GitHub Organization and name of this project

[jonathanschilling]

I would like to start a discussion on the name of this project and whether it could be turned into a GitHub organization. We would get almost all the available features since this project and contained repositories are open source.

The only remaining question is that of a good name. Stellarator-Tools describes what this codebase is mostly used for. A matching URL should be available which we can register for linking to an overview website or the main gh-pages website. mhd.org is still available but would cost $14500 and how would one apply for funding for this?

The URL serves a second purpose. I would like to publish the existing VMEC wrappers and the webservice core I wrote in Java and that are heavily used at IPP in the global repository Maven Central. The Maven coordinates typically feature a reversed-URL groupId, which means that if the root URL would be mhd.org, we could publish the code as e.g. org.mhd:VmecWebservice.

Here is a list of suggestions:

• MHD-Tools (mhd-tools.org) (hmm, not much better than Stellarator-Tools...)
• Stellarators For Fusion (s4f.org)
• Tokamaks and Stellarators (tokstell.org) (since VMEC can also do Tokamaks)
• ... your ideas ?

[gmweir]

The toolset and the VMEC work are essentially separate projects even though the toolset is dependent on changes to the code.

In my opinion it is unnecessary to purchase a domain and maintain a website for this kind of toolset. Please double-check the copyright status of the non-stellopt core tools before publishing on github or elsewhere. Things like MINERVA (and their constituent parts), MCONF, etc. are under strange copyrights.

My understanding is that this project is mainly focused on streamlining the VMEC codebase (consequently its dependencies and compilation) and exposing data in memory to external tools instead of formatting the data on disk. In order to prevent very nice improvements from becoming orphaned at the end of your PhD research I would suggest either a fork or branch of the stelllopt repo for the VMEC work. Something that would allow a developer to merge it in the future.

[jonathanschilling]

@gmweir Thanks for expressing your concerns on this topic.

Most, if not all, applications building on top of VMEC or using its output have implemented some sort of wrapper or toolbox to compute physics-relevant quantities. What I am aiming at is a clean room design of these toolbox functions. By definition, this involves complete specification of the functionality and documentation of the implementation. I have performed this work already for the basic routines used within the VMEC webservice at IPP. These routines are implemented from scratch and have been written in my spare time; therefore, they are completely independent from Minerva or MConf.

The branch experimental_vmec10 in the VMEC repository was created by me to track my changes and eventually merge them into the mainline VMEC.