TCP/IP on the way out?

[Arthur Chance]

On 08/08/2014 09:58, Steve Burton wrote:
> On 08/08/2014 08:00, Odhiambo Washington wrote:
>> I trust that all is well with everyone.
>>
>> I have seen this article which sounds much like a dream, but seems true.
>>
>> http://www.networkworld.com/article/2459286/why-tcp/why-tcp/ip-is-on-the-way-out.html
>>
>>
>> I'd love to hear the views of those who understand the network stack.
>>
>>
> My favourite part of the article is:
>
>   "The key to the process, they say, is a network coding and decoding
> element called RLNC, or Random Linear Network Coding. RLNC is the
> technology that they've patented and wrapped into C++ software at
> Steinwurf called Kodo. Steinwurf plans to sell its technology to
> hardware makers."

If anyone writing an application wants something like this they can 
probably code up their own transport protocol using raptor codes over 
UDP and get 90% of the benefits without having to persuade people that 
replacing the network stacks of 8-10 billion(*) devices connected to the 
net is a good idea, economically feasible and can be achieved in a way 
that doesn't fragment the network for months or years on end.

It's rarely the technology that decides anything, it's the economics and 
politics. That's a lesson most techies seem to have to learn the hard way.


(*) Figure taken from 
http://www.forbes.com/sites/quora/2013/01/07/how-many-things-are-currently-connected-to-the-internet-of-things-iot/