1 |
On Sun, Mar 30, 2014 at 7:54 PM, Rich Freeman <rich0@g.o> wrote: |
2 |
|
3 |
> On Sun, Mar 30, 2014 at 7:44 PM, Douglas James Dunn |
4 |
> <djdunn.safety@×××××.com> wrote: |
5 |
> > The system you are most familiar with really depends on what Operating |
6 |
> > System you use. if you don't use computers you probably were exposed to |
7 |
> > either the SI units or imperial base 10 units. |
8 |
> |
9 |
> SI units ARE in base 10. Most imperial units aren't in base 10, and |
10 |
> the SI prefixes aren't generally used with imperial units. You don't |
11 |
> usually report height in centiyards, etc. |
12 |
> |
13 |
> There seems to be some kind of misconception that this has something |
14 |
> to do with imperial vs metric units. |
15 |
> |
16 |
> Bits and bytes are such a modern concept that they were pseudo-metric |
17 |
> from the start, but programmers tended to use the SI prefixes in |
18 |
> non-SI ways - defining a kilobyte as 1024 bytes. "Kilo" is an SI |
19 |
> prefix, but the SI defines it as 1000, not 1024. |
20 |
> |
21 |
> The 1024-byte kilobyte was never metric or SI or imperial. Fairly |
22 |
> recently JEDEC codified the 1024-byte kilobyte, but also endorsed the |
23 |
> 1024-byte kibibyte, and the usage obviously predates that standard. |
24 |
> Before then, programmers never really had a "standard" for the |
25 |
> kilobyte. Since programmers don't tend to do a lot of compound units, |
26 |
> getting their terms endorsed by a standards body was probably not much |
27 |
> of a priority. If they had gone to the SI/ISO (or whatever was around |
28 |
> in the 60s) they'd almost certainly have been shot down on having a |
29 |
> 1024-byte kilobyte. |
30 |
> |
31 |
> Rich |
32 |
> |
33 |
> |
34 |
I called it imperial base 10, in the fact that you count 1-9 before hitting |
35 |
10 then 10-19 before hitting 20, rather than base 2, or whatever base you |
36 |
apply, not the fact that the units themselves are, and i realize that SI |
37 |
are in base 10 also. |