-
-
Notifications
You must be signed in to change notification settings - Fork 105
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Idea #1 for Team page #45
Comments
The code has now been committed to pandas, the new links are:
If you can be interested in the whole |
Very nice, @datapythonista! |
Hi, the first link team.html is not working. @datapythonista, @rgommers is this information available somewhere else? I checked datapythonista.me domain as well but couldn't locate it. |
Sorry @shaloo, those pages were deployed, and I removed my temporary copy. |
@datapythonista, thanks - it looks v v nice! |
Taking this up over the weekend...thx! |
Thinking about who to list. Some thoughts:
Does that sound okay to everyone? |
@shaloo would it be easy to have separate galleries for Web and Documentation teams? |
I was surprised when I read this at first, so to clarify for others who may also read this incorrectly: Pandas does not list all contributors first, they are mentioned as a group ("1500 contributors"). Maintainers are then listed, individually, with photos. |
Yes, we can do that. |
Ok can be done provided I have a list of who’s who. At present I simply fetched top 18 contributors from Github in order to try out @datapythonista suggestion above. |
AstroPy has done a good job on outlining roles and responsibilities within the project. |
That's indeed quite nice. We're definitely not there yet. Would like to get there at some point though - maybe approaching that level of role clarity by the end of the year. |
Putting together the names for this page was a little challenging, and because we don't have formal teams and roles yet, here is my best attempt:
It also highlighted again that we have multiple people with commit rights who haven't contributed in years, I'm inclined to clean that up. I listed here only active maintainers and web/docs team members. I also judged "maintainers" to be the key code-related group, and "steering council" much less important. @numpy/numpy-core-team comments very welcome. |
@rgommers Thanks Ralf, this looks great to me. |
Regarding integrating this script, here are some initial thoughts. Let me know if you have other/better ideas:
Thoughts? |
We should give Anne Archibald an emeritus entry, she had commit rights when we were on svn that got lost in the move to github because she had moved on. |
Looks great @shaloo! Thanks @charris, good point about Anne. @shaloo could you add her? I was actually thinking to have the emeritus maintainers just as a list of names, not a gallery (rationale: they're gone, so de-emphesize a little).
Agreed, your approach sounds good.
I suspect that this will be nicer than another separate page. |
And Alex Griffin had the cool snail avatar. See https://github.com/argriffing @argriffing. He disappeared so suddenly and completely that I worry. |
Anne is @aarchiba on github. |
I kind of like the gallery. |
@rgommers I'm not sure about the list of emeritus maintainers. Several of these folks do not seem to qualify, but it could be that I missed substantial contributions from them. What is the rule for inclusion here? I also notice that @jarrodmillman is not listed, who managed 5+ releases and coordinated several important community meetings. |
Thanks @stefanv, good catch - added Jarrod. And valid point on criterion. The ones you mean are probably Alex Griffing, Ondrej Certik, Jonathan Helmus and Francesc Alted. They all have commit rights, and were part of the NumPy team at some point. I agree that their contributions are all fairly marginal to be considered for this list. Happy to remove them all, seems fair indeed - there are many other people with similar amounts of contributions. If we disregard commit rights yes/no, then I'd draw the line at Jarrod and Jaime. Okay, removing them. |
fixed that too - they no longer have commit rights |
For the record, my contribution was that I put in many months of work to get NumPy 1.7 out and was the release manager for it. It didn't translate to a lot of commits, but it took a lot of time to figure out how to fix things and how to get everything working on all platforms and how to coordinate with all the contributors. |
@certik thanks for pointing that out. Yes I definitely remember you did 1.7.0/1! It was even longer than I remembered (Jul-Apr), and one of the more painful releases in NumPy's history. I think your contributions were more substantial than the others mentioned above. There's also still a bit of a gap with Jarrod/Jaime, who contributed for 3 or more years - let me confer with @stefanv on this list. |
Oops, somehow also forgot Pearu - added. |
@rgommers indeed, it was a lot of work, over a long period of time at a point when nobody wanted to do this job for NumPy. Whether that meets the bar for this issue is up to you. But I just wanted to set the record straight regarding my contributions to NumPy. |
Having too many emeritus contributors is not a problem we should worry about, having too few is more likely to ruffle someone's feathers. |
Thanks for the input everyone, and the new draft @shaloo. First off all, someone pointed out to me this phrasing didn't sound right: "fairly marginal to be considered for this list". I apologize if that came over wrong - I definitely did not mean to under-appreciate anyone's efforts! What I meant was that what I (and I think also @stefanv) had in mind for "emeritus maintainer" status was people who were part of the NumPy core team long term, typically multi-year sustained contributions. We have many many other people who made lots of valuable contributions over the years. Drawing a line somewhere is hard. @InessaPawson had the very nice suggestion to (on a separate page) list all contributors by name. This sounds like a good idea to me. We can use a script for that and the @shaloo could you make a PR with what you have? And can you add @certik back to that? I may have forgotten a couple of other people, I will go back and double-check everything. |
ok. Will do that in the next couple of hours. You missed Anne - (in your updated list above). I have kept her based on @charris inputs above. When I run the script, I'm having some GitHub data access issues at the moment may have crossed the daily limit for API calls. Will try again in a few hours to get het latest html with @certik. |
Done. Pls review.
|
Can you expand on this? It seems weird to me, since regenerating the html is very cheap. Only retrieving data from the GitHub API should be cached I'd think. |
Yes generating html is cheap and fast if you use stored content. What I meant by above is, in hugo forum I came across a conversation where folks were looking for a way to incorporate statically generated html files as part of their hugo site content and the response was to add the html file in static area and refer to it elsewhere in Hugo content. As of now, I have taken the same approach. |
Okay, I think that discussion may have been about some externally-generated html then. In this case I don't see why we would treat the Team page different from the other pages - it's all html pages generated by Hugo itself. |
I think I am missing something that you are suggesting here. Afaik, the static html page is generated by script and then plugged in hugo - so it is external in that sense. Who runs the script and when? It is run once and not at every Hugo build. So this team html page is like a png (sort of) for hugo. What am I missing? |
Ah I see the confusion. The last part of the script does use jinja2, I missed that. It does very little though. For each gallery all it does is a for-loop over the entries in However, we could open a separate issue and leave that for later, since it works right now. I'll comment on the PR, let's continue there. |
* Ref #gh-45 script to generate team gallery html page using team.yml specs * Ref #gh-45 integrate team/html in about page remove team.md not required * Fixes #gh-45 addresses Ralf's review comments * Replaces tabs with space=4 * team.html generated using gallery script is stored in static folder, referenced from about page in Hugo * Replaced tabs with space tabstop=4 * Fixes #gh-45 * improved html layout tab=2 * Some minor readability cleanups to the gallery script * Remove --ignore-errors, that hides errors that need to show up. Also fix some minor things. * UPdate cached GitHub data for NumPy Team Co-authored-by: Ralf Gommers <ralf.gommers@gmail.com>
This is generated from a simple yaml file that then fetches data from GitHub and creates a gallery: https://datapythonista.github.io/pandas-web/community/team.html
Code at https://github.com/datapythonista/pandas-web/blob/master/pysuerga.py#L43
Thanks for the suggestion, @datapythonista!
The text was updated successfully, but these errors were encountered: