PyCon UK 2015

Simple web services for scientific data

Alys Brett, Culham Centre for Fusion Energy

Would you like to let people access your data over the web or generate plots on the fly when someone loads a web page? This session will introduce the benefits of creating web services for accessing scientific data and let you try out the basics for yourself.

It is now common for online companies to provide programmatic access to data - web APIs where data resources in many forms can be accessed via a URL. This approach can be very useful for scientific data too. One benefit is that you don't have to worry about what platforms and languages to support - the data can be used by anything that can make HTTP requests. You might think that creating this kind of web service is solely the preserve of professional engineers but, with the power of Python, this is changing. There are very convenient packages (such as Flask and Requests) that make it incredibly simple to get started.

The session will start with a demonstration of some web services we've developed for nuclear fusion data from the JET experiment, including a plot server and a data browsing tool. This will be followed by a mini-tutorial to help you get started with harnessing the power of HTTP web services.