[fab] New project formation is out of control

[Rahul Sundaram]

On Thu, 2006-06-22 at 14:00 -0500, David Eisenstein wrote:
> Patrick W. Barnes wrote:
> > On Wednesday 07 June 2006 14:32, Greg DeKoenigsberg <gdk at redhat.com> wrote:
> > 
> >>Comments:
> >>
> >>1. I continue to assert that there are only two meaningful kinds of
> >>project: incubators/SIGs/whatever, and full-fledged projects.  Why
> >>distinguish between "ideas" and "incubator projects"?  Why does someone
> >>need "a plan of action" to "graduate" to incubator status?  What's wrong
> >>with having 2000 incubator projects, 1500 of which overlap?
> > 
> > 
> > A summary and a plan of action really aren't much to ask for, and the biggest 
> > reason for doing so is to avoid having to create and maintain 2000 wiki pages 
> > and 2000 new mailing lists that might not do anything.  We've had a number of 
> > people throw out an idea, give it no further thought and start requesting 
> > mailing lists and assorted other resources or privileges, and that can put a 
> > strain on the people who can provide those things.  If someone can't 
> > formulate a basic summary and plan, why should the rest of us put forth the 
> > time and effort to give them resources?
> 
> If the Fedora Infrastructure team is becoming strained by having to provide
> these things to a lot of proposed projects that may never fly, then it
> sounds to me like the infrastructure team needs to grow in numbers and
> perhaps also in resources -- to have more folks on it who can provide such
> things, or that Fedora Infrastructure needs more machines or something to
> provide "scratch space" on.  Perhaps Fedora Project can host their own
> mailing lists and Fedora people can help maintain them?

What exactly is preventing Fedora people (even outside of Red Hat)from
joining the infrastructure team or helping out in mailing list
administration now?

>   As for maintaining
> 2,000 wiki pages -- well, we *all* who are part of the Fedora Project pitch
> in on the maintenance of those...  especially folks like you and Rahul.
> Maybe we need to create a small team who can monitor these things a bit and
> put "dead" pages up into some kind of wiki "attic"?

Better yet, tell us which pages need work and we will do that but the
point being made is that if there are multiple projects being spawned
that duplicate from our each other, then we are losing contributors
momentum instead of gaining it. More machines wont fix it. Good
guidelines would.


> 
> I am a firm believer in the process of foment, and it best happens
> when dampers are removed.  It can be a bit chaotic, and sure, there will be
> duds, but it's the process and the communications that are important here,
> not just the end product.  The process includes encouragement, praise,
> open-mindedness, and tolerance for things that may not work out splendidly.

We can have chaos. Thats not a huge issue as long as new users and
contributors have a way to differentiate between new experimental
projects vs the established ones. The guidelines shouldnt be
bureaucratic and barrier to entry should be kept low and that's a
balance that we can work out through peer review. 

> 
> On the other hand, you have a point, Patrick.  Having some kind of summary
> and objectives available before allocating serious resources makes sense.
> It doesn't have to stop there, though -- maybe this is another instance of
> where good mentoring can come in.  If someone has an idea, but doesn't know
> how to make a plan or a summary, well I am sure that there are dozens of
> eager folks on http://fedoraproject.org/wiki/Mentors who can help folks come
> up with what is needed to get a new idea far enough for resources and time
> to be allocated.

Rahul