Although this is off-topic, I thought I'd forward it because (1) I like to read what Noel Chiappa writes, and (2) this is one of those IETF threads where people discuss how things might have gone differently if certain design and implementation decisions had been made differently. Date: Wed, 12 Apr 2000 14:01:37 -0400 (EDT) To: [EMAIL PROTECTED] Subject: Re: interception proxies From: "Dick St.Peters" <[EMAIL PROTECTED]> Charles Lynn writes: > You have misread the specification. That "source" is not (IP Header) > "Source Address" that you imply, but the "source" mentioned earlier in > the sentence, i.e., "address in the source route" (option). Apparently so - though given that the part I emphasized immediately follows mention of rewriting the destination address, this isn't obvious. > [If you think about it, the interpretation you emphasized above will > not result in a capability for bi-directional communication.] Actually, I've always thought that the first recorded-route address was the original source address so the route would indeed be reversible, but I'll admit to never having actually seen a recorded route. J. Noel Chiappa writes: > Pardon me if I emit a "balderdash". ... > They had great foresight, huh? The foresight to see the need for more > address bits (funnily enough, IP2.5 *had* this, and they *ripped it out* - > great vision there), traffic flows, etc (I could go on but what's the > point). Yes, just what was your point? > However, let's not mythologize anything, OK? It gets in the way of > objective analysis. Actually, the view that everything about IP was cast in stone forever in the IP spec is exactly the view that is being argued about. Would you settle for "The IP spec authors didn't have enough foresight to foresee a need to rewrite source addresses" ? :) Whatever anyone thinks of it, people are doing it. On the right are people saying it is immoral, evil, and dangerous, not to mention prohibited by the gods, and they refuse to talk about it. On the left are people doing it, each their own way because there is no standard and not even any public discussion. Only a few months after I first used the net, IP replaced NCP, giving me an instant impression that network protocols are ephemeral things, replaceable if they don't measure up. Presumably if they can be replaced, they can be adapted when necessary. -- Dick St.Peters, [EMAIL PROTECTED]