Emmanuel Lecharny wrote:
The problem with a release injected in maven is that it will be there forever. If a release has some problems (IP issues, etc), you can't remove it from maven, as some projects might depend on it, and the users will immediately carpet bomb the maven ML to get the release back into the repo. Sounds like a possible scenario, no ?
Ok; we are finally on to troubleshooting specific cases, woo hoo! W.R.T. Maven, if informed of an IP Violation, the jar will be dropped out of Maven, period. Although Maven is treated as a common-carrier, mirror-like entity, that won't absolve us of reacting to a specific complaint. So incubator is no different than any project, internal or external. Right? If it is a cosmetic defect like you cite, I don't recall that we have withdrawn any prior releases for a better, new release. Either it was a release or it wasn't, irrespective of code and packaging quality. Right? So if they live in Maven forever, who cares? c.f. http://archive.apache.org/dist/incubator/ Nobody has even come close to suggesting why releases @maven, a mirror like common-carrier, can be restricted. If the ASF permits any arbitrary third party of shipping their open source release through Maven, then we are in legally troubling waters if we prohibit certain content based on the sorts of issues folks keep bringing up. We don't *WANT* the Maven project or foundation to own the entanglements of "supervising" the Maven repository. That's up to the folks who submit the content for mirroring. (And in this case, incubator artifacts, that would be us). And nobody has convinced me why any podling committer cannot take the properly released Incubator podling artifact, post it to Maven, any more than they are encumbered from posting it on their personal homepage. What part of the AL is not clear, here? Bill --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
