Timing the search reply
July 10, 2007 Leave a comment
When you are searching, your client send out a request and the other clients in the hub replies (if there’s anything to reply). As you’ve noticed, the time for reply #1 and reply #20 can differ.
When people see the time difference, they feel the need for a “reply time column” in the search window. Basically, the column would display the amount of time it took for your client to recieve the reply. (This is commonly named “network latency” or “ping time”.)
The idea is that the reply time would reflect on the users’ own network speeds. The sooner the reply, the faster the peer. What most people don’t know is that this would be completely pointless and wrong. Why? Because the way the reply time is calculated;
Latency is directly related to the physical distance that data travels. Thus the time taken for a packet to travel from a computer server in Europe to a client in the same region is likely to be shorter than a the time to travel from Europe to the Americas or Asia.
Adding the column isn’t going to magically allow you to see who is the faster source. It will only add a false sense of speed. (Yes, one might say this, too, about the “connection type”.)
(No, this has nothing to do with ADC or NMDC.)