DOC: Document policy on colormaps and styles #28815
Conversation
f81f40b to
555351b
Compare
There was a problem hiding this comment.
Just remembered about https://github.com/matplotlib/matplotlib/blob/main/doc/devel/license.rst?plain=1
e36b3c2 to
2763352
Compare
doc/devel/api_changes.rst
Outdated
| - novelty: Does it support a new use case? e.g. slight variations of existing maps, | ||
| sequences and styles or yet "another divergent colormap" are likely not accepted |
There was a problem hiding this comment.
| - novelty: Does it support a new use case? e.g. slight variations of existing maps, | |
| sequences and styles or yet "another divergent colormap" are likely not accepted | |
| - novelty: Does it support a new use case? e.g. slight variations of existing maps, including divergent colormaps that are just a new combination of three colors, sequences and styles are likely not accepted |
Just tweaking b/c original wording sounded a bit insider baseball (maintainers annoyance) to me
There was a problem hiding this comment.
I've left out the "diverging colormaps" part. Stuffing it into the enumeration "maps, sequences, and styles" distracts the flow of the sentence too much. And also, we can leave it on the abstract level of slight variations without going into examples.
2763352 to
439e01d
Compare
Also closes matplotlib#27673 by removing that page. The relevant content is transferred to the new section. Co-authored-by: hannah <[email protected]>
439e01d to
06d5c27
Compare
|
This only had one approval for a policy documentation. Maybe this should have more time for discussion before becoming the defacto "policy"? |
|
I see this as rough guidelines rather than a fixed policy. But feel free to make a follow up issue or PR. |
|
I'm fine with it. Just if we are calling it a policy it may be good to have more time for discussion. |
|
I strongly encourage using the "needs discussion" label for things anyone feels need more time for discussion. But also I saw this as writing up the things we've informally been telling folks for ages, and that Tim had addressed all the review comments so figured there was consensus? |
|
Not a big deal - just think that, in the future, if something is going to be a "policy" we should have two sign-offs versus one. |
Also closes #27673 by removing that page. The relevant content is transferred to the new section.
Note: no backport necessary as we always use the devdocs version for
/doc/devel.