value of pi and 22/7

Ian Kelly ian.g.kelly at gmail.com
Sat Jun 18 01:18:25 EDT 2016


On Fri, Jun 17, 2016 at 10:30 PM, Lawrence D’Oliveiro
<lawrencedo99 at gmail.com> wrote:
> On Saturday, June 18, 2016 at 3:48:43 PM UTC+12, Random832 wrote:
>>
>> On Fri, Jun 17, 2016, at 19:12, Lawrence D’Oliveiro wrote:
>>>
>>> I’m not sure how you can write “30” with one digit...
>>
>> One *significant* digit.
>
> Like some credulous past-Bronze-age tribespeople understood the concept of “significant digits” ...

I don't see why they should need to in order to measure one thing as
"thirty cubits" and another thing as "ten cubits" and write those
numbers down. Remember, the cubit was based on the length of the
forearm, so it's not like it was a terribly precise measurement to
begin with; they might not have understood significant figures, but
they probably wouldn't have been overly concerned about the difference
between thirty and thirty-one.

Check out the rest of the chapter. Every single measurement in it
above seven is a multiple of ten.

> I wonder what the quality of their workmanship was like, if a measurement accurate to one significant digit was considered sufficient ...

You realize there can be a difference between the quality to which
something is constructed and the precision of the measurements later
used to describe it? "Threescore cubits long"  is an impressive
figure. "61 and a half cubits" doesn't do the job of communicating the
scale any better, and ultimately amounts to wasted words in what was
originally an oral tradition.


More information about the Python-list mailing list