Re: question about index cost estimates

Tom Lane wrote:

> I was considering a histogram for b-trees, but I have to admit I hadn't
> thought about r-trees. Seems like a 2-D histogram would be too bulky
> to be feasible. Could we get any mileage out of two 1-D histograms
> (ie, examine the two coordinates independently)? Or is that too
> simplistic to be worth bothering with?
>

i don't think it would do any good to look at them separately. you
might as well just assume a uniform distribution if you're going to do
that.

does anybody on the list know anything about fractals & wavelets? i
know _nothing_ about it, but i know that you can use wavelets to
compress photos (i.e., a 2-d data source) and my understanding is that
you're essentially converting the image into a mathematical function.
it's also my understanding that wavelets work well on multiple scales,
i.e., you can zoom in on a picture to get more detail or zoom out and
get less detail. my thought is if you've got a histogram, would
something like this be useful?

like i said, i have no idea how it works or if it is of any interest or
if it's even practical, just throwing something out there.

jeff