The performance of CBIR algorithms is usually measured on an isolated workstation. In a real-world environment the algorithms would only constitute a minor component among the many interacting components. The Internet dramati-cally changes many of the usual assumptions about measuring CBIR performance. Any CBIR benchmark should be designed from a networked systems standpoint. These benchmarks typically introduce communication overhead because the real systems they model are distributed applications. We present our implementation of a client/server benchmark called BIRDS-I to measure image retrieval performance over the Internet. It has been designed with the trend toward the use of small personalized wireless systems in mind. Web-based CBIR implies the use of heteroge-neous image sets, imposing certain constraints on how the images are organized and the type of performance metrics applicable. BIRDS-I only requires controlled human intervention for the compilation of the image collection and none for the generation of ground truth in the measurement of retrieval accuracy. Benchmark image collections need to be evolved incrementally toward the storage of millions of images and that scaleup can only be achieved through the use of computer-aided compilation. Finally, our scoring metric introduces a tightly optimized image-ranking window.