Does anyone here have a grounded opinion about NDN (Named Data Networking)? If that's as new to you as to me, I'm reading here: http://named-data.net It seems to have a lot of implications that I'm trying to think through, not the least of which is the idea that the Internet can and must evolve from an end-to-end communication system to being, itself, a content distribution network. The reason for thinking this way is that "YouTube, Netflix, Amazon, iTunes, etc., account for more than half of the world's internet traffic" and that reality "is inefficient and unsecure because the information-centric overlay is a poor match to the Internet's conversationally-oriented underlay." I am, myself, perpetually sceptical of efforts to start over, or to put policy into the network itself, such as NDN's aim to make every transmitted packet be signed and to make the network stateful including providing in-network memory. As always, "Who is this for?" needs an answer that looks well ahead. --dan
On Sat, Dec 5, 2015 at 10:41 AM, <dan@geer.org> wrote:
Does anyone here have a grounded opinion about NDN (Named Data Networking)? If that's as new to you as to me, I'm reading here:
back in the day, we called it IPoverATM / VBR-RT with SSCOP signalling.
It seems to have a lot of implications that I'm trying to think through, not the least of which is the idea that the Internet can and must evolve from an end-to-end communication system to [ insert speculative crap for craven private interests here ] ...
yeah, it's all bullshit. perhaps T-Mobile gives "named data" a free pass, but they're not the world, and they're not the endpoints, by any measure!
I am, myself, perpetually sceptical of efforts to start over, or to put policy into the network itself,
anyone putting policy into the network, rather than at the edges, is a robber baron being paid in filthy lucre :P best regards,
participants (2)
-
coderman
-
dan@geer.org