Erik Abele wrote: ...
...but the last part about restricting mirror usage seems a bit strong to me. IMO the 'Incubatee' should be encouraged and trained to use our mirroring system appropriately.
In fact this document, coupled with "at least a couple of releases to exit", is at least wierd, as Ted has duly noted.
If the release is clearly marked as being incubated and it's a public release (for a definition see apache.org/dev/mirrors.html), I don't see any issues in permitting the podling to use the mirrors.
The only downside of which I can think of at the moment is if incubation fails and the project vanishes (?, at least from the ASF),
the mirrored releases would be kept on archives.apache.org. We'd then
have to manually remove them or perhaps we could just keep them there
as a part of our history.
If they can release as incubated projects, I'd say that keeping them as history is quite logical.
hmm, I'm wondering if a apache.org/dist/incubator/ directory would do any good here?
After some thought, I now tend to favor this and agree with what you say. +1
BTW, I still believe that keeping all incubated projects under the incubator, with @incubator ML and incubator-cvs is the most correct and consistent thing to do, especially now that with this setting we would have [EMAIL PROTECTED] releasing as incubator-lenya... hmmm
--
Nicola Ken Barozzi [EMAIL PROTECTED]
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
------------------------------------------------------------------------------------------------------------------------------------------ To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
