Finding incorrect or unclear use of terms in ISO standards is not uncommon. In
my experience, however, its not just possible but highly welcome to suggest
fixes during review-- sometimes it just takes they a few iterations to get it
right. We are, afterall, not really asking to change "intent" but the choice
of wording to express that intent.
I think we all have little doubt what the original authors-- I'm sounding
almost like a hermeneut-- intended by "reduced accuracy".
In ISO8601:1988 it was implicit that dates were as accurate as they are
precise. Since they were aligned it did not matter (much) which term they
choose. They went with the one most used in vernacular rather than scientific
literature. With our extensions in Part-II it now matters. We have thrown
knowledge, belief, reliability etc. into the works. Dates now can be expressed
with explicit divergence of accuracy and precision. Any reluctance, I suspect,
hinges on not grasping this quite fundamental change!
On Thu, 25 Feb 2016 12:46:13 +0000, Byrd, Donald A. wrote
> I agree, it _is_ disappointing. But, Ray, would it be possible to at least
make the definition of "representation with reduced accuracy" less
misleading? Section 2.3.7 currently defines it as "abbreviation of a
representation by omission of lower order components". Could that be changed
by just adding at the end ", thereby reducing the precision of the
representation" or some such?
> On Feb 24, 2016, at 12:25 AM, Nathan Harrenstein
<[log in to unmask]> wrote:
> > It may not be a "serious defect" as you say, but neither is the
punctuation, spelling, or format, but those will surely be corrected during
the editing process. it is disappointing that this misuse of the term will
continue to be propagated, and to be frank it is a bit hypocritical coming
from an organization which creates and promotes standards.
> > On Tue, Feb 23, 2016 at 12:46 PM, Denenberg, Ray <[log in to unmask]> wrote:
> > We all among us understand that [WINDOWS-1252?]“reduced [WINDOWS-
1252?]accuracy” where it occurs should be [WINDOWS-1252?]“reduced [WINDOWS-
> > However the existing 8601 uses [WINDOWS-1252?]“reduced [WINDOWS-
1252?]accuracy”. It is beyond our charter to challenge and correct that. The
committee was chartered to fix serious defects, and some have been identified
and fixed. But this cannot be considered a serious defect: the misuse of the
term [WINDOWS-1252?]“accuracy” in place of the correct term [WINDOWS-
1252?]“precision” is not going to impede interoperability.
> > And so, when I adapted the draft spec to 8601 form, I needed to conform to
the template set in 8601. Doing otherwise would cause confusion during the
review and could put approval of this spec in jeopardy.
> > Please note though, in cases where a new feature (EDTF) invokes the
concept of precision, the term [WINDOWS-1252?]“precision”, has been retained.
(I was able to successfully win the argument in those cases.)
> > Ray
> Donald Byrd
> Woodrow Wilson Indiana Teaching Fellow
> Adjunct Associate Professor of Informatics
> Visiting Scientist, Research Technologies
> Indiana University Bloomington
Edward C. Zimmermann, NONMONOTONIC LAB
Federal Republic of Germany