Summary

• Compression is an application of information theory

• Encodes long strings as shorter strings

• All schemes depend on redundant / repetitive data

• Common gotchas

• A compression algorithm cannot compress its own output
• Cannot easily compress completely random data
• No / little redundancy