ietf
[Top] [All Lists]

Re: dubious assumptions about IPv6 (was death of the Internet)

2004-01-19 20:12:25
On Tue, 20 Jan 2004, Iljitsch van Beijnum wrote:

On 19-jan-04, at 23:55, Dean Anderson wrote:

As Kazaa, Napster, Groove, and other protocols have demonstrated, its
quite easy to create peer-to-peer applications without either expensive
external infrastructure or fixed, unique IP addresses.

These protocols require that at least one side in each transfer is 
capable of receiving inbound sessions. This can work if the user has a 
NAT and configures it such that inbound sessions are forwarded 
appropriately, but it can NOT work if the NATting is done by a service 
provider of some kind without any way for the user to configure this 
NAT box.

It can't work only if Both users have NATs with no inbound access.  

A simple connection broker can probably make this work even with both
sides having no inbound access.  While external, this wouldn't quite be
'expensive external infrastructure'.

So what else is new. In Shakespeare's time there were people that went 
to plays, memorizing them for later copying and performing the play 
without paying the author. And remember these are the same people that 
thought the VCR would kill their business.

Duplicating VCR tapes, DVDs, and CDrom's is still relatively
expensive--more so than sharing files over Kazaa.  I'm against the DMCA,
but the old situation is qualitatively and quantitatively different from
today's filesharing.  You used to pay perhaps half the price for an
inferior product.  That's an easy proposition to resist.  Now you get the
same quality for free.  That is much harder to resist.  When there is no
expensive infrastructure necessary to support the activity, it is much
harder to stop.  I can see why that is alarming.

                --Dean