Ian Clarke <I.Clarke at dynamicblue.com> wrote: > The problem with your's/Oskar's proposal is that I don't see how the > updates will be propogated to a sufficient number of nodes so that even > where requests can "slip through" nodes where they would normally be > answered, they eventually get to the updated version of the data. In > fact, unless I am missing something, this isn't really addressed by > Oskar's proposal at all! This is the issue that my idea addresses, and > moreover, I can't think of any more sensible way to do it.
Well, updates get propagated in more or less the same way that inserts do. If we accept the premise that performing an insert to only 10-15 nodes is sufficient to propagate data (in conjunction with the spreading performed by requests), then why shouldn't we accept the premise that updating 10-15 nodes is sufficient as well? (in conjunction with the spreading performed by follow-through requests) Perhaps the difference is that you want data to be updated everywhere as quickly as possible to minimize the number of out-of-date copies. However, no mechanism can provide even approximate guarantees that you will not get an out-of-date copy. Faced with that reality, it seems more in the Freenet "style" to allow updates to diffuse naturally with requests (Oskar's proposal) than to try to force the updates out everywhere (your proposal). Although updates may propagate a little more slowly, I don't think it will make much of a difference, and Oskar's proposal is certainly more consistent with the rest of Freenet, and uses fewer messages. theo _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
