[users] The rpmrepo problem (was Audacity vs. wxGTK 2.8)

RpmForge4Bill RpmForge4Bill at triad.rr.com
Sat Oct 4 16:59:23 CEST 2008


I thought I would offer some suggestions about the issue Dag faces 
regarding the continuation of the rpmrepo effort. Please pardon the odd 
snipping/marking below. When I did the copy from the original post, 
T'bird decided to be inconsistent about the quoting.

First, I tend to get (rather easily) long-winded about things like this. 
Keeping that in mind, I will try to post only the briefest talking 
points, rather than my usual in-depth consideration of all possible 
scenarios. This may engender a ... "healthy" discussion, pro and con.

My comments are based on many years professional experience, about 10 
years being involved in projects on the net (both as user and 
contributing participant) and a (rather jaundiced) view of the 'net 
denizens.

As with Dag, I once left a large company and went on my own. I think I 
have some understanding of his situation.

On Fri, 26 Sep 2008, Bogdan Costescu wrote:

 > On Fri, 26 Sep 2008, Dag Wieers wrote:
 ><snip>

========================== Dag spaketh ====================
The problem is that RPMforge does not scale very well. We are a bunch of 
guys, there is not central infrastructure. I have no problem helping 
with another project (or changing RPMforge) so that more people can help 
create packages in a way that there is an upgrade path from RPMforge. 
But doing what we have been doing is not viable.

Because it depends on me mostly.

So we need to remove the dependency on me, and that is one of the 
reasons why I don't want to change what we have now (the other reason is 
the time I am spending right now).
===============================================================

The key issue is exactly what you say above. "there is not central 
infrastructure". It is the nature of things that a project that grows 
and survives needs that, for many reasons that can be discussed.

<snip>

 >>  And I don't mind that people complain it is not working, I do mind that
 >>  they are not helping to solve the issue.
 ><snip>

In conjunction with infrastructure, a strong *stated* commitment from 
those who volunteer to work within that infra-structure is needed.

========================== Dag spaketh ====================
I am not saying people have to switch to apt. But it would be one 
solution. The other is helping with the rpmrepo project, or making yum 
more resilient.
===========================================================

MHO: trying to solve personal time-constraints by denigrating another 
projects efforts is a waste of your time.

The true solution you mention is "helping with the rpmrepo project". For 
that to happen, you need more than the occasional (or constant) whine 
about yum, folks not helping, etc. You need a determined "marketing 
effort". What this means to me: public announcements on the list where 
you should have a large body of users that feel what you do is very 
worthwhile that you are in need of committed long-term support of the 
user-base. Host a mail list/irc for this topic specifically. Get the 
initial committed group to help in planning and execution of the 
near-term campaign and development of infrastructure (both material and 
personnel).

 >><snip>

========================== Dag spaketh ====================
Of course. But frankly, I am spending a few hours every day with my 
packages. This will not happen for the next 10 years. I may have 
children, I may have other things outside RPMforge to do.

If nothing happens, that trust will be lost at some other point in time. 
Your trust does not feed me. I have been cutting back on the time I have 
put in the repository when I became self-employed last year. Maybe 
people haven't noticed, but that is the case.

I wanted a solution that did not depend on me for a few years now, but 
nothing was forthcoming and there still does not seem to be a solution 
right now. So where do we go from here ? I don't know.
=================================================================

I addressed exactly these same issues with a net project some years ago. 
The only difference was the principal did not see these issues - he was 
quite myopic about longer-term issues. Fortunately, many folks 
volunteered, made strong commitments by writing their own "job 
description" and committing to fulfill their stated obligations.

I can't state how "key" this is. The public statement of an individual 
for commitment to the group has a profound (my judgment) effect on that 
person and those who might potentially participate with that person. 
This is the "currency" that replaces $$ in a normal job, along with 
various forms of gratification one receives from contributing, feeling 
part of a cohesive dedicated group, etc.

<snip>


The problem you currently face is one of "the human nature". It can only 
be solved by understanding those issues and executing a program that 
addresses those things. Technical acumen, the environment around you, 
problems with other projects are all secondary issues that can and 
*will* be solved when a strong, cohesive, dedicated group of individuals 
operate as a team. It is the unfortunate nature of people that one of 
the things they need most is what they most likely dislike: management. 
The difference in the net environment, as compared to a work 
environment, is the altruistic nature required to make it work. Only a 
small percentage of the populace will have the needed traits, time, 
resources, etc.

That is the problem you face. If you are a "problem solver", you first 
identify the problem. The *symptoms* you experience are caused by the 
problem, not the other way around.

Of course, there are many time-consuming issues to address if you want 
to solve the problem. They are for another time and place. This post is 
only intended to offer one person's take on your situation.

As with others, your efforts are appreciated.

If you choose to address the problem, I would be interested in 
contributing in whatever limited fashion I am able.

-- 
Bill



More information about the users mailing list