I'm at hunch level, its strong enough to know there is something of value.

Though the bits I tend towards clarity are about extending the nervous system. 
Contracts will allow such an extension. So imagine a long line of people 
holding hands. Prick a hand on one side of the line and the SDR travel from 
hand to hand to the last persons brain intact and efficiently. Now replace 
different people with different protocols, thrift, protocol buffers etc it 
allows us to hop between protocols.

Going back to the people example.
Now put a region in each person's brain and create a pipeline of processing. 
SDRs will be able to traverse efficiently from one end of the line to another. 
If one person breaks the contract we can determine exactly who broke it and 
which party A->B->C->D did the error happen between B and C? The contract will 
tell us _exactly_ who broke it.

Only recently did we discover that the brains representations used are SDRs. So 
we also need to design something that converts our wired/wireless networks into 
successful conduits of SDRs. Things aren't really human readable as we cannot 
easily manually debug when things go wrong.

Having a ubiquitous core is one thing, we also need to design it such that its 
language can extend to the outer edges of the internet.

With the onset of named networking it makes total sense to not use human 
readable names but SDRs instead. Imagine an entire internet talking to each 
other with SDRs?

I believe the bytecode is used for efficiency reasons.

Kind regards
Stewart

Doug King <[email protected]> wrote:
>Stewart,
>
>Are you considering UBF for just data or for possibly bytecode in the
>network? I can see value in both.
>
>Would it be desirable (give a use case) for components to emit /
>consume /
>operate on little chunks of bytecode? Would this over-complicate things
>or
>simplify them? I'm thinking it may be part of the solution to the
>Networking conundrum, but it's just a hunch, not based on anything
>concrete.
>
>-Doug
>
>
>On Sun, Jan 26, 2014 at 6:21 AM, Stewart Mackenzie
><[email protected]>wrote:
>
>> Please read an interesting article by J Armstrong on UBF
>>
>http://armstrongonsoftware.blogspot.hk/2008/07/ubf-and-vm-opcocde-design.html
>> Kind regards
>> Stewart
>>
>> --
>> Please excuse my typos and brevity
>>
>> _______________________________________________
>> nupic mailing list
>> [email protected]
>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
>>
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>nupic mailing list
>[email protected]
>http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org

Kind regards
Stewart

-- 
Please excuse my typos and brevity

_______________________________________________
nupic mailing list
[email protected]
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org

Reply via email to