Home > Archive > Compression > April 2006 > Re: Please, PLEASE, hold your questions/comments/elsewhat til the end. Thank you. :)
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Re: Please, PLEASE, hold your questions/comments/elsewhat til the end. Thank you. :)
|
|
|
| >there is not enough information in N-1 bits to represent unambiguously all files of N bits??
who said that?
this would imply that entropy of all files of N bits were equal? but
who but a fool would enforce this limit upon themselves?
if all files of N bits occurred, but occured differing numbers of times
in the grand scheme of dowload then can you say that the average file
dowload will be less than N bits?
what if you treated the files as split into multiple downloads, would
that not lead to compression as exists today? then surely it is just a
matter of creating higly redundant streams of information from each
other and putting a critical user bit in some to make the information
content upto the nice average value.
so multiple related dowloads with minor information input is equivelent
to a single generator download when the input of critical user data can
be uniquly decoded to provide the previous small download by functional
generation from knowing the mutual information of the multiple related
small downloads.
this is absorbative rather than reductive.
cheers.
| |
| Earl Colby Pottinger 2006-04-03, 6:55 pm |
| "jacko" <jackokring@gmail.com> :
> files of N bits??
> who said that?
Anyone who has high school level math and spends more than 10 seconds of
serious thought.
> this would imply that entropy of all files of N bits were equal? but
> who but a fool would enforce this limit upon themselves?
Did you somehow miss the words "unambiguously all files"?
> if all files of N bits occurred, but occured differing numbers of times
> in the grand scheme of dowload then can you say that the average file
> dowload will be less than N bits?
Nope, because the statement was *ALL* files.
> what if you treated the files as split into multiple downloads, would
> that not lead to compression as exists today? then surely it is just a
> matter of creating higly redundant streams of information from each
> other and putting a critical user bit in some to make the information
> content upto the nice average value.
These tricks never work. At first you think you are saving data but the more
you try to cover, the more information needed to point to where the required
data is. The pointers grow in size and you end up writting the same or more
number of bits to represent the the original data files.
> so multiple related dowloads with minor information input is equivelent
> to a single generator download when the input of critical user data can
> be uniquly decoded to provide the previous small download by functional
> generation from knowing the mutual information of the multiple related
> small downloads.
Nope, the number of pointers keeps growing.
> this is absorbative rather than reductive.
Nope, It is not well thought out, sorry.
Please, Please, Please if you think you are right - don't argue - write code
and prove you are right.
Everyone here would love to see a working program that can do what you claim,
no-one believes that it can be done but if you do it - believe or not we will
move Heaven and Earth to insure you get to win the Nobel Prize. Really!
Earl Colby Pottinger
--
<Cruising, building a Catamaran, Rebuilding Cabin, New Peroxide Still Design,
Writting SF, Programming FOSS - What happened to the time?>
|
|
|
|
|