Interview: David Jackson at the UK Met Office

This post originally appeared on the Software Carpentry website.

The second of today's sponsor interviews is with David Jackson from the UK Met Office.

Tell us a bit about your organization and its goals.

You can find out more about us at http://www.metoffice.gov.uk, but broadly speaking, we:

predict the weather for tomorrow, next week, next season and beyond;
are a significant contributor to the global understanding of climate change;
are leading researchers of weather science;
provide forecasts for sporting events such as Wimbledon and Open Golf;
stand shoulder to shoulder with Armed Forces around the world;
help keep roads open and planes flying;
inform the decisions and policies of businesses and governments across the world;
help the National Health Service provide preventative healthcare.

Our group is focussed on research.

Tell us a bit about the software your group uses.

We have a lot of home-grown scientific models and related codes around weather and climate prediction.

The Unified Model is at the heart of this the code base.
Much of the compiled code is in FORTRAN.
We use Perl and shell scripting to bind things together, and also some Python.
Our MASS storage system is developed in JAVA.
We use IDL for data processing and R for statistical analysis. There is a lot of user code that sits on top of these base packages and a lot of bespoke development.
We use Subversion for version control and Trac for issue tracking.

Tell us a bit about what software your group develops.

The Unified model and related software is used internationally by a number of groups.
There is a whole range of software development from internationally supported software, like the UM to code scientists write for their own purposes, with other codes supported within the Met Office.

Who are you hoping Software Carpentry will help?

A scientist who does some software development for their own needs.
A scientific programmer who does not have a software engineering background.
In general, people who can code but would benefit from not-too-heavy software engineering principles.

How do you hope the course will help them?

The benefits of software engineering principles and disciples, such as testing and configuration management.
Good code structure practice.
How to make more code shareable and supportable.

How will you tell what impact the course has had?