2018-04-03 conda-forge meeting¶
New items
Community management
Community policy
Governance
Code of Conduct: https://docs.google.com/document/d/10dxX0Zse0Rx1HqsxC73Wfsghmy5m8PP8cHuBIOhWKpc/edit
Contributing guidelines
Org policy
Research into wikipedia’s governance https://www.youtube.com/watch?v=ZSQJYEVcMWM&index=27&list=PLwfR9EQmUekU3AprkcCVcoRKXXVIt-5E_
What to do with hostile maintainers, eg: https://github.com/conda-forge/lmfit-feedstock/pull/13 and https://github.com/conda-forge/asteval-feedstock/pull/6
Creating a core mailing list
Reach out to phil for email access to the
@conda-forge.org
Done
Adding people to cf/staged-recipes
Marius van Niekerk offered to help review on staged-recipes
Win32 poll results: https://docs.google.com/forms/d/1DbPWfHw1AhYWNsfsYzEo1AxZzKOpxoc-b7zaqY_AMls/edit?usp=sharing 91 responses, 5 are using Win32, only 1 is using Win32 exclusively.
nsis only works on 32-bit — it is used by constructor on windows so it’s a critical package to keep working
Filipe will craft an email to let the community know.
pypi/anaconda.org/wheel/conda-package
Collaboration (or at least communication) with the pypi/warehouse devs
Optionally building wheels for some packages. ( https://github.com/conda-forge/conda-smithy/issues/608 )
TODO: Chat at anacondacon about this
How do we want the bot to handle non-release releases (alpha/beta/dev/pre/etc.) https://github.com/regro/cf-scripts/issues/86 and https://github.com/conda-forge/conda-forge-enhancement-proposals/pull/3 and https://github.com/conda-forge/matplotlib-feedstock/pull/24#issuecomment-221496870
conda-forge-pre?
TODO:
We will blacklist all “non-release” releases and not support automated releases from these tags. Do not support rc releases for now.
Better maintainer training (branching structure, etc.)
John suggested (on gitter) that we reach out to intel and NVIDIA to get copies of their toolchains and development libraries.
We’ll cross this bridge when we have interest/time
GPU discussion - https://github.com/conda-forge/conda-forge.github.io/issues/63
pygdf as a test case with NVIDIA folks - https://github.com/gpuopenanalytics/pygdf
https://github.com/gpuopenanalytics/pygdf/tree/master/conda-recipes/pygdf
flit install as build step?
TODO: Write some dev documentation about supported modes of install