Re: Randomizing records in a file

[Rory Hewitt <rory.hewitt@xxxxxxxxx>]

All joking apart, height, hair color, eye color, first letter of name,
length of name etc. all correlate fairly closely with some racial and/or
ethnic characteristics. It's harder than you might think to find a
'pseudo-random' characteristic which really is random enough. Maybe day of
birth (within month).

I worked on one program which used the third letter in the surname as a
'randomizer'. It worked perfectly until it came to Mr Ng...

On Tue, Jun 5, 2012 at 12:20 PM, John Yeung <gallium.arsenide@xxxxxxxxx>wrote:

> On Tue, Jun 5, 2012 at 3:16 PM, Michael Ryan <michaelrtr@xxxxxxxxx> wrote:
> > I vote for alphabetical by height.
>
> I think by descending hair color might be slightly better, though.
>
> John