There is growing evidence regarding the importance of spike timing in neural information processing, with even a small number of spikes carrying information, but computational models lag significantly behind those for rate coding.  In this paper, we propose and investigate a minimalistic abstraction, using a reservoir model, for extracting information from spike timing by exploiting variations in axonal delays. Our model encodes input patterns into a sparse neural code, translating the polychronous groups introduced by Izhikevich into codewords on which we can perform standard vector operations.  The map from input patterns to codewords appears to be chaotic, and, for the appropriate choice of parameters, the distance properties of the code are similar to those for (good) random codes. Thus, the proposed model provides a robust memory for timing patterns.