At 02:19 PM 11/28/01, Fred Baker wrote:
>At 04:05 PM 11/27/2001, Anthony Atkielski wrote:
>>You'd think that an ISP, cable-company or not, would tend to charge by 
>>volume or
>>connection time rather than the number of IP addresses in use.
>
>dumb question.
>
>I see a longish thread about "so why does one need IPv6, there is in fact 
>no shortage of IP addresses, and IP addresses are not in fact a scarce 
>resource in the network today".
>
>I see a longish thread about the fact that some cable companies apparently 
>are desperate to charge per IP address (something one can only do if IP 
>addresses are in fact a scarce resource) and want to nullify a technology 
>that might mask the scarceness of that resource.
>
>It seems to me that these two can't both be true. IP Addresses cannot at 
>once be scarce enough to charge for and non-scarce enough that scarcity is 
>a non-issue.
>
>Does anyone else see something schizoid about this discussion?

Yes.

To add even more of a schizoid flair, ATT Broadband is selling a "business 
service" in New England now (and probably elsewhere). For a few hundred 
bucks a month, and a one time charge for equipment, they set you up with a 
router of some sort running NAT which they can remotely manage, an SLA that 
says they'll possibly support you better than they do the home users, and 
the option of paying more for extra downstream bandwidth. So here we have 
the cable company SELLING the NAT box... I'd hoped they'd offer routed 
subnets when they got to offering business service, but no.

The other interesting oddity about the discussion threads is the assumption 
on the part of many that the only way it'd be useful for a water meter, 
kitchen appliance or other device in the house to interact with the 
Internet is for it to be polled from outside. It's quite feasible for 
devices to initiate the communication instead. Interestingly, if the device 
initiates, this would just happen to work with most NAT boxes on the 
market. Of course that eliminates the address-space-shortage argument, so 
it's more fun to talk about polling the devices, even if impractical at 
present.


-----------------------------------------------------------------
Daniel Senie                                        [EMAIL PROTECTED]
Amaranth Networks Inc.                    http://www.amaranth.com

Reply via email to