OK I wasn't aware of this document, although I still say the first proposing of the problem publicly was Sagan's article, and this one is people trying to recollect a lunch time conversation years after the fact.

So I feel it is really Sagan's summation of the problem that is the original. So I would say maybe they said the only way it could work was if there was FTL or some similar formulation, but I think Sagan also showed that it could work in other ways.

So I guess we just sort of disagree on this.