All of electronic media is, in some sense, data that is reduced from an original copy. It begins with the capturing of the event. That captured moment or series of moments (e.g. sound, still image or motion picture) now exists in a portable format that is inherently limited in scope, function and capacity. The sample has now been encoded into data, which will be stored or transmitted over some sort of electronic medium. Although not exclusively, it is usually the case that the initial encode will need to be converted (or transcoded) to one or more additional formats, within certain parameters tailored to given storage and/or distribution requirements. This is where file-compression comes in.
File-compression’s main objective is to optimize storage and bandwidth as much as possible by breaking down files and/or eliminating their unnecessary characteristics. Two primary classifications of file-compression are lossless and lossy, which differ in a number of ways (including size, structure and performance). Lossless compression creates an exact copy of the original file when decoded, while lossy compression makes certain sacrifices to render a deliverable but slightly compromised output. Each classification encases a number of different encoding schemes that process data in several different ways, balancing fidelity and efficiency for functional and presentable data on the egress end.
A common implementation of lossless file-compression includes the use of Huffman coding, whose redundancy-limiting algorithm recognizes patterns in groups in order to conserve time, space and other resources. The model is able to compress and decompress digital media such that the output perfectly matches the input. The Zip file archival tool is a well-known format that supports lossless compression. Zip files compress digital media into much smaller data (often given the ‘.zip’ file extension) that can be uncompressed into their original form, with no loss or difference between the ingress and egress content (hence, the compression is lossless).
Lossy compression employs encoding schemes that sacrifice data and/or limit parameters in order to conserve space and optimize file-transmission. Transform coding basically approximates which data can and cannot be sacrificed based on the prospective delivery requirements, and is very common in lossy compression. One of the most well-known lossy formats is MP3 audio, which is part of MPEG (Motion Picture Experts Group) standardization and has had a very profound effect on modern audio. It is a very common delivery format for consumer audio, and many other formats are quite often converted to MP3.