Hi everyone, just to spend my two cts.. I don't like the idea of accepting the restrictions of an on-disc naming scheme as restrictions for a package tree. I really think it's not, well, state of the art?! I know the reasons for doing so, because of easy configuration, manual search, understanding of the "user" and so on. But I find the restrictions to hard. Wouldn't it be better to just give packages a unique id? One could then arrange the packages as one wishes, by category, size, liking what ever... The disadvantage would be one had to use a search function to find a package. (But why is that a disadvantage?) Why is it a problem to say: Well, we have more than 10000(?) packages in several versions and don't think you have to find your way by hand.
We shouldn't be that, say, technology agnostic not to use a search engine for coping with the package tree. To make it short: + unique naming opens a new dimension of package management - you need (special) software to handle the tree Obfuscating the tree with randomly chosen letters would have the same effect, though. Michael Raitza _______________________________________________ Exherbo-dev mailing list [email protected] http://lists.exherbo.org/mailman/listinfo/exherbo-dev
