* social experiments are dangerous, you risk pissing off end users and/or contributors and eventually reduce your user/contributor base (people vote against evil social experiments with their feet: "el último que se vaya que apague la luz...") * social experiments are good if there is a good reason (technical or other), so it is explained and everything is OK i.e. the change from kernel-source-n.n-n.arch.rpm -> kernel-sourcecode-n.n-n.noarch.rpm ** kernel-sourcecode (or whatever you call it) _is_ necessary to build drivers from other developers or vendors (madwifi, hostapp, intel, etc...) and it is explicitly required so in documentation and howtos; thus, this social experiment is _not_ good, unless we pretend to bury Linux in its own 'ghetto' making _more_ difficult for third parties to contribute and reducing the range of compatible devices... ** size is not an argument against because there are _enoumous_ 'debuginfo' packages and they are rebuilt and posted regularly. ** "you can do it yourself" is neither because with this same argument taken to the limit no distribution is necessary. * _this_ social experiment has been previously discussed so why it comes again? is Fedora open and community driven? or is Fedora a RedHat social experiment? -- Manuel Moreno <manolo@xxxxxxxxxxxxxx>
