Re: Motion 62: YES (with minor amendment)
On 2014-06-23 11:08:56 -0400, Ian McIntosh wrote:
> Standards and most technical documents should entirely avoid any use of
> "comprise", since it has two meanings that are almost opposites. One is
> that the parts comprise the whole, the other is that the whole comprises
> the parts. Words that have been corrupted into their opposites should be
> abandoned. Using them compromises clarity.
Note that P1788 defines its meaning in §1.5:
– comprise indicates members of a set are exactly those objects
having some property. An unqualified consist of merely asserts
all members of a set have some property, e.g., “a binary
floating-point format consists of numbers with a terminating
binary representation”. “Comprises” means “consists exactly of”.
but I'm not against a change (for this one, as you suggested below,
and possibly others). In particular, I think that
"The implementation may comprise a programming language as well as
extensions in the form of libraries, classes, or packages that are
necessary to satisfy the requirements for conformance."
is a bad use of "comprise", because the "exactly" doesn't make much
sense with "may".
> I propose "It consists of those subsets...".
--
Vincent Lefèvre <vincent@xxxxxxxxxx> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)