From: Emil Gilliam (emil@emilgilliam.com)
Date: Fri Nov 28 2003 - 21:59:02 MST
> +1, relevant. No time to comment though.
No one-line replies, especially useless ones.
>
> Most MMX does indeed *Single Instruction* on Multiple (64-128 bit
> registers, seen as array of 32, 16, 8-bit data types). It is a specific
> type of parallelism (not MIMD), and it still has to be applied
> sequentially
> on ~10^9 cells, using about 6 GByte/s throughput in best case
> (sequential
> stream with prefetch).
>
Chat about current processor architectures, programming languages, etc.
has been semi-tolerated in this list even when it doesn't address SL4
topics, but it ultimately belongs elsewhere.
> My keyboard just jammed so I'll be brief. People like Stafford Beer
> knew of this chaos first hand and would have been closest to breaking
> through the complexity barrier. Who knows where we would have been
> today if his cybernetic revolution wasn't crushed by the CIA when they
> planned to murder Chile's Salvador Allende back in 1973. It is also my
> view that the mistakes were made hundreds of years earlier when
> Capitalism scarred human nature into this condition. Not Egyptian
> mathematics but Capitalism and the early Christian concepts of zero,
> and how the debit and credit system evolved into our present monetary
> systems, is what did us in. Not respecting zero as the inverse of
> infinity is what stops us. In their naivety, what those computer
> pioneers didn't get right was how to cope with the frailty and fear
> inherent in human nature; - spam, FUD, etc.
>
That historical mistake known as the List Sniper crushes this thread
dead at this point, unless someone has something really new,
intelligent and SL4-worthy to say about the Lanier article.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:43 MDT