Compression

Compression is the process of cataloguing pieces of data or eliminating unnecessary bits to shrink data or files so that it can be transferred or transmitted faster. The sole objective of compression program is to reduce the overall number of bits and bytes from a file or data at the point of compression and to recreate the data or file to its original size, composition or nearest form when expanded.

Different compression programs use different methods to compress data or file, some uses adaptive dictionary-based algorithm while others do not. Those that uses adaptive dictionary based algorithm often times expand back to its original size or composition and as a consequence regarded as lossless compression programs. Lossy compression programs on the other hand works differently, it simply eliminates "unnecessary" bits of information, tailoring the file so that it is smaller, and it does not reproduce files or data to its original size or composition when expanded. It mere expands it back to its nearest composition using its interpretation to recreate truncated data. Note that this may affect the quality of compressed file positively or negatively. Take for example, truncated noise in a compressed music file that expanded back without such noise makes the quality of the file better. Lossy compression is very good for compression of picture, music, and video files because colour or sound bits could be swapped or even dropped without losing the image or sound completely but is certainly not suitable for compressing software applications, databases and others that must be recreated exactly.

Lossless compression programs usually track attributes or patterns within a data set or file in a catalogue. Thereafter replaces each attribute or pattern with certain representation, number for example within the data set or file. This catalogue file is also transmitted along side the compressed data, making it easy to regenerate exactly like the original when expanded.

The crude example below would help exemplify what happens when a file or data is compressed using lossless compression program and thereafter expanded:

Before compression:

"I hate talking but talking is the best way of expression. Fortunately talking is not the only way to show expression."

Cataloguing, compression in process…

What the compression program does is to catalogue the repeated words as in:

The new look of the statements…. "I hate 1 but 1 5 2 best 3 of 4. Fortunately 1 5 not 2 only 3 to show 4."

Now, the above shall be transmitted alongside with the catalogue. The compression program will use the catalogue data along side with the transmitted data to expand the file back to its original size.

How good a compression program is depends on its file-reduction ratio which is also affected by a number of factors, including file type, file size, compression scheme etc. For example, a compression program that can reduce the size of a document by 75% and recreate same back on expansion is better than one that can only reduce same document by 50%. However, if after expansion, it is discovered that part of the content is not reproduced exactly, then no mater what its compression ratio is, such program is not suitable for document compression.