Open Science

The label "Open Science" is a pleonasm, science must be open to be verifiable and useful to others. With the Internet and cheap data storage, there is no excuse to not share each other's work.

Let's be honest, a paper is never enough to grasp all details of a work, especially when a computer is involved. People interested in your work often look for details of implementation, or a certified way to reproduce your work and compare with their own work.

Never be ashamed of what you share, none of us is perfect, just do your best. With time and practice, organizing your work to be understandable by others will become easier. People that are really interested in your work will take time to dig into it, and they can always ask you questions if they need to.

Use open source software as much as possible, so people without access to expensive software can reproduce your work. Once we all use and contribute to open source software, they will have the same features than proprietary softwares. With the bonus that anyone could add the latest features they are working on.

Store your data in plain text as much as possible. The data format you use today might be obsolete in a few years, plain text won't.

Sharing your work also help establishing good scientific practice, which is really important: http://nautil.us/issue/24/error/the-trouble-with-scientists

Some papers:

Some links to get started:


Unrelated but some best practice for Python: https://gist.github.com/sloria/7001839