[2] We will elaborate on this further in the next section.

[3]Envoloping refers to a technique where packets of one type are packaged in the packets of other type for transmission purposes; for example, at present the MBONE packets which are normally in UDP format are usually enveloped in TCP/IP disguise.

[4]We address this relationship a little later in this section.

[5] There may be more priority classes than users are allowed to use in order to perform important network managemnt tasks, which may indeed take precedence over any other usage.

[6]This assumption is used to simplify the model, however, the results presented here can easily be extended to the cases where the delay is suffered at multiple nodes as shown in Gupta, Stahl, and Whinston (1995).

[7]From the users' perspective, in reality, the bottleneck is either the server's pipeline or the slowest data communication link in their path to the server.

[8]Note that some users might decide not to get the service because of the excessive delays, however, users with negligible delay costs will try to obtain the service regardless of the delays. Thus, with no pricing mechanism the services can potentially be accessed by only the users who value it the least.

[9]Realistically, this work would be done by a smart agent executing on a client on the users machine. We discuss this and more implementation issues later in the paper.

[10] Note that the perfect waiting time information scenario case is the "best-case" scenario for our implementation of the free access policy because users first check where can they get the fastest service and the information they get is exact.

[11] net benefits = total value - total delay cost.

[12]customer benefits = net benefits - total price. Note that with free access, net benefits = customer benefits.

[**]A version of this paper is available in HTML form on the World Wide Web server of Center for Information Systems Management, University of Texas at Austin. Click here to get this paper (long).