On Tue, Jul 30, 2013 at 4:29 PM, Brian E Carpenter <
brian(_dot_)e(_dot_)carpenter(_at_)gmail(_dot_)com> wrote:
On 31/07/2013 05:21, Melinda Shore wrote:
On 7/30/13 7:59 AM, Keith Moore wrote:
I don't think that's the problem; I think the problem is that most
users don't realize how much lack of transparency is harming them.
So "transparent Internet access" isn't a commodity. Transparency
would be cheaper if there were more demand for it, and there would be
more demand for it if people realized how much more utility they'd
get out of the Internet if they had it.
<n> decades in, I suspect that if there were going to be demand
for "transparency" we'd be seeing it by now. If VoIP wasn't the
kick in the pants that's been needed to change things, it's
difficult to imagine what else might be.
Users want applications to just work, but they (and many business
managers in our "industry") don't understand that when applications
fail unpredictably, it's often because of glitches in what we call
transparency.
However, we are in an arms race here. Every step to improve transparency
will be met by a further step in middleboxes that nibbles away at
transparency. We've been debating this for 15 years; have you seen
any real change in the balance of power?
Brian
Well this is the real issue. I really could not care whether my IP
addresses are constant end to end or change on each hop if the packets get
there.
What I do care a lot about is being able to work out what my network is
doing and which piece of equipment is responsible for any given fault.
One of the architectural shortcomings of the current situation is that the
Internet was originally designed to fill the gaps between the networks, the
inter-network. We now use the same architecture inside the local network.
Only it isn't optimized for network use in quite the same way or at least
the home user does not have a toolset that is as powerful for their
purposes as the Internet backbone providers have for theirs.
Instead what happened was that network devices that were poorly architected
to run under novel or windows or appletalk sprouted IP as a transport
choice. And in many cases this was done in a really shoddy way.
Filesharing, printers, the rest all hang off the network in ad hoc fashion.
Keith's obsession with NAT boxes is totally irrelevant to the core problem
which is that home networks don't use DNS as their
naming/directory/discovery infrastructure which is what they would do if
they were really Internet based. There is no model of what is 'normal' in
the network so no way to detect abnormal situations or the cause.
NAT is here to stay until the last IPv4 address is withdrawn which is 30
years off at minimum. And it will remain even then because there are so
many fun games to play with NAT.