Over the past 2 months I have attended 2 conferences about software development. The first, Python in Astronomy, focused on the software tools astronomers have developed to help them to do research and how we can support each other, mostly in Python. The second, DjangoCon Europe, was for developers who are either freelancers or work for software houses. There were almost 10 times as many people at DjangoCon (400) as at Python in Astronomy (50).
My interest in going to these was that LCOGT uses an awful lot of Python and Django, from our data analysis pipelines, Open Access archive, observing portal, to the website this blog is published on.
Some of my personal highlights from Python in Astronomy
- astro Toyz - A library which makes an astronomical data analysis website. It is a very early release but quite amazing,
- astropy - the most complete library of astronomy utility functions. From cosmology, FITS image analysis, to easy conversion between different units (even expressing the speed of light in Furlongs per Fortnight),
- astroquery - Simplifying the way to look for astronomical data in remote archives. (I've added a module for querying the LCOGT archive too).
Some of my personal highlights from DjangoCon
- Baptiste Mispelon's Adventures in Djangoland - a journey of discovery using Emoji!
- Ana Balica's terrific guide to Demystifying Mixins in Django,
- Erik Romijn's practical guide to improving security of your Django sites, including Pony Check Up,
- Advanced unittesting with py.test by Mathieu Agopian,
- Django REST framework - making an API in Django, the easy way, by Xavier Ordoquy,
- Klean Kanteen water bottles in the conference pack!
Open Source
From the first day of LCOGT we have had a commitment to being open source about all of our engineering (whether it is software, electrical or mechanical). We also maintain an "Open Access" image archive where all of our education and public domain observational data appears the instant it is observed. We also make all of our scientific data available after a proprietary period (in most cases it is 1 year - to give the scientists who took the observations a chance to publish papers on their finding).
I gave talks at both conferences and the reaction to each was totally different, although both sets of responses were very encouraging and friendly. Whenever I've spoken to astronomers (professional or amateur), schools or the public about LCOGT, almost the first question is "How can we get access to the telescopes?". I was totally unprepared for the Djangonauts' first question to be "Can we help with the coding?". Actually, it was not just the first question, but the second, third, fourth, fifth....
I know many of the Python in Astronomy people but almost none of the DjangoCon audience, yet I was bowled over by the warmth that they recieved me and my talk.
If you are a developer and would like to contribute to any of the LCOGT software, please contact me.
Ps. The snake reference is obvious (even though Python refers to Monty Python). The 'pony' in the title is a reference to Django's unofficial mascot, a pink winged pony, which is "awesome and has magical powers".
