Re: [WG-BB] Study Report: CATV assignment guidline

[Yoshihiro Obata <obata@eaccess.net>]

At 14:47 02/02/06 +0900, Takashi Arano wrote:

> >   Do you mean that NICs are going to trace cable technology so that 
> they always understand the relationship between bandwidth and number of 
> ports? If they are going to set rules based on bandwidth, how can you 
> maintain rules if a new technology comes in which will change the 
> relationship such like 30M with fewer number of ports.  I do not think 
> bandwidth is a valid information to be kept in the long term.
>
>Yes, this is a guideline not a policy. A guideline can be and should be updated
>appropriately.

   I do not think that NICs can follow system specifications in a timely 
manner.  We have been experiencing too many interactions just because NIC 
operators try to catch up with technology they have never used.  Even if 
we agree that 30Mbps equipment from certain vendors may handle a certain 
number of users for the time being, in order to allocate IP addresses 
according to bandwidth information, NIC operators should check every time 
there is no such equipment with 30Mbps that will handle users in a 
different manner or they should check the equipment's specification to 
see that it has not changed since we set the guideline.  This is a tough 
job and also it is an over disclosure of information considering that you 
need to plan for the number of IP addresses several months before deployment.

>Another important thing. This guideline suggests to give the applicants 
>two choices.
>To follow the current way the NIC is doing now, or to use more simplified way.
>
>So please make your position clear when you criticize any proposed idea.
>You can support the current WG-BB proposal, you can support JPNIC idea,
>you can even support no guideline (this is just to do it as it goes), or 
>whatever.
>All would be fine. However, please understand that just saying no without
>any alternative results in supporting the current practices and no change.

   If there is any existing guideline and there is no new valid guideline, 
considering that I do not think bandwidth as a good guidelin, my position 
is that we should stick to the existing one.  I think it is not good to 
see problems half solved by an idea that creates other fundamental problems.

Yoshi

-
- This list (wg-bb) is handled by majordomo@lists.apnic.net