On Thursday, Apr 13, 2000, Oskar Sandberg writes: >On Thu, 13 Apr 2000, Lee Daniel Crocker wrote: >> Here's a possible way to make "Depth" useful: have the originator of >> a Request set it to a small random value (say 1-10). On each forward >> hop of the request, Depth is incremented as HopsToLive is decremented-- >> but unlike HopsToLive, it is not further incremented on subsequent message >> forwards in response to RequestFailed's, so it really does represent the >> current distance from origin of the request (+ the random starting >> value). Replies, then, can use it to set HopsToLive to an efficient value >> (and replies themselves will not have a Depth). Will this work the way I >> think it does, or am I missing something? > >This is exactly what it does currently (right down to the initial random value >and not being incremented when messages bounce). What did you think it did?
I was unware this was the way things worked, and am a bit puzzled as to why Replies need to search their way back to the originator; can't each node track received/sent RequestId's internally so that replies to a certain request need only get sent back to the node that requested them? Is there a benefit to NOT doing so? The benefit to doing so is quite clearly a reduction of bandwidth consumed... plus each node only then needs to know that one nearby node requested the key; any HTL count is valid, and no Depth field is needed. --pj _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
