Thanks for the heads up on the site. Most journals are looking for exclusive papers and as https://arxiv.org is an electronic archive and distribution server for research articles this would equate to us publishing the paper, making the Parsec paper no longer exclusive.
EDIT: Also wanted to make it clear that we will ensure the paper if published by the journal will be open access.
Broken link now fixed, thanks for flagging 
Maybe I didnāt explain well. Most reputable journals do not consider preprints prior publication Preprint - Wikipedia. Definitely check the specific journal policy, but the norm is that itās allowable. I donāt mind whichever direction you guys go with this; just want to make sure you have all the facts. Will leave this alone now.
Yes we checked with our academic partner and they advised that it can only be published once the paper has been reviewed by the journal and accepted. So typically the pre-print is the final peer reviewed version. They confirmed that if the publisher detects that the full paper has been released somewhere else, theyāll drop it. I do appreciate you following up though!
@Nigel eventually, i suspect it will. Holepunching is sort of orthogonal to the protocol used. Currently the group is enforcing that nodes are externally reachable directly and hence minimising the need for tunnels etc. With Holepunching peers donāt need to be so stringently checked for ext-reachability.
QUIC OTOH is a protocol atop UDP. With TCP the main prob is that each connection is a socket consumed. With UDP we can do multiplexing atop a single socket - which is what QUIC/quinn does so we are likely not going to run into socket-descriptor limit problem on certain OS (notably linux as default in many distros is 1024 out of which many are consumed by open files etc - so much lesser available. Can be increased by commands ofc).
Then QUIC has multistreaming which prevents head of queue blocking and TLS inbuilt so encryption on the wire for free. Also more features (keepalives, IP migration etc.) as mentioned in dev updates. All this reduces maintenance in Crust and is done by a library which follows an IETF spec so much better than hand rolled boilerplate to warp TCP into doing these which is unnatural for it.
So if all works out (and that is what we are testing) then itāll allow huge section sizes for routing which is very helpful for security ofc.
Great work again safe team! Gathering a ton of momentum. pySafe has also progressed a ton, it is now a matter of a week or two before you all can play with v0.1. Riddimās on holiday, but I managed to spend a couple dozen hours architecting and implementing the library. The ultimate goal of having it almost exactly mirror the Java binding semantics is in sight. Demo apps are being coded and tested, among them were to be a CLI-like shell and a simple RPC server. Very excited to have more components in the ecosystem.
This is exciting news. I think adding Python is gong to have a big impact because it is both popular already, and an easy language for new coders.
Also, Zim Wiki, the app I use most of all is written in Python, and is what I use to manage and record all my work - so maybe one day it will be ported to SAFE! And Iām sure there are lots of others.
I think it will round out the ecosystem really well. Maybe not as webpage friendly as JS, but early experiments with the threading model show that it is possible to sub out our implementation with something like Celery, meaning it will be pluggable into flask. Other applications like pollers, clearnet bridges, mail (or any other app) notifications will be a strength. A lot of thought has gone into how to enable extensibility, but weāll be leaning on the community soon to help us identify these use cases and get the required functionality right (as well as test test test!).
Question of when not if now 
Better get that blue and white python logo ready!
:). Yippieeeeee
I donāt know what this means - it would be great to have a Python oriented topic or two, and I expect you are planning this already given the following, so I canāt wait! 
NodeJs, Web, Android, .NEt, Xamarin and soon python. These technologies have their own communities. What happens when you turn up to a community and give them a new toy?
In see stellar adoption in the near future.
So exciting.
Edit: Now this is a versatile graphic, poster, online advert, Twitter post, Facebook status post. This graphic is clean and professional, puts accross itās message well.
Should be placed up on university computer science notice boards.
āThe distributed internet needs youā
I think heās talking about making some soup. Sounds tasty!
Will definitely do once we are there. Sorry, I was just trying to bash in a reply before a Friday afterwork pint and enthusiasm got the better of meā¦ I simply meant that one of the most used python web frameworks (flask) can be hooked into the library and work well asynchronously. Incidentally, Celery (the job/queue library) is also used in many distributed applications, so future is bright!
Fantastic update team! Lots of good stuff happening here. Two thumbs up.
Ironically, the Rust logo is also missing.
I already asked this question in December: Is there a place where the way to use the authenticator in a rust program is explained? I didnāt get any answer on this specific question (but it was part of a much longer post).
Hi @tfa, is the auh CLI of help to that particular question? or you meant a Rust app which needs to authorise against the authenticator (either auth CLI or the one in browser)?
will this delay of publishing the whitepaper (up to 1 year) caused by the academic review cause any delays in the relase of the SAFEnet?
No it wonāt, any delays in publishing wonāt impact the release of code.
