Re: [dtn-interest] DTN without regions - I think I get it

It seems we are reaching agreement that end-point identifiers will be very general beasts:

>> Yes. In a sense, the "schema" takes the place of the "region" in

some of them will contain structure, some not, etc. Although I initially argued for a distinction between addresses and identifiers, I can see the merit of a more generalized scheme (after much convincing done by Mike and Kevin).

However, if I could bring back another point that I was pushing for earlier... I still believe we need to support an "intermediate-destination" header, which of course is similar to loose source routing, except that it is change potentially many ways along the way to the final destination.

The reason an intermediate-destination header is needed is that some schemas will not have routing entries for every EID in every node. Let's say node A somehow (dht, out-of-band, guess) knows that EID pgp://xxxxxxxx is connected to the earth system internet. However, other nodes on the path from A to pgp://xxxxxxxx, that are not internet connected, are not aware of this. A could then address its outgoing bundle thus:

destination: pgp://<key here>
intermediate-destination: region://in.earth.sol

Note that we cannot simply say destination = region://in.earth.sol, because this is not specific enough. However, if we keep the pgp;// destination, the first node connected to the internet will be able to determine a new intermediate destination for the same pgp:// identifier, perhaps something like ip://128.15.23.92.

Do you need help?
In addition to our free technical library, Pantek provides professional IT services to companies who utilize Open Source software: Learn more about Pantek's Technical Support offerings Place an order for 24/7/365 Technical Support here Learn more about Pantek, who we are, and what we do Chat with a Live Operator Related mailman.dtnrg.org Links dtn-interest Request more information about Pantek

Note that this is different from the "care-of" address that Scott suggested, or the "reply-to" address that has also been suggested earlier. The intermediate-destination merely states that "this is where the packet is going right now". Once the packet gets there, it may very well be replaced by some new intermediate destination.

• Jakob

  ---