PyCon 2015 in review
A couple weeks ago we headed off to PyCon 2015. If you missed PyCon, there is another opportunity to come see us. PyData Seattle is hosted at Microsoft’s Redmond campus in July 24-26. The call for proposals and early-bird registration are still open, so if you’re a data scientist using Python or simply interested in learning about it, we’d love to see you there.
Now, we are back from Montreal and all caught up on sleep; but, before we get back to work (now fully inspired!), we thought we’d share our experience at theconference.
We launched a new series of videos for PTVS at the conference. These videos are short (less than 5 min) and will help jump start your Python journey. They range from getting started guides to help set up Visual Studio for Python development through to deep-dive tutorials about Cross Platform Remote Debugging, How to create a Python Django Website, Creating Native Modules, and much more. This is just a short list we have started with and will continue to add to. If there is something in specific you’d like to see a tutorial on, please send us your requests.
Apart from these videos, throughout the PyCon tutorials, sessions, and sprints, we met lots of great people, saw some amazing talks and actively contributed to the event and community. Here’s a few things we discovered and shared at the conference!
One of the reasons we love going to PyCon each year is to tell people about Python Tools for Visual Studio. The community is very fond of simple editors like vim (and not-so-simple ones like emacs), so the functionality available in Visual Studio is often a very pleasant surprise. Tasks like setting breakpoints and getting code suggestions are so much simpler in VS that by the time we get to the really cool stuff like cross-platform and mixed-mode debugging, people have already started using it for their next Python project.
This year we also got to show off just how far we’ve come with Azure since last year. Our workshop helped about 50 people get started with great features like Azure Marketplace and Application Insights, and some of the easy on-ramps like try.azurewebsites.net encouraged even more people to try out Azure.
Our most recently launched service, Azure Machine Learning, was very popular. The Python community has a lot of data scientists and researchers, and being able to show them that R and Python are both tightly integrated went a long way towards relieving fears of vendor lock-in. We had a number of data scientists visit our booth and try it out (though they may have just come for the quadcopters…)
Both during and after the main conference, we spent a significant amount of time working with other Python developers and the core Python team to help make Python easier on Windows. One of the major problems we are aware of is that many packages won’t simply “pip install” on Windows, so we are actively working with other teams at Microsoft and external people working on open-source compilers to alleviate this concern.
Attendees at the Python Language Summit
During the development sprints we spent some time making further improvements to the Python installer for Windows, which you can preview in Python 3.5.0 alpha 4. We want to make it really easy for people to get and use Python, and a great way to help is by actively contributing to the core Python project.
We’d like to thank everyone who came to PyCon this year and chatted with us, as well as all the people who tried out PTVS, Azure or Azure Machine Learning for the first time. It’s great to be part of a thriving community, and the Python community is one of the friendliest and most exciting out there. We are looking forward to next year’s PyCon as well as PyData Seattle later this year. Hope to see you there!
|Steve Dower, Software Engineer, Python Tools
Steve is an engineer who tells people about Python and then gives them excuses to use it and great tools to use it with. He works on Python Tools for Visual Studio, contributes to many of Microsoft’s Python libraries, and is a core contributor and Windows expert for CPython.