On Wed, 28 Jan 2009, Ben Armstrong wrote:
or else eliminate the metapackage divisions entirely. (But in that case, yes, solving the workflow problem for tag acceptance would become more important.)
IMHO metapackages are a simple, established and widely accepted technique which we can easily relay on. I would prefer testing other things in parallel and see how they might work. As long as we do not have something that has important advantages over metapackages while having no important drawback I'd like stick to established things. Or did I missunderstood your mail? Kind regards Andreas. -- http://fam-tille.de