I would suggest that Zotero ignore such attempts by academic publishers to further limit access to the work of researchers that those publishers did not pay for.

@adamsmith and @tim820 - Good points about the potential conflicts of interests with respect to publishers and their tools. I hadn't considered that, but it is consistent with the list of approved SCNs (and Zotero's absence from the list).

@adamsmith to be clear, 'howcanIshareit.com' covers both sharing your own work and other's work, and it covers SCNs as well as other mechanisms of sharing, broken out by type (author's copy, accepted copy, copy of record).

The focus of my question isn't 'what behaviors can publishers practically or legitimately prevent' but 'what behaviors meet expectations of copyright law'?

To me and many of my researcher colleagues, the issues regarding copyright law in their written work are both legal and ethical. How did we end up in the situation where researchers give up their copyright to publishers for basically nothing, and those publishers then sell it back to researchers' universities for millions of dollars ? And those publishers prevent the taxpayers who often funded the research from their taxes (and also fund many of those institutions) from accessing that same published work (as well as for example doctors in poorer countries from accessing the latest published medical research - a deliberately emotive example). If you think that whole situation is unethical, then that will shape your position on publisher copyright, and your attitude to initiatives like these front websites for publishers.

On the legal side, the situation is perhaps not always as clear as assumed. For example, in the current sci-hub case in the Indian courts, sci-hub's lawyers argued that publisher's copyright transfer contracts with researchers should be deemed invalid, because the researchers did not get paid for their work, directly or in the form of royalties (ie the way in which the law assumes that copyright transfer works). Of course researchers do not on the whole want to be paid for the papers they write (they still get their salaries either way); but nor do they want publishers to profit to the tune of billions of dollars by restricting access to their work. In any case, that motion failed only on a technicality, so was never tested on its actual legal merits.

If you read the text of "What are Scholarly Collaboration Networks (SCNs)" at the howcanIshareit site you could be forgiven for thinking that they are a warm and fuzzy researcher-centered notion about how those researchers can share the work they authored with other like-minded researchers. But then you see how particular platforms are listed by name as either offering 'networking functionalities', or not. Then a more cynical view might instead describe SCNs as a publisher-invented euphemism for platforms that their corporate lawyers should monitor closely for potential copyright breaches.

I suggest again that Zotero should have nothing to do with publishers' transparent front operations designed to strengthen the assertion of their questionable copyright.