[PKG-Openstack-devel] Accepted python-eventlet 0.20.0-1 (source all) into unstable

Thomas Goirand zigo at debian.org
Mon Nov 13 09:59:35 UTC 2017


On 11/13/2017 10:04 AM, Daniel Baumann wrote:
> On 11/12/17 02:11, Thomas Goirand wrote:
>> It turns out that the classification is, in my opinion, wrong in many
>> ways.
> 
> great - then adjust that by *discussing* it.
> 
>> If you look at the case of Eventlet, it's used only in OpenStack. And
>> not just only once, but 38 times (in 38 packages both as runtime and
>> build dependency). So to me, it really doesn't fit the bill of being
>> "general purpose", and OpenStack is too dependent on it. Best is if it
>> can follow the OpenStack release too.
> 
> the idea of the classification is that it has no ambiguity and is
> therefore simple, easy and clear.
> 
> the original "line" was drawn at "is openstack.org upstream, yes or
> no?", which is a binary question, so easy to answer (no?).

I generally agree, though in a few cases, the line is blured. I'm
thinking here about things like python-tenacity & friends, built by the
telemetry team.

Cheers,

Thomas



More information about the Openstack-devel mailing list