incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nicola Ken Barozzi <>
Subject Re: [PROPOSAL] Apache Incubator Sandbox
Date Tue, 14 Jan 2003 13:05:13 GMT

Stefano Mazzocchi wrote:
> Nicola Ken Barozzi wrote:
>> Many projects have a sandbox where developers can try new things 
>> related to the project. But developers still want to have a place 
>> where to experiment new things, not necessarily tied with the scope of 
>> a project they're in.

> I tend to think that such a sandbox will only increase the probability 
> of the sourceforge effect and won't stop projects from extending their 
> scopes and grow subprojects inside.

Could you please expand on what you mean with the "Sourceforge effect"?

As for it not being effective per-se to reduce scope bleeding, it's 
true. It can *help* reduce it, but just the fact that it's there will 
not make it happen automatically. But if there is a clear policy that 
says that out-of-scope projects can be started in a certain place can 
surely help in reducing scope bleed.

> But I could well be wrong as I've been several times in the past, even 
> on this matter.

It all boils down to policy of course. The Japanese tend to think that 
anything can be done, given that it's done to certain rules, and in a 
certain time.

Where should Apache Committers try out new code, new projects? For 
Jakarta it's Jakarta Commons Sandbox. It works. Why not extend it to all 
Apache? Or should we prohibit new experiments at Apache by directing any 
effort in Sourceforge?

To be more specific: I started Centipede in Sourceforge.
Maven developers started Maven in Turbine.

What should the policy be?

I tend to think that Apache committers should be given a place where to 
experiment if they wish to.

Nicola Ken Barozzi         
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)

View raw message