[openstack-dev] [elections] TC candidacy

Ben Swartzlander ben at swartzlander.org
Sat Oct 1 03:10:53 UTC 2016

I'd like to throw my hat in the ring for the TC election.

My name is Ben Swartzlander (bswartz on IRC) and I've been PTL for the 
Manila project for the entire life of the project. I'm also relatively 
active within the Cinder project, and I've been part of the OpenStack 
community since Essex.

My reasons for running for TC are fairly simple. There are some changes 
I'd like to see and I think that I'll have more ability to effect change 
if I'm part of the TC.

The first thing I'd like to change is that I'd like to see OpenStack 
start acting more mature. It *is* fairly mature now but in many ways we 
still have the habits of a shiny new project. I would like to see way 
less time spent on new features and much more time spent on stability 
and quality improvements.

Specifically I'd like to see dramatically more automated testing of 
stuff we already have. Not gate tests that run for every checkin but 
serious automated nightly regression style tests that actually cover 
realistic use cases and take several hours to run.

I would like to see more frequent releases. I hold up the Linux (kernel) 
project as an example to emulate. Linux releases a new major release 
about every 2 months. OpenStack has held to a 6 month release cycle for 
it's whole life but I think we can and should move to shorter cycles. In 
a similar vein I think serious effort needs to be spend on LTS (long 
term support) -- specifically the ability to upgrade across multiple 
releases without anything breaking. The deprecation policy needs to 
change if we want to get this right.

I would like to see fewer new projects and more focus on existing 
projects and possible integration between them. Long ago it was decided 
that OpenStack should be a loose federation of related projects but many 
feel that OpenStack should be a unified product. This has created a 
cognitive dissonance that pervades nearly every discussion I have about 
architectural decisions within OpenStack and I feel the TC owes this 
topic more consideration. If we decide we're all working on a single 
product then we need to change the way we act. If we affirm that we 
really are just working on a bunch of loosely related things then we 
need to disband working groups/and cross-cutting projects that are 
trying to push for uniformity.

Lastly I feel strongly about community and the Python language. I am not 
a language zealot but I know from experience that adding more 
programming languages to an existing project is ALWAYS WRONG and I will 
fight any proposal to add more programming languages to OpenStack.

I don't expect everyone to agree with my ideas but if enough of you do, 
vote me onto the TC and I'll do my best to gradually change things for 
the better.

-Ben Swartzlander

More information about the OpenStack-dev mailing list