Is the inconsistency with the C++ standard enough to justify that the C
standards text be modified?

On Saturday, February 13, 2016, Kazutoshi SATODA notifications@github.com
wrote:

"one's complement" is inconsistent with the C standard.
"1's complement" and "2's complement" are also inconsistent.

FYI, I found that there was a lengthy discussion on Wikipedia at 2011
about "one's complement". The C++ standard was mentioned as a notable
use of "one's complement". There seems no consensus in the discussion,
but the main article has been settled as "ones' complement" for years
since then.
https://en.wikipedia.org/wiki/Ones'_complement

https://en.wikipedia.org/wiki/Wikipedia_talk:WikiProject_Computer_science/Archive_10#Ones.27_complement_or_one.27s_complement.3F

For me, a quote from "Art of Computer Programming, Volume 2" by
Donald E. Knuth is convincing.
https://books.google.com/books?id=Zu-HAwAAQBAJ&pg=PT343

... A two's complement number is complemented with respect to a single
power of 2, while ones' complement number is complemented with respect
to a long sequence of 1s. ...

While it is still debatable in general, I think inconsistency with the

C standard is enough for the C++ standard to justify this change.

You can view, comment on, or merge this pull request online at:

#619
Commit Summary

• Uniform "ones' complement" and "two's complement"

File Changes

• M source/basic.tex
  https://github.com/cplusplus/draft/pull/619/files#diff-0 (4)
• M source/declarations.tex
  https://github.com/cplusplus/draft/pull/619/files#diff-1 (2)
• M source/expressions.tex
  https://github.com/cplusplus/draft/pull/619/files#diff-2 (8)

Patch Links:

• https://github.com/cplusplus/draft/pull/619.patch
• https://github.com/cplusplus/draft/pull/619.diff

—
Reply to this email directly or view it on GitHub
#619.

@sigfpe I think it might be, for example, if the intent of the word in
question was to be compatible with the predecessor in C++.

Should the modifications be applied to the C standards draft instead?-------- Original message --------From: Kazutoshi SATODA notifications@github.com Date: 2/13/2016 8:06 AM (GMT-08:00) To: cplusplus/draft draft@noreply.github.com Cc: sigfpe sig.fpe@axiomatics.org Subject: Re: [draft] Uniform "ones' complement" and "two's complement" (#619)
@sigfpe I think it might be, for example, if the intent of the word in

question was to be compatible with the predecessor in C++.

—
Reply to this email directly or view it on GitHub.

@sigfpe I personally don't think so. I believe my above comments are
enough to explain why. If you think the C standard should be changed
instead, please explain why.

@sigfpe Sorry if I made the confusion. But no, I didn't.

It looks like "one's complement" is more common than "ones' complement" by a fairly large factor:

https://books.google.com/ngrams/graph?content=one%27s+complement%2Cones%27+complement%2Cones+complement&year_start=1800&year_end=2000&corpus=15&smoothing=3&share=&direct_url=t1%3B%2Cone%20%27s%20complement%3B%2Cc0%3B.t1%3B%2Cones%20%27%20complement%3B%2Cc0%3B.t1%3B%2Cones%20complement%3B%2Cc0

However, I'm inclined to take TAOCP as authority here, and I buy the argument that this is really the result of taking "ones's complement" and applying the usual English rule of Xs's -> Xs'.