Instead, I translated the most important papers into English. In a conference paper I'm writing right now I cite quite a few of these papers and I've tried to be extremely specific in their citations, going so far as to include OCLC numbers that I used for interlibrary loan. (Librarians will know what to do with this information.)
What's easily found today may not be so easily found in the future. Today, I find that many researchers can't find a paper that's not available online but is available in any decent academic library.
It's also possible that a hard to find paper is cited wrong or cited ambiguously, and having the correct, unambiguous citation is all one needs to find the paper.
To address papers which are practically impossible to track down, as gms7777 suggested, at least people can contact the author of the paper citing the hard to find paper in this case. (For the Hinze paper, as I recall Shell had a NDA as a condition of downloading the paper from their system, so I can't share that one.)
From a practical sense, it gives readers a way to find it for themselves -- contact the author that cited it.
That makes good sense.
> From a practical sense, it gives readers a way to find it for themselves -- contact the author that cited it.
Right, but in the case of a paper this old the author is most likely dead or extremely hard to reach. Maybe a standard part of any paper citing other papers should be a reference to an archived copy of the papers cited?