Hi Claude,
The MD file was likely added after M2 was released and would explain why it
didn't make it to the site (yet).
In the MD, you write:
"Bloom filter comparisons are extremely fast taking on approximately five
(5) machine instructions for the simple comparison."
I think this needs clarif
In the MD:
" This index arises from the observation that no target can match a filter
with a lower hamming value."
Shouldn't "hamming" be capitalized?
Gary
On Sat, Oct 12, 2024, 5:07 AM Claude Warren wrote:
> > ...
> > We can delay MultidimensionalBloomFilter unless introducing it later
> wou
Hi,
The MD refers several time to "10K filters". Where does this limit come
from? If there is a baseline CPU and RAM combination this is based on, the
document should state it IMO. Or, is it based on the width of a Java int or
a Java long?
TY,
Gary
On Sat, Oct 12, 2024, 5:07 AM Claude Warren wr
I read in the MD:
"...he data in transit and at rest is encrypted or at least strongly hashed"
What does "strongly hashed" mean here? Is this part of the Bloom Filter
vernacular? Shouldn't we define this somewhere?
TY,
Gary
On Sat, Oct 12, 2024, 5:07 AM Claude Warren wrote:
> > ...
> > We can
Thank you for explaining multidimensional filters here. It's not clear to
me yet if we need to introduce the new interface. I'll have to reread the
great docs you wrote 😀
The search() vs. locate() names are a bit confusing. Are these BF naming
conventions? If not, and if the difference is the exac
> ...
> We can delay MultidimensionalBloomFilter unless introducing it later would
> break binary compatibility. How beneficial would introducing this interface
> for users?
>
> Gary
>
>
> On Sun, Oct 6, 2024, 10:58 AM Claude Warren wrote:
>
> > This is starting tondelve into the realm of multidim
