If you know what link you're going to follow from an intermediate document, you do not need to download the intermediate document. You can instead save time and bandwidth by using the FOLLOW command.
Suppose http://www.healthnet.org/programs/ looks like this:
Some important conferences carried by SatelLife include: * ProMED: Program for Monitoring Emerging Diseases[1] * ProCAARE: Program for Collaboration Against AIDS[2] * E-Drug: Essential Drugs[3] ** References from this document ** [1] http://www.healthnet.org/programs/promed.html [2] http://www.healthnet.org/programs/procaare.html [3] http://www.healthnet.org/programs/edrug.html
If you send a message to the GetWeb server with the text:
GET http://www.healthnet.org/programs/ FOLLOW 2
The server will send you the description of ProCAARE.
You can chain together multiple FOLLOW directives. If you already know you're going to want the fifth document linked to the ProCAARE page, send a message like this:
GET http://www.healthnet.org/programs/ FOLLOW 2 FOLLOW 5
This would reduce the total turnaround time, since you would get the desired final document in one e-mail round trip rather than in three.