[e2e] Re: [Tsvwg] Really End-to-end or CRC vs everything else?

[Melinda Shore]

> >   Anyway, to some extent
> >it's already the case that there's an implicit rejection
> >mechanism, in that some things already fail when certain kinds
> >of middleboxes are introduced (firewalls break session-oriented
> >protocols, NATs break integrity protection).
> 
> This is good?  

Several years ago I would have said no, but now I think
it's neutral, to be honest.  A lot of crud that's popping
up in the network is the inevitable result of privatization -
you can't change the economic incentives and assume that
everything else will remain static.  A model based on selling-
you-a-service is a model that assumes that, among a host of
other things, the edge of the network is visible.  It
probably doesn't *need* to assume that, but until someone
comes up with an end-to-end business model, the end-to-end
networking model is going to remain under assault.  And that
implies that technologies, like integrity protection, that 
rely on end-to-end transparency will continue to have problems.
I don't see how you can fix the technology without fixing
the business model.

Melinda