Well, I step into this thread because I _really_ think it's complicating some (maybe) simple thing.
There can be many distinct actions: 1. search/browse CPAN for a module. 2. use (or require :) a module. 2.1 use it because I'm in a hurry or don't like to rewrite code. 2.2 use it because I don't know how to do without. 3. evaluate a module. 3.1 I test it myself. 3.2 I look for other's experience. 4. download a module. 4.1 modem, browse cpan.org, modules_by_category (Hi Fergal, why not some fast_CPAN_by_category without authors folders and ancient versions? Of course excelent readme files are _essential_ when browsing. PS: I'm this one!) 4.2 fast conection and CPAN tool. 5. develop a module, or collaborate on development. 6. keep my program and modules it use updated in my environment. 7. etc, etc, etc... I believe we should, before going on discussion about rating system and CPAN, clear the following points: 1. Is it possible/good/desireable to change CPAN engine or create/improve a excelent/visible/unique rating system, should this thread take us there? 2. If so, is it good idea start besides Perl6 (I suppose that a lot of module-writers will have to upload new versions, so perhaps Perl6 is _the_ suitable synchronizing point). 3. Finally, if so, well, we'd better to clear all of that distinct actions into a kind of taxonomy of purposes. Both browsing (the comprehensive universe) and searching (for one specific topic) are essential. It seems like none is good enough by now. Silvio
