[CentOS-devel] New centos developers 'invitation' -- was: Distribution update and centosplus

Fri Sep 8 03:22:47 UTC 2006
R P Herrold <herrold at owlriver.com>

On Fri, 1 Sep 2006, David Hrbáč wrote:

> Jim Perrin napsal(a):
>> The CentOS team is quite small and we're working on this for free.
>> Donations of cash, hardware, and talent are welcome to help improve
>> the project.

> ... There's no rule how to became the CentOs developer, as 
> far I have not seen any invitation in lists even. I'm trying 
> to improve/push Centos with every my post.

I am a bit late into the discussion, but would shed a bit of 
light on rules and invitations.

There are many ways that centos is 'developed' (as a rebuild 
effort, some but not all of the distribution integration 
effort, is pre-ordained -- base and updates package selections 
vary from the upstream by and large only as to trademark 
elidement and insertion, and updater decisions.)  Other places 
allow for creativity.  A guiding principle is: be a strict and 
reliabile rebuild, timely built, striving for stability first, 
and added features later, if at all.

Some roles lend themselves to delegation: communication 
efforts (mailing list, IRC, forum, wiki, advocacy), testing 
(candidates in the testing archive, bugtracker confirmation 
and traige), and so forth.  Other roles do not - rolling 
releases and 'official updates' signed with an official centos 
key, infrastructure such as mirror list maintenance or 
security matters, web editorial.

The project's needs are assessed, and available 'talent' and 
prominent contributors to the project are noted, discussed, 
and evaluated on a continuing and continuous basis, by the 
centos core team 'behind the scenes.' A few months ago, as a 
result of that approach, Jim Perrin was invited to additional 
public responsibilities for the project; more recently the 
planned EOL and transition of Tao into CentOS, and the 
addition of David Parsley occured.  Different people, 
different skills, but both now part of the core team. 
Friendly collaborations with other FOSS projects are part of 
Centos' history and a source of its strength.

Each of the core team are active in the project, have their 
eyes open, and know who the 'prospects' are, from direct 
interaction or published reputation.

So, there _is_ no one 'rule' which one might meet to then 
demand admission; rather the project is a self-perpetuating 
meritocracy of like-minded people _invited_ and choosing to 
volunteer increasing amounts of their time for the betterment 
of the project, at increasing levels of responsibility, when 
they have demonstrated a competence relevant to a given role.

Want to be a part? you're invited.  The quickest routes I can 
think of to be invited to assume _more_ responsibility for the 
project is to _do_ more; test, participate, be thoughtful and 
present and accurate.

The 'entry gateways' are the self-serve account signup routes; 
Write reproducable case bug reports when you see a flaw and 
propose the fix; weed and confirm existing reports; document 
in the forum or the wiki usage cases and solutions for the 
topics which recur over and over again in the mailing lists or 
in the IRC channels; 'staff' the mailing list and IRC with 
good counsel to explain and instruct to answer a tough 
question.  File reports and push fixes in upstream, not just 
up one level, but all the way back up to the underlying wells 
of projects from which the FOSS community drinks.  We'll 
notice.

-- Russ Herrold

ps - Think you have what it takes? Want to do more, but you 
are shy? If anyone reading this wants to test their mettle out 
of the glare of a public participation, but needs a project 
suggestion or facilitating resources for a project on behalf 
of centos, contact me off list and we'll come to a 
specification together to mentor you on, at a complexity level 
suited to your skillset. I have an ample backlog of wishlist 
tasks on behalf of centos. Some participants have picked up on 
suggestions I have described from time to time in the IRC 
channel from time to time, and some really nice work _and_ 
some personal skillset growth in the implementor has resulted 
from it.

- r