act in this consultant role.
Initially, we will take a cautious approach to protecting Master. One of the existing project team will review Pull Requests and Met Office staff have oversight of the quality processes. As we build confidence in each other, we will consider relaxing this. However, we expect good communication to be maintained to ensure Met Office team members retain good knowledge of Fab.
In terms of communication, we will leverage existing comms routes. We will also discuss the possibility of daily/weekly standups, depending on what works for all.
The Project Board should be updated approximately monthly on progress so that any adjustments can be made.
Long-standing resourcing challenge remains with Fab being out-prioritised. Contractor resource due this month.
No concerns at present. Keir Bovis will line manage our contractors and ensure the basics are taken care of.
The risk and issue log has had an incremental refresh.
- Staffing the dominating issue, but with a major mitigation in the form of contractor imminent.
- Also a minor niggle about GitHub actions changing under us. Possibly an illustrative point for NGMS more generally.
Appointment of contractors offers a good opportunity to ensure Fab is healthy.
At present, the stakeholder analysis is probably fit for purpose because it keeps us focused on our main goal. Once fab is on the way to being proven, then reaching out further would be desirable.
Once the plan has been agreed, we should determine if it needs to pass through the formal NGMS change management process. A further iteration may be needed later if our estimate of progress through using the contractors is too optimistic.
Project Board membership- Doug and Matt replacements?
Date TBC, aiming for early December.
- Future Release
- vn1.0 Release, March 2023
- 0.11 Beta Release, Jan 2023
- 0.10 Beta Release, Oct 2022
- 0.9 Alpha Release, June 2022
- Phase 2
- Phase 3
- Phase 4
- Repository Management
- Development Process
- Development Environment
- Releasing Fab
- Coding Conventions
- Glossary
- Concerning the Database
- Unit Test Coverage
- Issues With the System Testing Framework