Guy Chapman aka JzG wrote: In some corners of academia, particularly computer science, this is sometimes seen as a positive phenomenon. Papers freely available on the internet are cited more frequently (controlling for other factors) than papers that are only available in print or from pay archives [1]. This provides a nice bit of pressure for authors to put their papers on their homepages, and for journals to make their archives publicly accessible. That pressure has been successful in a number of cases (e.g. the editorial board of the journal _Machine Learning_ resigning en masse to support an open-access alternative [2], which pressured _Machine Learning_ itself into making many of its own archives freely available online). The result is more information available to everyone, not only the wealthy or those who have affiliations with wealthy first-world universities---a goal that seems rather in keeping with Wikipedia's spirit. -Mark [1] S. Lawrence. Online or invisible? _Nature_ 411, 2001. Online version: http://citeseer.ist.psu.edu/online-nature01/ [2] Editorial Board of the Kluwer Journal, Machine Learning: Resignation Letter. _SIGIR Forum_ 35(2), 2001. Online version: http://www.acm.org/sigs/sigir/forum/F2001/sigirFall01Letters.html

Ken Arromdee wrote: I support "Ignore All Rules", but I don't see that as being applicable here. It looks too much like the easy way out. One does not deal wih a fundamental problem by ignoring it. I agree with the importance of sources, but if we truly want this to be a project for the world we need to avoid imposing western Aristotelianism in places where it was not the cultural norm. That would just be another form of imperialism that we made ourselves believe was gone when Europe's former colonies became independant. We do Wikipedia a disservice when we insist on a rigorous interpretation of reliability. Sources allow us to trace the origins of concepts that are expressed in an article; having them implies nothing about the validity of the concept. There is certainly no necessity that a source be on the internet, and some benefit might be derived from insisting on at least one source that is not from the internet. There is certainly no necessity that a source even be in English (or whatever other language is used in the Wikipedia inquestion); somebody who understands the source language can always translate it later. Ec