[30.6] Python/Django meetup


#1

On Friday night, a couple dozen Pythonistas and those curious about coding got together at the new Effinger underground in Bern for a meetup about code and data. This is one of the first user groups using our newly launched space, and we are very excited to host them all! Several Powercoders also joined in who were having a monthly get-together.

Thanks to Andreas Keller, who is organizing the upcoming React Native event, for his help with the meetup.

Read the recap on the blog or below, but first, check out these events coming up next at E-0010!

Python and open data

Photo by Stefan H. via Twitter

Expanding on my thoughts from the Tuesday Data Infrastructure workshop, I gave a tour of topics central to the open data movement with a focus on work in Python - talking about how we are working on improving data access and quality through various open source projects.

It was described how portals provide simple and secure access to open data sources, of them most prominent the Python based CKAN which powers DataHub.io, opendata.swiss, Stadt Zürich, open.canada.ca, and many more. I spoke about the way Python is used both for specialized tools as well as general web publishing, like through the Pelican static generator, which powers several projects I work on including handbook.opendata.swiss and index.okfn.org.

We had a rather extensive discussion then about data formats, a supportive audience helped both to explain topics like the Frictionless Data, Linked Data, Graph databases, and covered a few data management principles. I gave a sneak preview of the Django-based Wagtail CMS which we are using at Datalets.ch, among other things to build API-powered, single-page or “headless” websites.

I didn’t really have time to talk about the Flask-based Dribdat project I’ve been developing for open data hackathons. Or Datacentral - the alternative open data portal, or the Good Tables service, both Python based. Or describe the way Python is used at Jugend Hackt events which I’m a fan of, and the role of Python in data rescue initiatives like DataRefuge - all good topics to cover another time :slight_smile:

Many thanks to 89grad and to Liip for sponsoring the event. Speaking of whom, for provocative further thoughts and an introduction to open data (about time we saw some python posts, people!), check out what Liip’s Isaline Mülhauser has to say:

See you at the next meetup! Please leave a comment if you would like to run your own presentation for the Python Meetup Group, or if you’re interested in using the new E-0010 space, or if there’s anything else at all you’d like to share.


#2

A couple of people asked about the open data I showed as being quite useful to the developer community…

A similar more Python-centric notification service I use for some projects is http://pyup.io … Leads me to think that perhaps a good title for a next time could be “open data at the service of the Python community” :wink:


#3

Speaking of Libraries, a fresh post from our friend Timo Grossenbacher outlines some of the challenges in dependency tracking from the standpoint of reproducible data science. He talks about R Notebooks, but I think the issues are just as valid for folks working with Python / Jupyter: sometimes just a small change to the version of one dependency has deeper consequences. Keeping our code running in an up-to-date environment remains a challenge. Worthwhile reading and links: