tobac icon indicating copy to clipboard operation
tobac copied to clipboard

Add zenodo json file to fix title and contributors

Open mheikenfeld opened this issue 2 years ago • 15 comments

Currently the doi on zenodo creates title and authors automatically from the repo info. This leads to some unwanted behavior.

  • Not really consistent and complete author/contributor list such as partially cryptic Github usernames and lacking the author information for those who have not contributed code via Github).
  • Not enoguh control over the title

This can be fixed with a .zenodo.json file, where we can give some of the information explicitly, while other things like the version will be taken directly from the repo.

I have added my author information, maybe everyone can add their information as part of this PR..

mheikenfeld avatar Jun 09 '22 21:06 mheikenfeld

Thanks for filling in your name and Info @JuliaKukulies

mheikenfeld avatar Jun 27 '22 17:06 mheikenfeld

It would be great to get everyone's information in before we merge this in so that we get a complete list before zenodo makes use of this.. @freemansw1 @w-k-jones @fsenf @snilsn @galexsky @zssherman @zxdawn @deeplycloudy @kelcyno @rcjackson @w-herbst

mheikenfeld avatar Jun 27 '22 17:06 mheikenfeld

Please tag/add anyone that has slipped through my short search through the contributions in the different branches, thereis probably a few additional names, whose Github accounts I did not find here...

mheikenfeld avatar Jun 27 '22 17:06 mheikenfeld

@mheikenfeld just added my affiliation via PR to your branch.

deeplycloudy avatar Jun 27 '22 17:06 deeplycloudy

Em ... How could I fork @mheikenfeld 's branch to my current old branch? I come up with creating a new branch and cloning data. But, it sounds like a wrong method.

zxdawn avatar Jun 28 '22 15:06 zxdawn

There are a number of people who should be on this list that aren't on/aren't active on GitHub, so I think this is better resolved simultaneously via email. I'll send an e-mail to everyone that I think should be on this list.

freemansw1 avatar Jun 28 '22 18:06 freemansw1

Em ... How could I fork @mheikenfeld 's branch to my current old branch? I come up with creating a new branch and cloning data. But, it sounds like a wrong method.

@JuliaKukulies @snilsn @deeplycloudy: Maybe one of you could quickly explain how you set up the pull request to the pull request, struggling with the same right now..

mheikenfeld avatar Jun 28 '22 20:06 mheikenfeld

How about about adding something like the "tobac community" as the first author and then all the individual names. I've seen this in a few examples of citations on the web, often with big collaborations or institutions, but I'll try to dig out an example of something more similar to us... This would keep us from having the order of the contributors play too much of a role and also mean that we will have a constant way that the citation would look like in a name/year format, whatever happens to the individual contributors to the package in the future....

mheikenfeld avatar Jun 28 '22 20:06 mheikenfeld

Agree! The first author always has a special role among all the contributing authors in a paper. Using a synonym would help us to reflect that our developers base is dynamically changing ...

fsenf avatar Jun 29 '22 07:06 fsenf

Maybe one of you could quickly explain how you set up the pull request to the pull request, struggling with the same right now..

I used the edit-button in your fork, here: https://github.com/mheikenfeld/tobac/blob/zenodo_json_contributors/.zenodo.json

After filling in my information, clicking 'Propose changes' created a branch in my own fork, from which I could then do a PR.

snilsn avatar Jun 29 '22 08:06 snilsn

How about about adding something like the "tobac community" as the first author and then all the individual names. I've seen this in a few examples of citations on the web, often with big collaborations or institutions, but I'll try to dig out an example of something more similar to us... This would keep us from having the order of the contributors play too much of a role and also mean that we will have a constant way that the citation would look like in a name/year format, whatever happens to the individual contributors to the package in the future....

Great idea!

JuliaKukulies avatar Jun 29 '22 10:06 JuliaKukulies

Em ... How could I fork @mheikenfeld 's branch to my current old branch? I come up with creating a new branch and cloning data. But, it sounds like a wrong method.

@JuliaKukulies @snilsn @deeplycloudy: Maybe one of you could quickly explain how you set up the pull request to the pull request, struggling with the same right now..

Maybe one of you could quickly explain how you set up the pull request to the pull request, struggling with the same right now..

I used the edit-button in your fork, here: https://github.com/mheikenfeld/tobac/blob/zenodo_json_contributors/.zenodo.json

After filling in my information, clicking 'Propose changes' created a branch in my own fork, from which I could then do a PR.

Maintainers can also directly push to the PR by adding the fork of @mheikenfeld as a remote to their local clone, check out the zenodo_json_contributors branch and then push the changes from you local branch to the fork (see here for commands).

JuliaKukulies avatar Jun 29 '22 10:06 JuliaKukulies

Maintainers can also directly push to the PR by adding the fork of @mheikenfeld as a remote to their local clone, check out the zenodo_json_contributors branch and then push the changes from you local branch to the fork (see here for commands).

This is the way how I did ...

fsenf avatar Jun 29 '22 14:06 fsenf

I sent out my e-mail this morning, but I'm going to give it a week or two before I start filling everything in. Given the timeline on this, I'm inclined to push out the milestone here to include this as v1.4.0.

freemansw1 avatar Jun 30 '22 02:06 freemansw1

This is a good idea, thanks for setting this all up. It will also be good to list this as tobac community. Initially we could then have an alphabetic author list, which will of course be re-ordered by contributions for each paper etc. Alternatively, we could have a core contributor group listed first, but that definition quickly becomes a mess.

PhilipStier avatar Jun 30 '22 10:06 PhilipStier

Let's keep the first author as the "tobac community", as discussed here (this is mostly a reminder for myself).

freemansw1 avatar Sep 09 '22 14:09 freemansw1

Codecov Report

:exclamation: No coverage uploaded for pull request base (RC_v1.4.0@0d098ac). Click here to learn what that means. The diff coverage is n/a.

@@             Coverage Diff              @@
##             RC_v1.4.0     #139   +/-   ##
============================================
  Coverage             ?   35.00%           
============================================
  Files                ?       11           
  Lines                ?     2100           
  Branches             ?        0           
============================================
  Hits                 ?      735           
  Misses               ?     1365           
  Partials             ?        0           
Flag Coverage Δ
unittests 35.00% <0.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

codecov[bot] avatar Sep 12 '22 18:09 codecov[bot]

OK, I've made the updates here to both add in the information of everyone who e-mailed me and to make things compliant with the Zenodo JSON Schema. I'll open an issue here soon on doing CI for the JSON schema in ~1.4.1.

I've made the first author "tobac community" and reordered the author list to be:

  1. Original Creators
  2. Core Group/GitHub Admins
  3. All others

As @PhilipStier suggests above, I am imagining this as a dynamic list as new versions are implemented.

freemansw1 avatar Sep 12 '22 19:09 freemansw1

I've re-requested a review from @JuliaKukulies and will wait to merge until one of @fsenf or @w-k-jones reviews. @mheikenfeld can you also take a second look at this?

freemansw1 avatar Sep 12 '22 19:09 freemansw1

Author list looks good in general, but I have actually no idea how to test if rendering within zenodo is working well ...

fsenf avatar Sep 13 '22 13:09 fsenf

Oh, I approved too fast. Nils affiliation is wrong. I will update this in the next commit.

fsenf avatar Sep 13 '22 13:09 fsenf

OK, now approved!

fsenf avatar Sep 13 '22 13:09 fsenf

Thanks @fsenf, I'll merge once @JuliaKukulies re-reviews.

freemansw1 avatar Sep 13 '22 14:09 freemansw1

And yeah, there isn't a good way to check rendering within zenodo to my knowledge beyond validating the JSON schema. I've put an issue about that here: #178

freemansw1 avatar Sep 13 '22 14:09 freemansw1