ip-precedence for management traffic

Mailman List Mirror (Read Only)
1

Joe wrote:

>I am still failing to see why what you're talking about cannot be done
>with today's technology.
>
>And if it can be done with today's technology, and isn't being done with
>it, either that's a business opportunity for you, or it says something
>about the model.

The later. It can be done today. So why is it not being offered?
There must be other forces at work.

It is (/was). You had things like WebTV. Spectacularly unsuccessful as
time went on. The problem is that the Internet is very powerful and very
big. Offer people a basic box that does basic things, and one person will
want to (also) pull up a PDF, another will (also) want to be able to
install a more modern version of Flash to support the latest video
capabilities being offered by $YOUTUBE_LIKE_SITE, a third will (also) want
Java support in order to play some trite online game, etc. Offer people a
basic Web-only Internet connection, and they'll want to know why their
gaming console doesn't work, etc.

Fundamentally, we have worked very hard for a very long time to create a
sort of baseline level of what is expected as part of "Internet"
connectivity. That's a basic IP connection. We've created standards on
top of that to allow applications to interoperate. It is hard to dip down
below that bar in terms of functionality. Anyone who tries is going to
end up eating support costs. How do you offer a "cheaper" level of
(let's say) Web-only Internet access, when the support costs will be
higher? Where's the value? What's the business plan? Where's the profit
in that?

I really meant what I said:

[I]f it can be done with today's technology, and isn't being done with
it, either that's a business opportunity for you, or it says something
about the model.

... JG

2

As an unrelated footnote, these are questions which will become highly
relevant when RIR address space depletion occurs and when providers
initially believe that they can create viable product sets based on
provider NAT, once their v4 address space can no longer service their
customer base.

[As a further sub-note, the "believe" bit is not a statement of scepticism,
but rather a statement of fact; many providers will almost certainly
believe that end-users will swallow provider NAT, regardless of whether the
product models turn out to be viable or not.]

So, although it's a different context, I think you'll see the answer to
these questions in this context in the next couple of years.

Nick