On 8 Aug 2007, at 20:10, C. M. Sperberg-McQueen wrote: > (But my experience suggests that "may not" > can be ambiguous for some users between the meanings "must not" > and "may or may not"; you might consider saying something like "it > is possible that the result of applying the transformations identified > in these schemas will not be a faithful rendition" or something like > that. Emphatically so. I think the only people for whom this is _not_ ambiguous are those who have memorised RFC2119. I'm a standards junkie and I still get 'may not' the wrong way round (that is, I always initially parse it as 'must not'[1]). Perhaps it's a transatlantic thing, but I feel it would be a lot clearer if 'may not' were avoided in careful prose, and replaced by whichever of 'must not' and 'might not' is appropriate. Or periphrasis: `...may be an unfaithful rendition'. Note that although 'must not' and 'should not' are described in RFC2119, 'may not' isn't mentioned; I feel this is significant. Best wishes, Norman [1] The use of 'may not' in this particular context happens to be reasonably clear, but that doesn't subtract from the general point. -- ------------------------------------------------------------ Norman Gray : http://nxg.me.uk eurovotech.org : University of Leicester, UK
Received on Thursday, 9 August 2007 10:24:04 UTC