While much has been made of the ocean of information available on the Internet, much less emphasis has been placed on how Web surfers might actually be able to process it. This article compares an interactive (user-inputted attribute importance weights) linear computer-assisted decision aid (CADA) format with three passive CADA formats in two studies. The decision environments for these tests are difficult, involving 20 to 30 brands rated on 6 attributes, one of which is negatively correlated with the others. Because the Linear CADA rank order is based on user-inputted attribute weights, it is expected to be more in concordance with the users preference and hence is predicted to offer higher decision quality and be better liked than the passive formats. Contrary to expectations, however, a passive Equal Weight format performed as well or better than the Linear format on all objective and subjective comparison criteria, while the other two passive formats were not significantly worse on several decision quality criteria. The implications of these findings for information providers on the Internet are also discussed.