Re: Re: [PEPr] +1 for RFC::VersionNaming

Lukas Smith wrote:

Laurent Laville wrote:

Greg Beaver wrote:

This could throw a wrench into the idea of creating an xml-rpc-less channel server, as the user would need to know both the version and the stability. I think adding a postfix with state is going to be a bad idea whether it is automatic or not.

I'm +1 on this last point of view. I don't see what advantages it should give. If someone can tell me these advantages, i'll be glad to heard it. We never did it in past, and <state> in package.xml was enough to mark a release as beta, alpha or dev.

the advantage is to make it more transparent to users what kind of stuff they are installing. its hidden a way a bit during the install process .. alot of people end up setting preferred state to < stable since we are also not very good about moving our packages to = stable in a timely fashion. that was why I proposed adding the state in a more visible location. then again maybe we should just make sure people know about the "pear install [PackageName]-[state]" syntax .. or maybe we should just kill the preferred state all together?

People should know about [PackageName]-[state]. I put this in the FAQ not too long ago. preferred_state is absolutely necessary for upgrade-all. Greg