[e2e] using p2p overlays to overcome recursive NATs/realms

[David P. Reed]

At 02:56 PM 2/9/2002 -0800, Randy Bush wrote:
> > I think it is pretty clear that ISPs have no interest in deploying v6.  A
> > fair number of them would love to stay in v4 because the lack of addresses
> > creates a steep entry barrier for competitors.  Others like the idea that
> > they can differentiate service for business users by charging an arm and a
> > leg per IP address, even when they aren't scarce.
>
>let's see.  the evil isps stay at v4 because the address space scarcity
>allown them to monopolize.  but they should give address space away because
>it's plentiful.
>
>perhaps you should at least try to understand the operational business
>before you foam irrationally and inconsistently about it.
>
>address space is a pita to the isps.  it is easily gotten if justified, but
>it is hard to track and manage and produces zero income.

Funny, the five or six ISPs I've bought from seem to bill for the number of 
addresses routed.  And it's a pretty fair piece of the bill.

>   and the cable
>jocks have a major pita because, with a million customers, provisioning
>special needs just does not scale, as there is no useful support in the
>technology (hint: you can't listen to customer igp because they'll announce
>network 16/8 or whatever at you).

Once upon a time in the mid 1990's, NYNEX seemed to think modems were a 
special need (a Sr. VP here said that there was "no demand" for people to 
connect PCs to their work from home over modems, much less faster 
links).  It's possible that cable ISPs are clueless about what the demand 
would be if they enabled customers to have routed networks.  (the same Sr. 
VP, a typical operator executive, said in the same conversation: aren't 
modems illegal on residential lines?)  Of course if "you can't listen to 
customers" or don't even try you'll never know whether they'd ask for 
something.  Maybe that customer's IGP is a legitimate wish that they'd pay 
for, rather than just a "pita".


>it's not the evil isps, but the clue-free ersatz researchers who foist v6

I don't know who you're talking about.  v6 has had 10 years for product 
developers to solve these problems or develop alternatives.  Why blame 
"researchers"?

>upon the world with a half-assed solution, more address bits and not one
>iota of work on routing, no transition design (pity the poor folk trying to
>sort out the dns v4/v6 transport mess), yet pontificate on the evils of the
>greedy bastards who are having trouble deploying the crap when only one
>significant router vendor can even support it.

I'm a business guy most of the time.  I'm not saying the operators are evil 
(though you interpret me as saying that).  I am saying that they make 
choices to optimize their ROI.  As you point out, if they can't make more 
money with IPv6, they won't deploy it.  But instead of helping them make 
money with v6, by enabling an architecture where it is easy to attach 
networks and build new applications, the IETF and equipment vendors have 
allowed the creation of a godawful kludge that is tangled in its own 
underwear, while at the same time the operators have fallen in love with 
the kludges like tight-fisted address allocation and NAT "features".


>get a clue
>
>randy