Re: [xml-dev] No XML Binaries? Buy Hardware

[Rick Marshall <rjm@zenucom.com>]

can we go back to analog computers?

On a more serious note. Can I add that it is the tree structure of the 
DOM that is expensive to parse - binary or text.

If you look closely, the best speedups come from binary schemes that 
effectively drop the tree structure.

If we flatten it (and this of course is trivial for data, but not so for 
documents) then text/binary is no longer as big an issue.

flatten == normalise

Rick

Elliotte Harold wrote:
> Len Bullard wrote:
>
>> My question was more along the lines of what I asked the other Mike 
>> (the one
>> with the accent):  what are these appliances doing that can't be done in
>> software or is the hardware more effective and why?
>
> Just as there are some people who believe that a binary solution is 
> always faster than text just because it is binary, so too are there 
> those who believe that a hardware solution is always faster than 
> software just because it is hardware.
>
> I've seen enough counterexamples to both to be deeply skeptical of 
> such claims, especially when someone starts touting either the 
> binary-ness or hardware-ness of their solution as an advantage in itself.
>
> Indeed these beliefs may not be unrelated. Both binary and hardware 
> solutions have the common distinction of being more opaque than their 
> competitors. Perhaps it is their very opaqueness that makes them 
> attractive.
>