[zeromq-dev] Nanomsg postmortem

[Pieter Hintjens]

Waiting for CI is an option for any maintainer. There are clear cases
where it's not helpful, e.g. as people add new CI targets that don't
work consistently. The skillset needed to fix CI is different from
that needed to make patches, and it's important that we don't create
dependencies between people that slow down our work.


On Mon, Feb 8, 2016 at 1:38 PM, Osiris Pedroso <opedroso at gmail.com> wrote:
> Very interesting read.
>
> It confirms Pieter's view that every pull request should be merged, no
> matter the direction it takes us.
> This adds/keep a contributor interest in the project, which ultimately keeps
> the project alive in the long run.
>
> But I still think the minimum bar should be that a PR should only be merged
> if it builds for all target platforms that we have CIs for.
> Otherwise the PR is not being constructive to the overall project, if it
> will be creating more work for others.
>
> PS: I found this article that shows how to setup Travis-CI on my branch. I
> will try it and see how it goes.
>
> On Mon, Feb 8, 2016 at 6:03 AM Pieter Hintjens <ph at imatix.com> wrote:
>>
>> Hi all,
>>
>> This might be of interest:
>> http://sealedabstract.com/rants/nanomsg-postmortem-and-other-stories/
>>
>> -Pieter
>> _______________________________________________
>> zeromq-dev mailing list
>> zeromq-dev at lists.zeromq.org
>> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>
>
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>