# Forum on project ideas¶

This page collects project ideas coming from either the Scikit-HEP team or more broadly from anyone from the Scikit-HEP community. Do get in touch if you would like to add an idea to this forum. It will be the best way to raise awareness about your idea and attract contributors!

## Google Summer of Code 2017¶

Scikit-HEP is participating in the Google Summer of Code 2017 program with CERN as an organization, and under the umbrella of the HEP Software Foundation, see the direct link.

We have put forward 2 successful proposals! The direct links to the project proposals are the following:

## Data aggregation¶

• Data aggregation largely means histogramming. An interesting idea would be to exploit the Histogrammar package implementation in Python as a powerful way of dealing with data aggregation in Scikit-HEP.

## Datasets¶

• Datasets are central in HEP and Scikit-HEP too. This important package needs further development and lack parts of the implementation. Get in touch with Jim and Eduardo if the topic is of interest to you.

## Math modules¶

• The 3D and Lorentz vector classes need to be improved to exploit NumPy arrays. There is also functionality to be implemented. | Contact Eduardo if such a development is of interest to you.
• Now that 3D and Lorentz vector classes are available, though in development, it would be handy to build on them. Needed are a 3D point class and then a set of mathematical functions implementing handy geometry-related calculations commonly used in HEP. Contact Vanya and Eduardo if such a development is of interest to you.

## Modeling¶

Means “data models and fitting utilities” at large. Ideas?

## Simulation¶

• Creation of a decay generator base class. A concrete class using ROOT’s TGenPhaseSpace should be an immediate example. Contact Eduardo if you want to get involved.

## Analysis Development¶

• Give users as early as possible an idea on how the project may contribute to and/or help with their analysis, e.g. by giving a small example analysis using the project toolchain.
• Think about including “framework” tools for an analysis, e.g. workflow managers (for example luigi).

## Continuous integration, testing, code health, etc.¶

• Worth adding CI tests with CircleCI?
• Add CI on Windows for as long as possible - nice since a totally different plaform compared to Linux.
• How do coveralls and codecov compare? Worth having both? That would mean adding codecov.
• Look at Landscape and decide if worth having such code quality metrics automatically handled.