What data are we actually allowed to send from station to station? If we start at StationA, can StationA send its UDP hostname and port to the neighboring stations so that overtime a list of visited stations can be built up? Or are we only allowed to send station names with no networking info along what so ever.

ANONYMOUS wrote: Stations can only communicate with their neighbours, using UDP and their port number provided on the command-line. A station is not permitted to know the hostname/port of any non-neighbouring station, so that information should not be sent over the network. You can send station names in your 'outgoing' queries, and station names will be returned in replies to those queries. I hope that's clear?

But what's the issue in simply sending only the original station's network information into the datagram that's been sent out? If once a response is found, that station can simply just use that information and directly send the response back to the original station by sending just a single UDP response without the need to backtrack through all the stations its visited which would essentially increase the chance of it being lost as well the time the original station to get the response. I don't really understand why it shouldn't be allowed.

ANONYMOUS wrote: Because the station that determines the best result, possibly by receiving and comparing multiple responses, might not be a neighbour of the original station that started the query. There are many possible ways to implement this application; our project specifies that only neighbours may communicate with each other.