pyOpenSci Meeting Notes - 21 March 2019#

https://hackmd.io/Pma1-c6eRSCrvG-ixv17Jg

Attendees#

Leah Wasser - CU Boulder
Jenny Palomino - CU Boulder
Kylen Solvik - CU Boulder
Carson Farmer - External advisor, Textile.io
Luiz Irber - DIB Lab, UC Davis
Leonardo Uieda - University of Hawaii at Manoa
Filipe Fernandes - NOAA

Agenda Items#

Conversation with JOSS - should know in the next few weeks if our review process is accepted
- they are interested in partnering with us, similar to their partnership with rOpenSci
- when someone submits a package to pyOpenSci, they can also submit it to JOSS to get a DOI from them
  - robust review process, as compared to Zenodo DOI
  - You can have both: a DOI for the package from Zenodo and a DOI from JOSS for the paper/publication
Discussion of review process
- When should a package be submitted?
  - should be fairly well-developed; if early in the package development, potential submitter should complete pre-review steps: https://www.pyopensci.org/dev_guide/packaging/packaging_guide.html
- pyOpenSci should provide a reason for why go through the pyOpenSci process, rather than just JOSS?
  - When someone submits – maybe we have a curation … support process to help people with testing , docs, etc etc
  - If we make it easier for people to submit, to setup tests, and docs etc, we might get more submissions, does rOpenSci do mentorship with testing, docs, etc etc ? FOLLOWUP with Karthik and Carl – do they mentor submissions?
    - cookie-cutter: https://github.com/pyOpenSci/cookiecutter-pyopensci
  - Maybe look into how much hand-holding/support does rOpenSci provide for submitters?
    - maybe we help them with testing, docs, etc
    - pyOpenSci is dedicated to education around OS development, so this fits mission
    - Possibly have three categories for submission:
      - pre-submission with support (submittter requests guidance on specific tasks such as testing, docs; could also be labeled as such by the reviewer)
      - pre-submission
      - submission
  - We should allow for some flexibility in the review process – people might be set in their ways
    - Link to current requirements: https://www.pyopensci.org/dev_guide/packaging/packaging_guide.html#overview
    - Checklist for the reviewer after the submission issue is opened: https://www.pyopensci.org/dev_guide/peer_review/reviewer_guide.html
    - Review Template: https://www.pyopensci.org/dev_guide/appendices/templates.html#review-template

What happens after a package is accepted?
- packages goes to JOSS if it was requested and in scope for JOSS
  - technical review is done already
- decision on where the repository will live
  - they can keep in their space if organization
  - or they can decide to port it over to pyOpenSci
- accepted packages will be listed on pyOpenSci website and send out via social media announcements
- One goal of pyOpenSci is provide a space/mechanism for faster discovery of packages
- So one benefit of the review is that it is a community approved package and will now be easier for others to discover and use

Outreach efforts - https://docs.google.com/document/d/1OpAgnky2NxD0KghRhINVLmtvoaq4PCBikfHgH9HjgCI/edit
- updates on who has reached out to whom… please update the google doc with your name and who you’ve contacted
Website (http://pyopensci.org) - how do people get involved with us?
- Landing page: https://hackmd.io/JaJtuFKJSQm3lc9WVLTI7w?edit
Packages - we need a package or two to review!!
- Scope: https://www.pyopensci.org/dev_guide/peer_review/aims_scope.html
- Goal to review 1-2 packages before BoF
- We need to find ~2 reviewers
- Filipe has suggestions for package
  - wrapper for an API to obtain data
    - https://github.com/ioos/erddapy
- Maybe Max Joseph - streamstats?
  - https://github.com/earthlab/streamstats
- Striplog is a package for well log data
  - https://github.com/agile-geoscience/striplog
- Create a new issue in repo to kick off review process:
  - https://github.com/pyOpenSci/software-review/issues/new/choose
Community meetings? do we want to have a monthly or bimonthly meeting??
- continue bi-monthly at this time
- will revisit as the group grows
BoF - scipy invite!!
- Call is open now until May 31st: https://docs.google.com/forms/d/e/1FAIpQLScNO_scayKeQapQxZRNeFOqql35HY9nRc6gCpxEk8VkZBAkLQ/viewform

To Do:#

rOpenSci – has discourse forum to answer questions. it might be better to document things here! we could set this up.
Filipe will submit a package to the review process
Should we submit earthpy? or something else?

Resources:#

Dev Guide: https://www.pyopensci.org/dev_guide/intro