[
Lists Home |
Date Index |
Thread Index
]
At 12/09/2002 18:05:13, Adam Turoff <ziggy@panix.com> wrote:
# On Thu, Sep 12, 2002 at 05:22:35PM +0100, Alaric B. Snell wrote:
# > I can't lose the feeling that something here is terribly, drastically,
wrong
# > here. Nobody ever made an XDR (eXternal Data Representation, not the other
# > one) accelerator, an ASN.1 BER accelerator, or anything like that.
#
# Nobody ever made XDR or ASN.1 a household word, either.
ASN.1 is in some *speicialist* quarters, or so I was reading.
# > Why should
# > you need special hardware for the minor issue of how you encode your
# > information? I mean, if it's a complex encoding technique like data
# > compression or encryption then sure, but XML isn't like that.
#
# You *need* special hardware to accelerate your XML processing
# because that $1,000,000 StarCat you bought last year is burdened
# by unused *generic* processing components. Everyone knows that
# special purpose hardware can easily get an order of magnitude
# performance increase over general purpose computers. It doesn't
# matter what performance metric you use -- MIPS/pound, TB/liter or
# GFLOPS/watt -- the special purpose hardware *always* offers better
# performance.
I suppose if they produced an XML dialect especiailly for climate/supernova
modelling, maybe.
# Plus, this XML Accelerator comes in a pretty green 1U enclosure. And it
# has those cute *go-faster* stripes on the side!
It looks like a lawnmower.
Roger
|
