John,

I "think" we have had this discussion (several years ago) on here about the difference. My vague memory is unimportant as to how it started but that it has caused confusion ever since. When the author wrote his (secondary?) email it raised my suspicions and decided that the author wasn't clear and decided not to pursue it as he was not giving an accurate picture.

Ed

On Dec 26, 2012, at 11:35 AM, John Gilmore wrote:

Tom,

Thank you for the silent correction.  The 'exa' in 'exabytes' is
certainly a radical improvement over 'exo', which was not
confidence-inspiring.

That said, it seems to me that for these magnitudes the binary prefix
'exbi' should be used.  We have

(2^10)^6 = 115_2921_5046_0684_6976  exbibytes

(10^3)^6 = 100_0000_0000_0000_0000  exabytes

and there is thus a non-trivial 13+% difference between these two numbers.

All this began with the notion of the rough equivalence of 2^10 = 1024
and 10^3 = 1000, which is a 2+% difference.

The practical difference between a kibibyte and a kilobyte was thus
unimportant, particularly in discussions among highly numerate people
who understood what sort of approximation they were using.

Things have, however, changed.  We are now often dealing with the
easily confused  innumerate, and the differences are large enough to
make dissimulation attractive to some, certainly not all, marketing
types.

John Gilmore, Ashland, MA 01721 - USA

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Reply via email to