Closed bryevdv closed 11 months ago
noting I plan to delete bokeh/badges
in the very near future as well -- the conda download badge generation does not work anymore and SO badge generation is no longer needed (shields.io picked it up)
Also asked in Slack but maybe here is better:
Does anyone know why we have a fork of distributed in the Bokeh org? https://github.com/bokeh/distributed It does not seem to be different from the main distributed repo in any way or therefore serve any purpose. I would like to propose deleting it
(It's entirely possible I forked it into bokeh
by mistake when I needed to fork into my personal awhile back, but I don't actually know. Want to see if anyone else has any other explanation before acting)
It also only just occurs to me that bokeh/bokeh-notebooks
has been superseded by bokeh/tutorial
and could at least be archived with a pointer to the new repo?
noting I plan to delete
bokeh/badges
in the very near future as well -- the conda download badge generation does not work anymore and SO badge generation is no longer needed (shields.io picked it up)
I know how to fix the error generating the conda badge if you would like to keep that part of the process?
if you would like to keep that part of the process?
To be honest, not really. It was unreliable even when it did "work" owing to uncertainty when the data was published. But more importantly, GH has taken a stance against being everyone's cron job. GH actions are disabled in any repo that does not see activity for two months (I had to re-enable manually in settings yesterday). It's just a badge, not that important. If we need numbers for grant applications, etc. we can derive them at the time.
I'm +1 for archiving bokeh/bokeh-notebooks
and deleting bokeh/distributed
(also don't know why this is here, and I could've added it accidentally as well)
I do feel strongly about not deleting the other repos -- I think having the project's history for future reference is helpful, and if we delete the repos we'll lose data on all the past contributors to these repos (similarly, contributors also lose attribution). So, I'd vote for creating a bokeh-attic
org to store these repos.
Sidenote: Sorry for closing+reopen+edits, had a mouse glitch!
@bokeh/core I'll plan on taking these actions by the end of the week unless there's a discussion opened here:
bokeh/bokeh-notebooks
bokeh/distributed
bokeh/blog.bokeh.org
bokeh/bokehplots.com
I'll make a new issues about moving archived repos to a new bokeh-attic
org or similar at some other time in the future, no real hurry.
We currently have a half-dozen archived repos:
Can some of these be deleted? I'd like newcomers to the Bokeh org to be confronted with as little irrelevant and useless informatio as possible. Offhand I personally think these could definitely de deleted:
blog.bokeh.org
(useless, all material copied over to medium at this point)bokehplots.com
(useless, nothing relevant or useful for most of a decade here)bkcharts
(completely irrelevant today)It seems like there may be others archived soon (e.g. RBokeh and possibly some vendored dependencies). Would it be reasonable to make a
bokeh-attic
GH Org to move old archived repositories to? That would include vendored dependencies archived more than two minor releases ago. since we would never expect to make any more point releases on those branches ever. In the highly, extremely unlikely event that we did need them, the repos would technically still be accessible with some effort.cc @bokeh/core