In my paper on incorporating X-voting into preference voting by STV[1], without saying so I had treated it as axiomatic that a method of mixing X-voting and preference voting should reduce to either X-voting or preference voting by STV should all the voters be of one sort.
In a comment at the end of my paper, the Editor suggested an alternative formulation which, sadly, would not reduce to X-voting as it is always practised should all the voters be X-voters. The Editor's formulation would not therefore satisfy the axiom mentioned above.
The answer to the question at the end of David Hill's paper[2] 'Is there a way of doing it that everyone would think fair in all cases?' is surely 'No'.
There are the hardliners on both sides - those who think that anything other than X-voting is not fair, and those who think that anything other than preference voting by STV, which I imagine includes David Hill, is not fair.
The most that one can hope for, then, is not a way of doing it that everyone would think fair, but a way that a majority of considered opinion would think fair.
The major response that I have had to my paper[1] so far is that 'it is a good idea'.