Re: [nbs] NBS and TCP connection identification

[Christian Vogt <[e-mail]>]

Erik Nordmark wrote:

> It is pretty clear to me that if we want some form of id/locator 
> separation architecture, then some things will change. We definitely 
> have to be cognizant of this and understand the tradeoffs, but starting 
> off with "we can't change X" for any value of X isn't likely to be a 
> good approach.


Erik:

You are right.  We should be clear on what will change, and what does not have to change.

And I think the plan is fairly simple:  Operating system vendors will take on the burden, since they will need to add name-based sockets to their operating systems.  But they will be incentivized:  Name-based sockets will simplify application development, which has always been an objective of operating system vendors.

Once operating systems support name-based sockets, we are good.  Of course, we want applications to adopt the new API -- and they eventually will because this will make the life of application developers simpler.  But there is no time pressure.

Finally, we won't change routers, neither will be require new infrastructure.

- Christian