[compsciphd]

As someone who published multiple papers during his academic days under auspices of ACM, USENIX and IEEE, I never saw a case where they demanded that authors not host a copy of the paper for themselves for their "own" audience on their own website.

Furthermore, the concept of a "public domain" paper in academia just seems weird to me. the concept of "public domain" means that the contents can be reused in whole or in part without attribution of any sorts to the original author(s). that goes against the ethos of academia (i.e. plagiarism) in regards to authoring papers, so unsure what public domain for the actual paper gives users vs. the document being what I'd refer to as "freely available" (i.e. no one else can charge for access to the document, only the 'copyright' holder can).

If the author has a right to freely distribute the document (and anyone who gets the document from the author maintains the same right), I don't see what public domain "assignment" gains anyone. i.e. copyright assignment (to the publisher) with the ability to freely distribute the paper accomplishes all these goals. The only thing (I can imagine) that it doesn't accomplish is giving others the ability to collect a bunch of papers together and sell it for "profit". But that doesn't seem to be a something DJB views as needed (and in fact, rails against the publishers who are requesting the copyright assignment for that very purpose).