Filters

[leo vegoda]

Hi,

On May 28, 2005, at 12:58 am, Iljitsch van Beijnum wrote:

[...]

>> When you look at the statistics published by the RIPE NCC you will  
>> often see several prefixes reported for a single allocation. This  
>> is because we report each allocation event separately.
>
> Argh, that is bad!
>
> Now I have to go and filter those out.
>
> I'm now trying to do this for v4 but it's incredibly slow.
>
> For v6, I don't even know how to go about it as I can't do 128 bit  
> integer math.
>
> Why do you guys do this?

Some people use this information when they use our raw statistics.  
Seeing as we store the information in the 'very raw' format  
internally, it's easy to help those people by reporting it in that  
format.

We're reviewing the statistics we publish at the moment, along with  
the other RIRs. If you have a particular request, or set of requests,  
please let me know.

[...]

> Another thing that worries me is that AFAIK, the current stats only  
> show the last allocation for an address block. So if a block is  
> allocated in 1995 and then returned in 1999 and subsequently  
> allocated again in 2001, I only see the 2001 allocation. So I can't  
> make good stats for long ago as it looks like returned address  
> space is never allocated.
>
> What I'd like to see is all allocations along with a "deallocate"  
> date or some such, so it's possible to see the entire history for  
> an address block.
>
> This will probably make the files a lot longer, though.

We might be able to do something along these lines.

Regards,

-- 
leo vegoda
Registration Services Manager
RIPE NCC