Re: ULA-C (Was: Re: IPv6 will never fly: ARIN continues to kill it)

[Iljitsch van Beijnum <iljitsch at muada.com>]

On 20-sep-2007, at 14:42, Thomas Narten wrote:

> A key point here is that when it comes to sales and marketing, it's
> problematic when your competitor says "we offer X," if you yourself
> don't. Given the commodity nature of ISP service, it doesn't take long
> before everyone is offering similar terms, even if there are
> technically bad implications

[...]

> The concern is that pretty soon, everyone will route ULAs because they
> feel like they are at a competitive disadvantage if others are doing
> so and they are not. And that would a huge mess.

The point you're missing is that one ISP can't provide global  
reachability for a prefix, you only get this if everyone cooperates.  
That just isn't going to happen unless someone with a huge amount of  
clout is going to force the issue. If Google wants to be reachable  
over ULA space then people may open up their filters. If it's IBM or  
Boeing, nobody is going to care.

And to people who can get PI or PA space, there is no point in  
forcing the issue, because even if they're successful in the end,  
it's going to be painful and expensive for them, too.

But even if it happens: who cares?

> And what if only _some_ of the ISPs routed them? We'd still have a
> mess, because now we'd have a Balkanized Internet, where univeral
> connectivity wasn't the norm anymore.

That sounds like an apt description of the current IPv6 internet. It  
works well in Europe and Asia, but North America is a wasteland:

$ ftp ftp.ietf.org
Trying 2610:a0:c779:1a::9c9a:1095...
ftp: connect to address 2610:a0:c779:1a::9c9a:1095: Operation timed out
Trying 156.154.16.149...
Connected to ftp.ietf.org.


_______________________________________________
Ietf mailing list
Ietf at ietf.org
https://www1.ietf.org/mailman/listinfo/ietf