This is a reply in part to Leo's email and to the comments from Sefano on the same subject:
You may be wondering why I issue a veto against the modifications committed by Peter Donald to the Phoenix code base. I would line to take this opportunity to express those reasons in more detail, and invite Peter Donald or any other member of the Avalon community to present counter arguments.
1. I am opposed to the introduction of an external dependency within Avalon code when a perfectly valid solution exists within ASF and published under the ASL.
2. I am opposed to the modification of the Phoenix code base when the modification is directly in contravention of an established concensus to migrate to Commons where possible (and in particular in this case the Commons CLI application).
Peter Donald is free to respond to the veto as he sees fit. Should Peter Donald ignore the veto then I would request action from the Avalon PMC to address the matter should that situation arise. I think it is appropriate the Peter Donald revert the changes that he made following which he is welcome to make a proposal to the community concerning the maintenance of the Excalibur CLI API (or derived work). This difference here is that it is the responsibility of Peter Donald to make that proposal and present a sufficiently compelling case to this community supporting his view. Should Peter Donald prepare such a proposal, I remain hopeful that he will address the specific issues I've raised above. I do think that the action taken by Peter Donald with respect to the modifications to Phoenix without any prior community consultation are inappropriate but I recognize that these actions are within the scope of the privileges granted to him by the Avalon PMC within the established policies and procedures. As such I feel obliged to address this issue via the veto mechanism - a mechanism available to me in accordance with the same policy and procedures.
Cheers, Steve.
Leo Simons wrote:
Steve, vetoes are _not_ how we want to work things out. This is *not* productive (see my earlier mail about my view of productivity). Please reconsider throwing more annoying vetoes into the party about stuff which has not been aptly discussed.
Here's how development works: we interpret were we're moving as a project and what we need as a project, and act based on that. If we're not sure everyone agrees with those acts, we draft a proposal before committing stuff. When we are, we just commit, and things are subject to lazy consensus.
If we shouldn't have been sure because there isn't in fact a consensus, the thing to do is to bring the actions and commits up for discussion. *Not* just throw a veto out without discussion. Vetoes make it more difficult to build consensus, and consensus is what we like.
Why don't you retract your veto, take a look at Eung-ju's patch, rework it to be fully compliant with the existing phoenix cli options, and propose the modified patch as a replacement. Everybody happy.
vetoes -- bleh.
- Leo
Stephen McConnell wrote:
-1 (veto)
donaldp 2003/03/12 05:01:38
Modified: . build.xml
src/java/org/apache/avalon/phoenix/frontends CLISetup.java
Removed: lib/container excalibur-cli-1.0.jar
Log:
Remove away from excalibur-cli which has been axed.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--
Stephen J. McConnell mailto:[EMAIL PROTECTED] http://www.osm.net
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
