why, hello boys and girls
jules Gilbert
I have something to say but don't want to get attacked. Any chance?
Of not getting attacked?
--jg
Fibonacci Code
missle targeted, arsenal arrays positioned,
status : Fire at will.
chance of diplomatic : 0％
regards,
fibonacci.
-joke.....
Thomas Pornin
> I have something to say but don't want to get attacked. Any chance?
> Of not getting attacked?
If by "getting attacked" you mean "finding out that not everybody
automatically agrees with you and that some people might even dare
not to feel greatly honoured to bask in your radiance", then the
precise answer to your question is: "not a single chance".
If by "getting attacked" you mean "being mauled by a rabid bear",
then let me remind you that copper wire have a hard time transmitting
impulses other than electrical; grievous bodily harm rate is pretty
low in Internet-based telecommunications. In that situation, the
anwer to your question is: "no worry, you will remain physically
unscathed".
Your best strategy is probably to be, or at least act as, a grown-up
adult.
--Thomas Pornin
jacko
The entropy pigeon crew will always dis yer talkings. the fact that
zeroless number systems do exist, that you may have some interesting
vhdl source, and that your reality factorization part of your brain
may take years for wordage to stick to it so that its resonance is
brought forth in no way gives you batfink wings of steel.
there is no money in replacing an industry with smart logical use of
lesser resources. get over it. they need to not be able to source a re-
source (sic) before anything changes, with the resulting arguments
placing the top dogs in the pickings of a bone war.
the world is run by idiots with money!
cheers jacko
http://sites.google.com/site/jackokring for an understanding of
uncertain geometry. Nice but who really needs further development of
the nuclear pistol??
Industrial One
> jules Gilbert wrote:
> > Can we play nicely?
>
> > I have something to say but don't want to get attacked. Any chance?
> > Of not getting attacked?
>
> > --jg
>
> The entropy pigeon crew will always dis yer talkings. the fact that
> zeroless number systems do exist, that you may have some interesting
> vhdl source, and that your reality factorization part of your brain
> may take years for wordage to stick to it so that its resonance is
> brought forth in no way gives you batfink wings of steel.
>
> there is no money in replacing an industry with smart logical use of
> lesser resources. get over it. they need to not be able to source a re-
> source (sic) before anything changes, with the resulting arguments
> placing the top dogs in the pickings of a bone war.
>
> the world is run by idiots with money!
>
> cheers jacko
>
> uncertain geometry. Nice but who really needs further development of
> the nuclear pistol??
Hahaha, this is the first time I've seen Jacko address Jules directly.
Maybe he can play the role of Special Ed Math teacher and actually
succeed.
stan
I believe there is a common saying that goes something like:
Repeating the same behavior and expecting different results is the
definition of insanity.
My guess is that if you post working code you'll be fine; post
unsupported claims and you will get what you've always gotten.
sebastian
Perhaps the best thing to do would be to provide some sort of proof
that your method works. I suggest that you publish a decompressor -
even if it is just to one person in the group here (such as Mark
Nelson, given his reputation) - compress the files of our choosing and
then send the resulting output to whoever you choose to entrust your
program to, for verification. That would lay to rest, for once and for
all, the doubts that have been cast on your method. I've even put
together a simple verification driver program for you - all you have
to do is supply the 'decompress' function, compile, and then publish
the executable. Sound fair?
#include <stdio.h>
#include <stdlib.h>
typedef int boolean;
extern boolean decompress( FILE* in, FILE* out );
/*
// For testing purposes...
boolean decompress( FILE* in, FILE* out )
{
int
ch;
while( ( ch = fgetc( in ) ) != EOF )
fputc( ch, out );
return 1;
}
*/
void error( const char* message )
{
fprintf( stderr, "Error: %s\n", message );
}
int usage( const char* exe_file, const char* message )
{
if( message )
error( message );
fprintf( stderr, "Usage: %s <input_file> <output_file>
<compare_file>\n", exe_file );
return EXIT_FAILURE;
}
int main( int argc, char** argv )
{
FILE
* in,
* out,
* compare;
const char
* input_file,
* output_file,
* compare_file,
* exe_file = argv[ 0 ];
int
c1,
c2,
result = EXIT_FAILURE;
puts( "Gilbert Decompressor" );
if( argc != 4 )
return usage( exe_file, 0 );
input_file = argv[ 1 ];
output_file = argv[ 2 ];
compare_file = argv[ 3 ];
in = fopen( input_file, "rb" );
if( !in )
usage( exe_file, "cannot open input file" );
else
{
out = fopen( output_file, "wb+" );
if( !out )
usage( exe_file, "cannot open output file" );
else
{
if( !decompress( in, out ) )
usage( exe_file, "invalid format" );
else
{
compare = fopen( compare_file, "rb" );
if( !compare )
usage( exe_file, "cannot open compare
file" );
else
{
rewind( out );
for( ;; )
{
c1 = fgetc( out );
c2 = fgetc( compare );
if( ferror( out ) || ferror
( compare ) )
{
error( "file read error" );
break;
}
if( c1 != c2 )
{
error( "test failed" );
break;
}
if( feof( out ) )
{
puts( "Test passed" );
result = EXIT_SUCCESS;
break;
}
}
fclose( compare );
}
}
fclose( out );
}
fclose( in );
}
return result;
}
jules Gilbert
Do you actually expect me to publish my materials, even as an EXE
file?
Sorry I bothered you folks. Go back to your TV shows.
Pete Fraser
news:09874005-183c-4ae7...@g23g2000yqh.googlegroups.com...
> Sorry I bothered you folks.
I don't think you are, otherwise you'd stop doing it.
sebastian
>
> Do you actually expect me to publish my materials, even as an EXE
> file?
>
> Sorry I bothered you folks. Go back to your TV shows.
That's quite alright. Just remember that dishonesty reveals itself. We
should be thankful for that, I think.
jules Gilbert
> "jules Gilbert" <jules.sto...@gmail.com> wrote in message
>
> news:09874005-183c-4ae7...@g23g2000yqh.googlegroups.com...
>
> > Sorry I bothered you folks.
>
> I don't think you are, otherwise you'd stop doing it.
When I used the word "sorry" I was trying to indicate that I see the
futility of attempting anything serious with most of you; I wasn't
using the word in the usual sense that it is most frequently used.
Jim Leonard
> When I used the word "sorry" I was trying to indicate that I see the
> futility of attempting anything serious with most of you; I wasn't
> using the word in the usual sense that it is most frequently used.
That's ok. Eventually you'll expire and we will finally be spared
your non-verifiable details about your non-verifiable claims.
To be taken seriously, you have to act seriously. Publish working
code.
In case you haven't noticed, storage prices continue to plummet.
There will always be an academic reward for pushing compression
research, but the financial reward will all but disappear soon. I
don't really know what you're holding out for, unless you're afraid
someone will point out that the emperor has no clothes.
jacko
Ooooh. Sarcastic use of the word sorry. Publish all the code just
like...
-- (C)2008 K Ring Technologies Semiconductor
-- BSD http://nibz.googlecode.com (no support)
-- Comercial licencing available
-- Tel: +44 796 797 3001 (a real space odessy)
-- Maintained by Simon Jackson, BEng.
-- E-mail: jackokr...@gmail.com
-- Hardware support kodek for nibz
library ieee;
-- library altera;
use ieee.std_logic_1164.all;
use ieee.std_logic_arith.all;
entity k is
port (
do : buffer std_Logic;
di : in std_logic;
busy : buffer std_logic;
comde : std_logic;
clk : in std_logic;
-- a rising edge signal (run)
run : in std_logic;
-- for setting the prbs
-- may be of use as a general prbs ...
load : in std_logic_vector(15 downto 0);
save : buffer std_logic_vector(15 downto 0);
rw : in std_logic;
sel : in std_logic
);
end entity;
architecture rtl of k is
signal asymproc : std_logic;
signal which : std_logic;
type s is (action,motion);
signal state : s;
-- intermediates
signal wind : std_logic_vector(4 downto 0);
signal procp : std_logic_vector(1 downto 0);
signal tapx : std_logic;
signal ends : std_logic;
-- process control flags
signal procdo : std_logic;
signal winddo : std_logic;
signal modo : std_logic;
signal runlast : std_logic;
signal whenok : std_logic;
signal tmp : std_logic;
constant tap : natural := 3;
signal pmux : std_logic_vector(15 downto 0);
begin
process(clk)
-- prbs
begin
end process;
procp <= pmux(1 downto 0);
procdo <= procp(1) and procp(0);
-- set motion velocities via probability switching
modo <= procdo xor which xor do xor '1';
-- set sampling window
wind <= pmux(7 downto 3);
winddo <= wind(4) and wind(3) and wind(2) and
wind(1) and wind(0);
process(clk)
-- k state machine
begin
if(rising_edge(clk)) then
runlast <= run;
end if;
if(rising_edge(clk) and busy = '1') then
case state is
when motion =>
asymproc <= asymproc xor modo;
state <= action;
when action =>
if(asymproc = '0') then
if(which = '0') then
-- sample
if(winddo =
'1') then
busy
<= '0';
--
sample do
end if;
else
-- random
do <= do xor
procp(0);
end if;
else
which <= which xor
procdo;
end if;
state <= motion;
end case;
-- NB. if 2nd rising edge run before busy zero
-- then misses a radix slot
elsif(rising_edge(clk) and whenok = '1' and busy =
'0') then
-- set busy, cleared by kodek finish bit
busy <= '1';
whenok <= '0';
do <= tmp;
end if;
if(rising_edge(clk) and runlast = '0' and run = '1')
then
whenok <= '1';
tmp <= di;
end if;
-- prbs
if(comde = '1') then
ends <= save(15);
else
ends <= save(0);
end if;
tapx <= save(tap) xor ends;
if(rising_edge(clk) and busy = '1') then
if(comde = '1') then
save <= save(14 downto 0)&tapx;
else
save <= tapx&save(15 downto 1);
end if;
end if;
if(comde = '1') then
pmux <= save;
else
pmux <= tapx&save(15 downto 1);
end if;
-- load
if(rising_edge(clk) and sel = '1' and rw = '0') then
-- NB. a load while busy can be a random
number gen
-- but not much use in general operation
-- do a kind of reset
asymproc <= '0'; -- rand/samp side
which <= '0'; -- samp side
state <= motion; -- perform motion methods
first
-- (begin and end on them)
-- assign essentials
save <= load;
end if;
end process;
end rtl;
Pete Fraser
news:f915f754-4705-4e18...@c34g2000yqn.googlegroups.com...
> -- Hardware support kodek for nibz
What does this do?!?
Why do you use std_logic_arith?
Did you really intend comde and save to be unclocked?
If so, why no sensitivity?
Everybody wants a part of asymproc.
I'm surprised Quartus deals OK with that.
Pete
jacko
> "jacko" <jacko...@gmail.com> wrote in message
> news:f915f754-4705-4e18...@c34g2000yqn.googlegroups.com...
>
> > -- Hardware support kodek for nibz
>
>
> What does this do?!?
k bit is an entropy reduced bit when only sample at the signal valid
times.
>
> Why do you use std_logic_arith?
>
why not.... blah... swings... roundabouts...
> Did you really intend comde and save to be unclocked?
> If so, why no sensitivity?
>
clock is fastest! veey shrt tand.
> Everybody wants a part of asymproc.
> I'm surprised Quartus deals OK with that.
>
the asymetric bit process...
> Pete
Nimo
With more than 40 years of programming experience,
you are repeating the same age old 'Mistake'.
The Mistake of 'Lossless' compression.
Instead of that,
why don't you invest yourself some of your time
to unravel the mysteries of Light.
"The Strange theory of Light and Matter" by Richard Feynman
is a nice starting point.
As we are entering into the era of Quantum Computing,
May be your dream
(somehow My dream and Everyone's dream: The dream of Lossless
compression)
would be possible the day we over come the problem of 'Quantum
Entanglement'
{at least I think so..., } http://en.wikipedia.org/wiki/Timeline_of_quantum_computing
until then, you keep your self busy with this cat.
http://en.wikipedia.org/wiki/Schr%C3%B6dinger%27s_cat
_____
It’s a good thing we have gravity, or else when birds
died they’d just stay right up there.Hunters would be all confused
StevenWright, comedian.
jules Gilbert
> On Nov 14, 8:15 am, jules Gilbert <jules.sto...@gmail.com> wrote:
>
> > Can we play nicely?
>
> > I have something to say but don't want to get attacked. Any chance?
> > Of not getting attacked?
>
> > --jg
>
> With more than 40 years of programming experience,
> you are repeating the same age old 'Mistake'.
> The Mistake of 'Lossless' compression.
>
> Instead of that,
> why don't you invest yourself some of your time
> to unravel the mysteries of Light.
>
> "The Strange theory of Light and Matter" by Richard Feynman
> is a nice starting point.
>
> As we are entering into the era of Quantum Computing,
>
> May be your dream
> (somehow My dream and Everyone's dream: The dream of Lossless
> compression)
> would be possible the day we over come the problem of 'Quantum
> Entanglement'
> {at least I think so..., }http://en.wikipedia.org/wiki/Timeline_of_quantum_computing
>
> _____
>
> It’s a good thing we have gravity, or else when birds
> died they’d just stay right up there.Hunters would be all confused
> StevenWright, comedian.
No, "Nimo" -- whoever you are.
I have an actual program that includes a routine written in C, that
processes a stream of random appearing data (say, organized as bytes,)
to produce a compressible list of tokens and a dictionary; The
tokens refer to dictionary entries which, (now at the receiver
station,) can be cheaply interpreted to reproduce the random input
stream. Moreover the size of the token list plus the size of the
dictionary is less than the length (again, using bytes as the units,)
of the input file.
The system is fast and requires little extra memory.
The code does not contain any GNU licensed code and the underlying
algorithm is not the subject of an existing patent.
I do use Julian Seward's bzip2 data compression libary, but although
Julian doesn't require this, I give him a one line credit on my
copyright notices.
jules Gilbert
(I just saw this or I would have responded sooner.)
First, I am aware that data storage sub-systems are increasingly
efficient -- but I disagree with you as to what it implies wrt our
industry. And my program actually undoes randomness -- no other
program, no paper, not even the musings I see here -- comes anywhere
near this claim. And I don't simply have an empty claim, I do this.
It's real.
In point of fact I took almost 40k uncompressed source files from a
popular OS and reduced them to numbered bzipped files; My test
environment picks 25 at random for an 'experiment' a run of the
system, does statistics, and then does this again, producing another
line of statistics.
And my investors are aware that I've made the decision that without a
large one-time payout, (stock plus cash,) I do not intend to otherwise
market the program, -- and in fact, I keep my work well protected and
I alone know the passwords, so when I go, that's it.
I do expect to show MN though, that only makes sense. But I want
everything in place when I do this, I want to be able to field
prospect inquiries, etc. right now I am not setup to do this.
Noob
> In case you haven't noticed, storage prices continue to plummet.
Right.
> There will always be an academic reward for pushing compression
> research, but the financial reward will all but disappear soon. I
> don't really know what you're holding out for, unless you're afraid
> someone will point out that the emperor has no clothes.
I'm not sure this is true in the context of digital video.
Recently, resolutions have been pushed from 720x576 to 1280x720 to 1920x1080,
to probably even higher in the future.
http://en.wikipedia.org/wiki/High-definition_video#Standard_high-definition_video_modes
AFAICT, video compression (as in H.264, dirac) still is an active research topic.
Regards.
Noob
> Recently, resolutions have been pushed from 720x576 to 1280x720 to 1920x1080,
> to probably even higher in the future.
cf. http://en.wikipedia.org/wiki/Super_Hi-Vision :-)
"""
On December 31, 2006, NHK demonstrated a live relay of their annual [music show]
over IP from Tokyo to a 450 inch (11.4 m) screen in Osaka. Utilizing a codec
developed by NHK, the video was compressed from 24 Gbit/s to 180�600 Mbit/s
and the audio was compressed from 28 Mbit/s to 7�28 Mbit/s. Uncompressed,
a 20 minute broadcast would require roughly 4 TB of storage.
"""
Jim Leonard
> And I don't simply have an empty claim, I do this.
> It's real.
Without any published verifiable data, your claim is empty. Surely
you don't expect us to believe hearsay as fact? Do you treat hearsay
as fact yourself?
> In point of fact I took almost 40k uncompressed source files from a
> popular OS and reduced them to numbered bzipped files; My test
> environment picks 25 at random for an 'experiment' a run of the
> system, does statistics, and then does this again, producing another
> line of statistics.
Lines of statistics don't equal compression. Analyzing data and
finding patterns isn't compression. Putting all of that data into a
file 1. that is smaller than the size of the original that 2. can be
used to reconstruct the original -- THAT is compression.
Jim Leonard
> AFAICT, video compression (as in H.264, dirac) still is an active research topic.
Yes, but the context of our conversation was limited to the lossless
domain.
Jim Leonard
> I have an actual program that includes a routine written in C, that
> processes a stream of random appearing data (say, organized as bytes,)
> to produce a compressible list of tokens and a dictionary; The
> tokens refer to dictionary entries which, (now at the receiver
> station,) can be cheaply interpreted to reproduce the random input
> stream. Moreover the size of the token list plus the size of the
> dictionary is less than the length (again, using bytes as the units,)
> of the input file.
As a sign of good faith that you want to be taken seriously, please
make available the output of your compressor for this file:
http://marknelson.us/attachments/million-digit-challenge/AMillionRandomDigits.bin
No need to provide any working code or source, just the output.
jules Gilbert
>
> No need to provide any working code or source, just the output.
To Jim Leonard:
Jim, the moment I do this someone will want me to do another file,
possible a file that has been changed slightly from MN's test file.
And again, etc...
And soon people will be able to go to work figuring out how I do what
I do.
However, if you will state that you won't do this or help others who
might do this, that you will accept the file produced from MN's test
file, then I can accommodate you, In fact I'd love to.
I do promise that what I send you will have been produced from MN's
test file and that it can be used to restore to reproduce the
original. (I and my friends call this "round trip.")
Do you have an FTP site I might use to upload the intermediate file
produced by compressing that file?
--jg
jules Gilbert
>
> No need to provide any working code or source, just the output.
Oh, one more thing Jim;
My testing apparatus is based on almost 40k files, all bzipped and
based on available source code. But one file in that bunch is Mark
Nelson's baby!, and it's almost not fair to you to ask me this as I do
it all the time. But you asked, and it is a duzey, (a really
difficult file to compress.) Mark set out to make a file impossible
to compress, at least using conventional methods, and he certainly
succeeded, didn't he!
--jg
jules Gilbert
Ah, after thinking about what I've said here, and wanting to be
completely and fully accurate, I've decided to retract this claim.
Yes, for compression I have what I say I have.
But I recall years ago how researchers in the field of neural nets
thought that because they had proven that a neural net could be used
to approximate any continuous function that this was great and that
suddenly, the door to AI would swing open. Except it didn't happen.
Wrt compression, yes -- once I sell my program (details as to how to
buy to be published soon,) "perpetual" compression will be available
to each of the purchaser's, (nothing in nature is actually perpetual,
say, several to many times, depending on the original file size, large
files compress more times than small files.)
Anyway, I want to retract my implicit claim to have a general purpose
AI machine that can read any random stream and make sense of it.
Nope, I can't do that. Wish I could...
--jg
Jim Leonard
> Anyway, I want to retract my implicit claim to have a general purpose
> AI machine that can read any random stream and make sense of it.
So you are finally admitting it is impossible to create a system
capable of compressing *all* input sets? After 16 years, have you
finally understood information theory?
Earl_Colby_Pottinger
Did you note that his retaction has little or nothing to do with his
previous claims over the last decade plus? Infact, when on earth did
he even change his claim to that of "An AI Machine"? This retaction
is a lie, because it does not admit to most of the wrong/unproven
claims he has for more than a decade.
He will be back with another claim that he has found a way to compress
random data and if you point out his retaction he will claim that this
method does not use AI. As always it is best to treat him as a scam
artist first and foremost.
Thomas Richter
>>> Anyway, I want to retract my implicit claim to have a general purpose
>>> AI machine that can read any random stream and make sense of it.
>> So you are finally admitting it is impossible to create a system
>> capable of compressing *all* input sets? After 16 years, have you
>> finally understood information theory?
>
> Did you note that his retaction has little or nothing to do with his
> previous claims over the last decade plus? Infact, when on earth did
> he even change his claim to that of "An AI Machine"? This retaction
> is a lie, because it does not admit to most of the wrong/unproven
> claims he has for more than a decade.
>
> He will be back with another claim that he has found a way to compress
> random data and if you point out his retaction he will claim that this
> method does not use AI. As always it is best to treat him as a scam
> artist first and foremost.
It is more like filling the gap in a proof by saying "and here a miracle
happens", just that the (non-disclosed) "AI" jumps in to make said
miracle happen. After all, it is the "holy grail" of informatics, and
known over years for promising much, and providing little. (And, as soon
as something was provided, it was no longer called "AI"). IOW, if a "AI
could think, why couldn't it do anything else as well - for example
satisfying wishes?" (-:
I would believe in Jules' convoluted way of thinking he believes that an
AI is able to fill in the loss created by his compression scheme, being
able to fill in the gaps and errors created by lossy encoding "as much
as a hmn cld fll n th mssng lttrs in a txt". Of course, that wouldn't
still make it a lossless compression scheme, only one that works
reasonable for "natural text" with a pretty low loss rate. (And, I
wouldn't need an "AI" for it, but a dictionary, or at a higher level,
word statistics.) Now, Jules' might conclude that for any reasonable
source a AI that is sophisticated enough would be able to fill all
missing pieces in all cases without loss.
Not so, of course. Counting strings is enough to see this. I'm pretty
sure a native speaker (unlike me) could drill up a nice example of an
English text where a single mis-reconstructed letter would make a
tremendous difference in meaning, such that a native reader (leave alone
an "AI") would be unable to reconstruct the intend correctly, even in
presence of some context.
So long,
Thomas
Earl_Colby_Pottinger
> Not so, of course. Counting strings is enough to see this. I'm pretty
> sure a native speaker (unlike me) could drill up a nice example of an
> English text where a single mis-reconstructed letter would make a
> tremendous difference in meaning, such that a native reader (leave alone
> an "AI") would be unable to reconstruct the intend correctly, even in
> presence of some context.
It can get worse than that even. One time my father and I spent over
an hour fighting about something (I don't remember what now as that
was decades ago) but it turned out that we were in complete agreement
because we both did not assign the same meaning to the same words.
Simple example: That cat is blue!
Do I mean the colour of the cat, or it's emotions/behaviour? It is
possible to has a number of statements like that together such that
even the context is of little help.
Niels Fröhling
> Mark set out to make a file impossible
> to compress, at least using conventional methods, and he certainly
> succeeded, didn't he!
Mark didn't "make" the "file". And it's not difficult to find out what's the
"file" about, if you'd show some interest.
Ciao
Mark Nelson
>
> Mark didn't "make" the "file". And it's not difficult to find out what's the
> "file" about, if you'd show some interest.
>
Actually, I did make the file, although it is clearly a derived work,
not an original one.
The original work, by now-forgotten mathematicians at the Rand
Corporation, was an ASCII representation of 1 million base-10 digits,
produced by a random physical process which was carefully checked for
various types of correlating biases. See:
http://en.wikipedia.org/wiki/A_Million_Random_Digits_with_100,000_Normal_Deviates
I took the million digits and concatenated them to form a single large
integer, then created a file that contains the binary representation
of that integer.
Admittedly a fairly trivial step, but one which makes the terms of my
compression challenge much clearer.
While the million-digit compression challenge has not saved the world
from trolls, I feel that it has in fact helped to make it a slightly
better place. It seems to serve as a honey-pot, attracting dreamers to
a known location and giving them something to focus on. I regularly
hear from people who make it out to the other side unscathed. And it
does give those on this newsgroup a good, consistent response to
overly optimistic claims.
If I were ever to rewrite the challenge, I think the only thing I
would change would be a requirement that you produce a program that
can compress the million-digit file, as well as the million-digit file
after having been encrypted using a random key of my choosing. This
would help flush out schemes involving hidden data. (I still imagine
somebody hiding that file in glib.)
- Mark
jules Gilbert
> On Dec 9, 5:46 pm, Niels Fröhling <spamt...@adsignum.com> wrote:
>
>
>
> > Mark didn't "make" the "file". And it's not difficult to find out what's the
> > "file" about, if you'd show some interest.
>
> Actually, I did make the file, although it is clearly a derived work,
> not an original one.
>
> The original work, by now-forgotten mathematicians at the Rand
> Corporation, was an ASCII representation of 1 million base-10 digits,
> produced by a random physical process which was carefully checked for
> various types of correlating biases. See:
>
>
> I took the million digits and concatenated them to form a single large
> integer, then created a file that contains the binary representation
> of that integer.
>
> Admittedly a fairly trivial step, but one which makes the terms of my
> compression challenge much clearer.
>
> While the million-digit compression challenge has not saved the world
> from trolls, I feel that it has in fact helped to make it a slightly
> better place. It seems to serve as a honey-pot, attracting dreamers to
> a known location and giving them something to focus on. I regularly
> hear from people who make it out to the other side unscathed. And it
> does give those on this newsgroup a good, consistent response to
> overly optimistic claims.
>
> If I were ever to rewrite the challenge, I think the only thing I
> would change would be a requirement that you produce a program that
> can compress the million-digit file, as well as the million-digit file
> after having been encrypted using a random key of my choosing. This
> would help flush out schemes involving hidden data. (I still imagine
> somebody hiding that file in glib.)
>
> - Mark
Mark, I can process your file.
And to the nay-sayer's; I have zero interest in meeting your
requests. Rather quite the reverse. My ideal satisfactory resolution
is to show MN that I have what I say I have and politely say goodbye.
I recognize that I am unlikely to receive the $1B I am asking. Okay,
I accept that. But that is what the program is worth. And in fact I
am very familiar with several varieties of AI and I believe I
described my latest method accurately. (For about 15 years I build
specialized compilers tools and systems; My last project provided
programming services to programmers, much as computer algebra systems
provide mathematic-related services to mathematicians.
And Mark, no, I haven't attempted to store your 415k file in a souped
up load module, to me that seems like cheating.
Pete Fraser
> My ideal satisfactory resolution
> is to show MN that I have what I say I have and politely say goodbye.
So why don't you stop fucking about, and just do it?
jules Gilbert
> "jules Gilbert" <jules.sto...@gmail.com> wrote in message
First Pete, I compress Marks' file at least once a week. It's not a
big event for me. In fact on occasion can be found at a local site
(my machines are very secure, I've offered demos in the past -- if I
lost a laptop it would simply make a good platform for running
windoz.) I repeat: I've offered New Englander's demo's before. I
will do a little bit of vetting first, I'm not foolish, but several
people have seen my program function.
But in fact I have reason to be in Florida soon, which isn't all that
far away...
Pete Fraser
[I intend to] show MN that I have what I say I have
and politely say goodbye.
Pete replied:
> So why don't you stop fucking about, and just do it?
Jules replied:
[I've already compressed Mark's file three times before breakfast.
I don't trust some people. People might run an MS OS on my laptop,
I offered some nameless New England people to see my unspecified
demo, and mumble, mumble.]
But in fact I have reason to be in Florida soon, which isn't all that
far away...
Pete now:
So Jules, that was a lengthy waffle to a simple and strightforward
exchange. Have you agreed with MN to give him a demo?
If so, when will this be, and what will it show?
Pete
jules Gilbert
Pete: I don't crunch your remarks into "mumbles" -- please don't do
that to my remarks.
jules Gilbert
And Pete, maybe to you (being in California,) you don't realize that
Mark and I don't reside in the same community. (Probably a good
thing, good for both of us.) Above all Pete you should get it clear
that I don't move as a result of your requests.
This is a commercial endeavor for me; I recognize that to you, this
is just recreation -- well that's fine for you. But to me the subject
of repeatable compression is vitally important.
Industrial One
> This is a commercial endeavor for me; I recognize that to you, this
> is just recreation -- well that's fine for you. But to me the subject
> of repeatable compression is vitally important.
Which is why you have contributed no insight to the "subject" and
never released it the day you first shat on Usenet when bandwidth was
way more restricted and expensive. Nothing like crunching your own
imaginary Nobel prize you lousy shit-eater.
Pete Fraser
> [I intend to] show MN that I have what I say I have
> and politely say goodbye.
Later Jules wrote:
> And Pete, maybe to you (being in California,) you don't realize that
> Mark and I don't reside in the same community. [] Above all Pete you
> should get it clear that I don't move as a result of your requests.
I'm confused. I thought that was your plan.
I was just looking for confirmation.
Pete
jules Gilbert
wrote:
Fist of all, earning Nobel prize isn't what it used to be... (And not
just because of Obama, look at the other recipients. But the really
funny thing was Obama's acceptance speech, because he defended the
US's beginning new hostilities in Afghanistan.) Make no mistake,
that's the single country the US has the right to turn into a parking
lot; Residents of that country attacked the US and then, when we
asked for those individuals, their legislature said "no." That's why
we invaded and we have the perfect right to turn Afghanistan into one
giant parking lot.
But about compression. Folk's it's simple: Everyone here thinks that
it's impossible to decompose a random string into something else --
well I have a method that does this.
Please re-read that last sentence, that's my claim.
I do expect to get down to Texas but at the time of my choosing. It
won't be my first demonstration, I'm fielding local inquiries first.
Third, Lance, watch your language, your really destroying yourself and
you don't have to.
jules Gilbert
Pete, I intend to demo. But I have things happening up here. It's
not like I'm an ivory tower academic -- I have things I'm doing that
are important to me. (I'm not anti-academic, I'm just pointing out
that my agenda isn't based on "publishing" or other academic credit.)
Two local large companies have basically said to me, "call us" -- and
I'm interviewing market reps.
I have an unusual method by which I intend to sell my program, because
patenting doesn't work for small inventor's. (Think it does, just try
suing Microsoft -- because in the past they've stolen compression
technology and when taken to task, told the owner's "So sue." The
thing is, MS nearly won.)
And that company that owned the patent, Stack Electronics, spent $15M
fighting MS. But the last time I checked I didn't have that much
money in the bank and also, I am not certain that I'm going to live as
long as that law-suit took. Patenting just isn't the right solution
for me.
And guess what -- although the court award them $160M, MS was going to
appeal. In the end SE settled for about $60M -- as I recall. (You
can look it up.)
Only four times more than the $15M investment.
And these are not figures that would interest an investor. VC's
looking at these numbers would say the 4:1 ratio is too low, they
believe that a 10:1 gain is necessary just to keep a VC company
running, because most investments are flops.
vic
> I have an unusual method by which I intend to sell my program, because
> patenting doesn't work for small inventor's. (Think it does, just try
> suing Microsoft -- because in the past they've stolen compression
> technology and when taken to task, told the owner's "So sue." The
> thing is, MS nearly won.)
>
> And that company that owned the patent, Stack Electronics, spent $15M
> fighting MS. But the last time I checked I didn't have that much
> money in the bank and also, I am not certain that I'm going to live as
> long as that law-suit took. Patenting just isn't the right solution
> for me.
>
> And these are not figures that would interest an investor. VC's
> looking at these numbers would say the 4:1 ratio is too low, they
> believe that a 10:1 gain is necessary just to keep a VC company
> running, because most investments are flops.
Investments in random-data-compression algorithms have historically
shown a failure rate of far more than 90%, so you are correct to be
cautious.
Vic
Jim Leonard
> But about compression. Folk's it's simple: Everyone here thinks that
> it's impossible to decompose a random string into something else --
> well I have a method that does this.
> Please re-read that last sentence, that's my claim.
Nobody has a problem with that claim; we can all do that. Do you also
claim that your decomposition method also results in usable/practical
compression? That's what we have a problem with.
Jim Leonard
> And guess what -- although the court award them $160M, MS was going to
> appeal. In the end SE settled for about $60M -- as I recall. (You
> can look it up.)
You should do the same, as the court case had *nothing* to do with
compression. It was over the use of certain programming techniques
and (in the counter-suit) undocumented interface/APIs. The actual
method of compression was not the issue.
Thomas Richter
> But about compression. Folk's it's simple: Everyone here thinks that
> it's impossible to decompose a random string into something else --
> well I have a method that does this.
Me, too. It is called sorting. Guess what, the outcome is almost
completely non-random.
> Please re-read that last sentence, that's my claim.
I'm fine with that. But decomposing a random string into something
non-random is still not compression. You will find that inverting this
composition requires as much information as in the original string.
Actually, if that would be not the case, the input would hardly be random.
IOW, I see no sense in your postings here. You surely don't want to sell
people the "technology" here, and neither you participate in anything
I would call "discussion". You just make unprovable claims on an
impossible task.
So, what do you expect to gain in this forum? Keep the noise level up?
Prove successfully you're a con-artist? Attract stupid investors? Study
the psychology of the news-net?
Did I say I hold a pink elephant in my left trouser pocket? You can buy
it for $100,000 if you like. Deal?
jules Gilbert
I'm sorry -- I forgot that I have to be exceedingly literal, otherwise
you and the others don't get what I mean. Let me use your remark to
make my point;
A sorted string compresses very much better than a random string.
Thomas Richter
Yes, so we agree. However, you also need to store how to unsort it to
get back the original. And that is where your error is.
Any interest in my pink elephant, btw? It's really a bargain. I've
investors that want to pay ten times of the price at which I'm offering
it here.
So long,
Thomas
Thomas Pornin
> Any interest in my pink elephant, btw? It's really a bargain.
If the elephant fits in your trousers pocket, then I am much more
impressed by the trousers than by the elephant itself. A pink elephant
can be obtained with mundane technology, namely going to the zoo with a
few gallons of paint.
--Thomas Pornin
Ernst
Hey Jules,
I just started collecting threads with Pan news reader. For some reason
this thread was pulled today.
Anyway good luck with that. I'm looking at deconstruction of data
objects and splitting it into meta-data bit streams today.
Working in the shallow end of the pool so-to-speak.
So your ( and others ) fear of loss of reward is real.. I know that.
Just wanted to share that I had to think about sharing a codex or not
sharing a codex with everyone..
Well Since I am way over my head, with what data about data I have
access to, in using the data to the fullest advantage for data
compression I decided that a little recognition is better than none. Also
I could learn from others..
So the point of this is how does one properly publish something?
I would like to see what folks think and get some credit for presenting
an encoding system.
Does this publishing mean other will make money and I wont?
Good questions to ask.. Jules I figure you have given it some thought so
i will ask you first but welcome other replies.
What is the best path to publishing some discovery of a codex?
I'd think if I am doing this for public then some public license like the
GNU or such may be wise..
So what is suggested.. Patent it then give away the rights to use for
free?
--
01000010011001010110011101101001011011100110111001100101011100100010011101110011
00100000010101000110100001110010011001010110000101100100001000000101001001101111
01100011011010110111001100100001-*-Math:Symbolism:Logic ~ Three_Amigos-*-