Re: Where is the flaw in my thinking?

On Mar 13, 5:44 am, moogie <budgetan...@xxxxxxxxxxxxxx> wrote:
I would like for someone who is more knowledgeable in the area of
compression to see the flaw in my thinking.

My thought process goes something like this:

- the more tightly a file is compressed, the more the compressed file
begins to look like random data.

.....

This is your first mistake. The fact is if you have a good compressor
that
is not adding control information or header information. When you run
compression hopefully the file is shorter thats all compression can
do.
The only reason it might look random is that there are more random
looking sort of files than nonrandom. But any file is possible
including
a file of all zeros or ones.
Many people wrongly assume that if the output is all zeros that the
compressor messed up and that you can compress again. True you can
compress again but then you would have a different compressor and
something else would go to the file of all zeros. Some files would be
longer and a few shorter. Compression is just a reordering of files
so
that hopefully more useful files are ordered first as you fill the
shorter
files inculding files that have random and nonrandom appearing
patterns.

I think people lose it when it comes to compression because when
done right which means lossless bijective compression. Each optimal
compressor is also an optimal decompressor for some criterian that
we be optimal for something. All each of these is, is no more than
a reordering of the infinite file space.

Even my BWTS is nothing more than part of a loop. Take any file
BWTS it enough times any your back to same starting file. Its just
that certain structured files the forward direction seems to make
useful compression better. But there exist files as measured by
some ceitera than UNBWTS would allow for better compression.

Since no compressor is best for all types of files we are chasing
our tails.

Hope this helps

David A. Scott
--
My Crypto code
http://bijective.dogma.net/crypto/scott19u.zip
http://www.jim.com/jamesd/Kong/scott19u.zip old version
My Compression code http://bijective.dogma.net/
**TO EMAIL ME drop the roman "five" **
Disclaimer:I am in no way responsible for any of the statements
made in the above text. For all I know I might be drugged.
As a famous person once said "any cryptograhic
system is only as strong as its weakest link"

.

Relevant Pages

  • Re: Arithmetic coding and continued fractions
    ... One "problem" when implementing an arithmetic coding algorithm is to ... Continued_fraction) have the potential of yielding a shorter ... does anybody know if the representation of an interval in AC ... of a given arithmetic compression. ...
    (comp.compression)
  • Re: File exist
    ... then every read-write key would, in your "no shorter" proposal, ... map through SD to a shorter value -- but that would, in effect, ... you can't have a compression function which uniquely maps every input ...
    (comp.lang.c)
  • Re: For Sean Pitman: Review of "Meaningful Information"
    ... For a binary alphabet, there is one string of length zero, two strings ... shorter strings must be moved to a slot taken by a longer one. ... Compression is useful for text because text has a very biased ... symmetrical complexity to the pattern. ...
    (talk.origins)
  • Re: Deflating a File
    ... Compression works by collapsing redundant sequences of information into ... shorter bit strings, typically through some sort of ... PRNG output, regardless of how many digits of output you have. ... data and detect the RNG algorithm. ...
    (comp.lang.java.programmer)
  • Re: OT but maybe interesting-compression schemes?
    ... >or zeros are required to describe a string of 1024 ones and zeros. ... Compression is basically rearrangement of this set of numbers. ... Frequent items, since they're near the base, need ...
    (rec.crafts.metalworking)