When comparing two infrared spectra it seems like it would be good to have a numerical value for the comparison. This value could be used to rank a series of comparisons. Or the value might also be used to give a "confidence" to the match. Infrared search software often has a numerical ranking of the sample spectrum and a library spectrum but the algorithm used for that ranking might be obscure and not easily described to others interested in the results. And it is difficult to compare two sample spectra. Pure compounds usually do not present a problem but complex mixtures can be difficult determine a comparison.
Through this discussion I would like to see what thoughts are within the ACS community about comparing infrared spectra.