🚀 go-pugleaf

In the GNU bulletin I see that a replacement for compress
is wanted.  I have studied some data compression techniques,
mostly variants of lz and huffman + some aritmetic coding.
I have even written a program that compress better than
compress, although it is too slow to be of much use.
(It even compressed files compressed by compress some 1-3%)

.....

Speed, compression ratio, non-patented, how good must it be?

Is a program with about the same capabilities as compress,
but without any known (to me) patents good enough?

I think a variation on huffman coding could be both fast
and good enough, but how many variations on huffman coding
are patented?

Urban Koistinen - md85-epi@nada.kth.se
Current writer of GNU Chess 5.0 - it will be great!
-- 
Urban Koistinen - md85-epi@nada.kth.se

[mod- I was under the impression that Huffman coding was not patents,
but dont quote me on this...]