[IPython-dev] Should I still contribute to IPython ?
Matthias BUSSONNIER
bussonniermatthias at gmail.com
Wed Dec 19 03:12:40 EST 2012
Le 19 déc. 2012 à 00:51, Nicholas Bollweg a écrit :
> open question: where can automation help?
>
> what is causing the bulk of time waste in doing community management to really dig into what a particular issue or PR is trying to accomplish? perhaps a cloud asset, along the lines of Travis-ci, that allowed single click access to a running kernel, web notebook or other aspect of the system?
I think the problem is you don't really know who is responsible for which PR/issue, so you lost time wondering wether you have to be involved,
or if you are in a good place to review. I have no problem getting the code to run. For me it's a one line
$ git checkout origin/pull/2463 ; ipython notebook
Sometime 2 if I need to do a git fetch before and adding /head or /merge at the end is enough to know which state of the PR I want to test.
The longer part is to read the PR, unpacked the new UV laser, and wondering what the other dev thought about that are.
I miss a small "involvement indicator" on PR that say "involved"/"neutral"/"don't care" on the side to know which PR I should
take care of.
Also the "who agree for merging" is a little confusing sometime.
I would like a "ok to merge" button, that you just click and you get the count of +1 vote that goes back to 0 if a new commit
is pushed.
I think this would highly improved the merging /review rate with :
* This pr has 3 "merge ok", I just have to read add mine and I can merge.
* This PR has already 2 dev with are "involved" I can worry of others.
* This one has nobody I'll look after it.
--
Matthias
>
> while GH is great, and is cited as one of the factors leading to involvement blowing up, would a more workflow-enabled system like jira or rt offer better triage? with the API, Github could still be an entry point into the ecosystem.
>
> too much success seems like a happy problem... I hope everything works out!
>
> On Dec 18, 2012 10:39 AM, "Thomas Kluyver" <takowl at gmail.com> wrote:
> On 17 December 2012 20:00, Brian Granger <ellisonbg at gmail.com> wrote:
> * How do we manage communication? Verbal communication is much more
> efficient than emails or even IRC. The 4 people at Berkeley will have
> an incredible advantage in being able to talk daily. We don't want to
> cripple or remove that advantage, but we need to figure out how to
> include other core devs and people from the community. This is
> particularly relevant to myself as I am the only person involved in
> the Sloan work that is not at Berkeley.
>
> I think this issue - communication - is becoming key. I spent a day away from my computer, and came back to >40 new e-mails in my IPython folder (in addition to the 20 odd unread that I'm planning to get round to one day). That's a mixture of our two mailing lists and the Github notification messages. I get the feeling we're approaching a critical point, where it's no longer possible for us as individuals to keep up with all the discussions going on.
>
> To my mind, we need to split things up. We already have an informal separation of interests - for instance, I leave most of the discussion about the notebook front-end to others, but get more involved with IPython.core. I think we need to make this a bit (not too much) more formal, so that no-one falls down the cracks as everything speeds up.
>
> This could mean, e.g. more specialised mailing lists, or having a consistent process for handling incoming issues.
>
> Best wishes,
> Thomas
>
>
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
>
> _______________________________________________
> IPython-dev mailing list
> IPython-dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/ipython-dev
More information about the IPython-dev
mailing list