It's quite common that the algorithm itself is the trivial part in an algorithms paper, and the real contribution is the analysis. If you have a sequence of sy

It's quite common that the algorithm itself is the trivial part in an algorithms paper, and the real contribution is the analysis. If you have a sequence of symbols emitted by a known source, it's often easy to compress the sequence optimally. But if the properties of the source are unknown, optimal compression becomes much harder. The key part of the LZ77 paper was showing that a simple algorithm is asymptotically optimal for a wide range of sources. That the algorithm is a universal compressor rather than a mere heuristic that seems to work pretty well with some test inputs.