A computing research repository

The Computing Research Repository (CoRR) described by Halpern is potentially a powerful tool for researchers in computing science.In its current form, however, shortcomings exist that restrict its value and that,in the long term,might strongly undermine its usefulness.Important aspects that have insufficiently been taken care of are (1) the quality and consequently the reliability of the material stored, (2) the still restricted submission of material,which implies that other sources have to be consulted by researchers as well, (3) the still unsound financial basis of the project, and (4) the confusion that may easily arise when a preliminary version is stored in the CoRR, while a different final version is published in a journal.