The beginnings of Onion Routing: I need an explanation about this 1996 chart

Question

The graph above comes from the document "HIding Router Information", used to formally present the concept of 'Onion Routing' at a convention in Cambridge in 1996.

I am analysing this graph for university research and have a number of questions:

1. At the dawn of 'Onion Routing', was the layered encryption similar to the one Tor uses today, or has there been any substantial change?

2. In this graph, is the information between the client (Initiator Machine) and the first node (W) encrypted, or is it the first node that encrypts the information? Who currently initiates layered encryption on the Tor network, the first node, or Tor Browser from the client device?

3. The graph says that the first node (W) is controlled by "Secure Site". I have read the document "Hiding Router Information" and I am not sure what "Secure Site" is. Does anyone know?

Sorry about the writing. But English is not my mother tongue.

Answer 1

I found the paper you reference. It was written in 1996 but the first version of Tor was written in 2002. This was just a theoretical precursor that was written years before the actual Tor project began. It doesn't have much relation to the actual Tor network today.

From the paper:

Figure 1 illustrates the topology of an Onion Routing network with five nodes, one of which (W) is the Proxy/Routing node for the initiator's site.

• (W) is what we now call a Guard Node in Tor.
• "A secured site" Is protected by a firewall/intranet. Remember this is 1996 and before the widespread advent of https.
• (Z) is an exit note
• (U), (Y), (X) are middle nodes. In Tor today there are only 3 notes total between you and the internet site that you are trying to connect to.

You can find a much better paper detailing the modern Tor project here. You can also write to one of the authors for better answers. Paul Syverson's website with contact information is here.