The don't use the same network, and never have. Jami uses OpenDHT (https://jami.net/ces-2019/), GNUnet uses its own DHT (see for example https://grothoff.org/christian/nss2011.pdf, or the rather hacky (and not complete) bib page: https://bib.gnunet.org/#DHT)

Would this help? https://docs.gnunet.org/#Introduction

If anything could be made more clear, let us know as feedback.

We should maybe include a "GNUnet in 5 minutes or less" on the website. The old LEGO analogy is nice, and lynX did a great comparison here: https://secushare.org/img/stack-comparison.png We don't really need the ISO OSI model here, but even this can be added, as we've already used it throughout the years to compare to the GNUnet stack.

I think somewhere we mentioned before that gnunet does not require ISPs and IXPs, but using them today is the de-facto default state.

May I draw your attention to my old comment here, as well as (more importantly), the entire thread of even older comments /it/ links to:

https://news.ycombinator.com/item?id=19959687

Which I shall also link to directly, but the above also has some relevance:

https://news.ycombinator.com/item?id=19864808

See especially here:

https://news.ycombinator.com/item?id=19867532

And here:

https://news.ycombinator.com/item?id=19867467

Please give me a summary of what you intend to express here. I just don't have the time to cherry-pick the information here.

A short addition: (1) various forms of onion routing are in active research. (1.1) see for example the "big data, little data, no such data" presentation of Christian Grothoff from 2017

(2) what we are still trying to solve here, mainly, is making it harder to know the path of communication. Today, if User Lora with Node A talks to User Charly on Node B, the data goes through N nodes. The N nodes between A and B are aware of (parts of) the path, but the communication remains encrypted (in very short terms explained).

(edit: no markup here)

> but they used 'replace'. Them's fightin' words.

You are welcome to communicate feedback to our (GNUnet) mailinglists. We are in the process of replacing the website (Drupal is good, but doesn't get the job done anymore for us), so if there are problems (even if just interpretation problems or tiny things that throw you off) I invite you to send them to the appropriate mailinglist listed at https://gnunet.org/contact_information

We appreciate any kind of feedback.

I've read into this again. 'Replacing' will be clear from context. This was added very recently (2016?) to the homepage by myself and to the official description.

The Manual (the pushed content as well as my stacks of old fashioned paper corrections) is not mentioning it. For the new website it (the description) either will be described as it is right now (unlikely) or rewritten to be easier to understand.