Initial adaptation from Jupyter to matplotlib #1
Conversation
| more legal and financial decisions and interacts with other | ||
| institutions, we see a need for a more formal governance model. Moving | ||
| forward The Project leadership will consist of a Steering Council with | ||
| the power to appoint final decisions to an individual. We view this |
There was a problem hiding this comment.
Something is wrong here--I'm not sure what you intended to say.
There was a problem hiding this comment.
Yeah, I'm not sure how "with the power to appoint final decisions to an individual" is better than "a BDFL", as was in the original text from jupyter.
There was a problem hiding this comment.
Sorry, we started trying to remove the BDFL and then went back on it, but forget to revert this change.
|
There's a phrase "The BDFL can appointing his/her successor" (came over from Jupyter, so not in the diff) that should also be cleaned up. Otherwise, I have no problem with this. I think the model has served Jupyter well to this point. |
|
I've just pushed some changes that I hope address the comments here. Additionally, on a second read I decided to elaborate on the history section a bit. It draws the linkage from John through Mike to Thomas, so it isn't quite so jarring when Thomas is first mentioned in the BDFL section. |
| - Thomas Caswell | ||
| - Michael Droettboom | ||
| - Phil Elson | ||
| - Eric Firing |
There was a problem hiding this comment.
Subcommittee members must have contiguous last initials, apparently.
| project affiliated with the 501c3 NumFocus Foundation. The goal of The | ||
| Project is to develop open source software and deploy open and public | ||
| websites and services for data visualization. The Software developed | ||
| by The Project is released under the BSD (or similar permissive) open |
There was a problem hiding this comment.
Assuming this covers everything under the Matplotlib GitHub organization, basemap bundles GEOS (LGPL) and GMT data (GPL), and natgrid's license is "Restricted" (whatever that means), which I think don't strictly fall under "permissive".
There was a problem hiding this comment.
The only weird thing about natgrid is src/nncrunch.c. The main source is BSD-2 clause and @jswhit 's code is MIT (I think?).
Is there a way build basemap without bundling those dependencies?
thoughts @WeatherGod ?
There was a problem hiding this comment.
with respect to basemap and GEOS, the source tree merely vendors a version of the library, but doesn't require that it be used, IIRC. As for the GMT data, I have always been highly skeptical of the GPL claim for the data. The software (which is not included) is GPL, and that is fine, but I have had a hard time believing anyone can even apply GPL to a dataset.
As for unbundling the data from basemap, it is technically possible (conda-forge supposedly does it), but no one has submitted a patch upstream to do it there as well. However, basemap is pretty much useless without that data. It is theoretically possible to use a different dataset (Cartopy uses another dataset), but converters would need to be written to generate them into the internal format used by basemap (which is pretty much undocumented).
There was a problem hiding this comment.
Should we just explicitly exclude basemap as falling under different rules (wrt licensing) so we can move forward with everything else?
|
All: I've made the final "simple" corrections and merged this. I think it was most important to begin to have something in effect than waiting for perfect. Specifically, I think @Carreau's discussion about "broader" community, and @OceanWolf's points about more specifically defining certain groups are good points, and those discussions can continue in other issues and PRs. Thanks everyone for helping with this, and congrats, @tacaswell. |
This is a first pass at adapting the Jupyter content for matplotlib.