If it hits enough three-bit runs in the data it might be okay though =)
three bit runs have maybe 1 in four odds... so 6->8 is probably pretty typical..? long runs drop n/3 bits rounded down =)
Name:
Anonymous2014-01-20 22:44
It can compress 3 bytes to two if its 1x 24bit run or 8x 3bit runs... which is, uh, 4 / 2^24
Name:
Anonymous2014-01-20 22:49
oh, and combinations of 3-bit multiples ^^ so a few more
Name:
Anonymous2014-01-20 23:15
Where can we mere mortals obtain this software?
Name:
Anonymous2014-01-20 23:20
And you can just squeeze in a single secondary control sequence in 2-bits, but it can't follow a run =P
What's the best use for control structures?
Markers and repeats?
Name:
Anonymous2014-01-21 1:59
And how to find the longest repeating sequence of bitcode?
or most repeated sequence?
Name:
Anonymous2014-01-21 2:49
>>1
Whatever the scheme short sequences of bits cannot be compressed in any way. Common sense.
There is not way to compress 2 bits to 1, 3 to 2, 4 to 3, but as bit count increases there is a larger chance to compress a single bit/byte. Taking a 4GB file there is very high chance some form of compression will compress it to 4GB-1byte.
>>18 Taking a 4GB file there is very high chance some form of compression will compress it to 4GB-1byte.
There are 2^2^32 different 4GB files. There are 2^(2^32 - 1) different 4GB-1byte files. Therefore no more than half of 4GB files can be compressed by 1 byte (and even in that case you have to do something with all 4GB-1byte files). Note that the file length is actually irrelevant.
>>18
There is no way to compress/fit all eight 3-bit combinations in the four 2-bit spaces..
But actually, in a four gig file there might be fairly long repeated sequences, random or not... then it's just how efficiently you can find, mark and repeat them =)
Name:
Anonymous2014-01-21 8:16
I think this is going to be able to repeat ambiguous sequences too ^^ so that'll be interesting
Another source described as a current NSA analyst said: ‘In a world where I would not be restricted from killing an American, I personally would go and kill him myself.’
What about the other 96% ?
Name:
Anonymous2014-01-21 20:26
Such a typical american attitude...
The only reason that some american can't kill snowden is that He is American too!
Name:
Anonymous2014-01-21 23:46
Burn ^^
Name:
Anonymous2014-01-21 23:48
>>25
Another source?
WHAT WAS THE FIRST SOURCE?
And what was this source?
Holy shit, man.
Name:
Anonymous2014-01-22 0:00
le pentagon sage Everywhere!
Name:
Anonymous2014-01-22 0:38
Hmm...
there seems to be a little issue with ambiguous control structures..
And probably dorks trying to patent other people's things before they even work xD
Name:
Anonymous2014-01-22 0:45
Silly dregs...
This ain't the jewish olympics ^^
Name:
Anonymous2014-01-22 3:23
So, anyway, spec 0.2
First two bits of a chain (bitcode) determine number of true/on bits in the first three spaces (data), and if they are all zero (+0) or all one (+3), the chain terminates with instant 3->2 bit compression..
If not [ie, one or two true/on bits], then the sliding window kicks in, encoding -1/+0/+1 changes in the window sum.. Only the +0 code is properly Ambiguous, -1/+1 codes always disambiguate (up/down encoding), and can also terminate a chain by reaching either +0/+3 window sum..
Since chained bitcodes only use 3/4 of the two-bit space, there is room for a single control sequence, but it needs to follow either a +1 or +2, so we get 2x 4-bit control sequences. That's probably still not going to be enough to be useful, so i think i'm going with another extensible 4-bits for mark & repeat address codes, bringing the control structures up to 1-byte per piece, or 3 bytes minimum for a mark/mark/repeat set.
To be effective, data being repeated only once needs to be longer than 3 bytes, but if repeated multiple times it could be as short as 10 bits.
However, three of the five panel members concluded that the NSA spying programme "lacks a viable legal foundation" under the Patriot Act.
It "represents an unsustainable attempt to shoehorn a pre-existing surveillance programme into the text of a statute with which it is not compatible", they said.
The programme also raised constitutional concerns, including "serious threats to privacy and civil liberties as a policy matter, and has shown only limited value".
"As a result, the board recommends that the government end the programme," said the report.
However two panel members - both lawyers from the Bush administration - strongly dissented, saying the issue of legality should be left to the courts to decide.
One, Rachel Brand, also argued that declaring the process illegal could affect the morale of intelligence agencies and make them overly cautious.
saying the issue of legality should be left to the courts to decide.
But declaring the process legal could affect the morality of intelligence agencies and make them even more overly ambitious.
Name:
Anonymous2014-01-24 0:52
How can the courts decide if something is legal without knowing what it is, in it's entirety?
And if the public disagrees, does that make the public illegitimate, or the other?