Julie Porter wrote two interesting articles in MMD 02082002.  Her
comments about Babbage are interesting; I had not realised that von
Neumann was inspired by Babbage's work.  I hope she has seen the
working reconstruction of the Difference Engine at the London Science
Museum.

Her forthright view of "snake oil" software is an excellent reality
check which should be applied more often when something seems too good
to be true.

However I would like to pick a very small nit.  She said that the
"FFT/DCT/DSP is a one-way lossy compression."  DCT (discrete cosine
transform) is not by itself either a compression or a lossy process.
The set of coefficients generated by a DCT can be transformed back
exactly to the original, subject only to normal rounding errors.
Losses occur when you discard some of the coefficients to obtain
data compression, for example, in JPEG photo images.

Jeffrey Borinsky, London