At 3:11 PM -0500 5/2/07, Richard Lynch wrote:
On Wed, May 2, 2007 7:55 am, tedd wrote:
Let me estate the problem. From your dB of things (the population),
you're going to pull out the top 100 most popular items and then
divide them into five groups using labels t1 through t5. Considering
tha
On Wed, May 2, 2007 7:55 am, tedd wrote:
> Let me estate the problem. From your dB of things (the population),
> you're going to pull out the top 100 most popular items and then
> divide them into five groups using labels t1 through t5. Considering
> that it's css guy asking for this, he's probably
On Tue, May 1, 2007 4:48 pm, Daniel Brown wrote:
> I don't think I'm quite following why you wouldn't just want to
> break it
> up into groups of 20
Because in the real world, it's not an even distribution of popularity.
There are only 1 or 2 Legends, and a handful of Rock Stars, and a
co
At 4:42 PM -0500 5/1/07, Richard Lynch wrote:
Little help here?
Richard:
Let me estate the problem. From your dB of things (the population),
you're going to pull out the top 100 most popular items and then
divide them into five groups using labels t1 through t5. Considering
that it's css gu
I don't think I'm quite following why you wouldn't just want to break it
up into groups of 20
On 5/1/07, Richard Lynch <[EMAIL PROTECTED]> wrote:
My boss and the web designer have decided to do something that
requires statistical formulae well beyond my statistically-challenged
capabil
My boss and the web designer have decided to do something that
requires statistical formulae well beyond my statistically-challenged
capabilities, so I'm turning to y'all...
Basically, the current query looks something like this:
select * from
(select whatever, count(*) as popular
from the_t
6 matches
Mail list logo