add note from 2019-09-11

Author: mattip

status_meetings/status-2019-09-11.md

@@ -0,0 +1,71 @@
+---
+tags: NumPy
+---
+
+# 2019-09-11 NumPy Development Meeting
+
+- Time: 11am Pacific Time
+- Join via Zoom at https://berkeley.zoom.us/j/762261535
+- [Trello workboard](https://trello.com/b/Azg4fYZH/numpy-at-bids)
+- [Previous meetings](https://github.com/BIDS-numpy/docs/tree/master/status_meetings)
+
+**Present:** ZJ, Tyler, Warren, Chuck, Hameer, Ralf, Matti, Sebastian, Inessa, Stéfan
+
+## Follow-up from last meeting / discussions
+
+- Ralf opened [PR 14454](https://github.com/numpy/numpy/pull/14454) to document the public API. Do we need to discuss it?
+  - Polynomial namespace, multiple ways of calling the poly functions; Eric Wieser has a [suggestion](https://github.com/numpy/numpy/pull/13601#issuecomment-525591808) on how to fix
+  - TODO Follow up on surveying new public API surface in `np.random` (Matti will try and clean that up)
+
+# Topics
+
+- NEP repo updates github.com/numpy/neps do not reflect to https://numpy.org/neps
+  - Github deploy is not triggered (reliably?) for some reason
+  - Sebastian asked support for help
+- In discussing build warnings, @rgommers raised the following ideas:
+  - hide the _configtest output by default. we regularly get issues about those where people see lots of errors and for example hit Ctrl-C
+    - add a `--debug-config` option to show this output?
+  - introduce a CI run that's clean with the latest GCC. we have a `-Werror` build somewhere, but only with old GCC (existing build matrix entry could be updated)
+  - introduce a CI run that's clean on macOS with Clang (lots of warnings right now)
+    - clang on linux too?
+- Hypothesis in testing. The mailing list discussion was very short. [PR 14440](https://github.com/numpy/numpy/pull/14440) is an example for discussion
+  - Ask for some more examples, especially around bytestrings/unicode, pickling. Don't pin the version.
+- [NEP 32](https://numpy.org/neps/nep-0032-remove-financial-functions.html): So far, there have been only a few responses to the announcements on the numpy-discussion and pydata mailing lists.
+    - Well before the release of 1.18 we will have a numpy-financial package on PyPI and deprecation stubs in numpy.
+- NEP 31
+    - Add `determine_backend`:
+        - Few choices: `__ua_domain__` or making `__ua_convert__` mandatory.
+    - Any additional comments?
+  - The general feeling is that we need to get the correct people on a high-bandwidth discussion. Perhaps at the Nov 22-23 meeting? 
+- Social media channels - reserving for now?
+  - Maybe trademark the NumPy name?
+- Website redesign update
+  - Ralf, Inessa showed off the new site.
+- Logo refresh is going on [numpy.org PR 36](https://github.com/numpy/numpy.org/issues/37)
+
+
+## Additional Details
+
+- Matti
+  - Participated in EuroSciPy. It was nice to meet up with folks. A number of people sprinted, among them we got a PR to use hypothesis in our testing.
+  - Pushed some small changes to Cython, trying to help them get 3.0 out. Also a small fix for the `@property` decorator in numpy.pxd in SciPy.
+  - Adding a new dtype via (ab)use of the object dtype for geodata in pandas
+ 
+- Sebastian
+    - [Extensible Datatypes for NumPy](https://hackmd.io/kxuh15QGSjueEKft5SaMug) Matti had some feedback/clarifications.
+    - Started (prototyping): https://github.com/seberg/numpy/pull/6
+        - Start on `PyArray_InitDTypeMetaFromSpec(...)`
+        - Not quite there yet for AbstractDType for value based casting, but next step.
+
+- Warren
+  - Repo for my experiments with custom dtypes using the current API is https://github.com/WarrenWeckesser/numtypes
+  - Spent time on this issue in numpy.random:
+    - Bug report: https://github.com/numpy/numpy/issues/14457
+    - Initial fix by ahaldane: https://github.com/numpy/numpy/pull/14458
+    - My follow-up PR: https://github.com/numpy/numpy/pull/14465
+
+
+---
+
+Please remember to archive this file and commit it to github.com:bids-numpy/docs
+