there seems to be great 'sharing potential' across this community for materials others may not be in position to acquire and thus great practical interest for mathematicians world-wide.
However, I don't think Math Overflow should be known as the place with this 'sharing potential'. First, this would fit other forums better. Second, typical institutional library's Terms of Use say something about not sharing the pdfs, so the question would be asking people to break their library's rules even when no copyright law is at stake. Third, yes, copyright law.
]]>(1) User A: My school doesn't subscribe to The Rocky Mountain Journal of Knot Theory. Could someone scan in Smith's recent article and send it to me?
User B: Here you go.
(2) User A: Does anyone have a reference for the connection between the Jones polynomial and untangling climbing ropes?
User B: Smith explained this very well in The Rocky Mountain Journal of Knot Theory. See this link.
I don't like scenario (1) for two reasons. The first is that it feels sketchy to me, although I understand that it is controversial whether or not it is illegal for us to host it. The second is that it is of very little benefit to the rest of us reading it.
I think that we should ignore scenario (2), even if the link is to a copyright violation.
I should probably acknowledge here that I am not a moderator, so this is just one guy's opinion, not official math.MO policy.
]]>I would advise emailing the authors.
]]>