How is no one talking about this?
>completely decentralized
>host ANY service through low latency onion routing
>nodes financially incentivized to not suck
>don't need any complex configuration to host something, just install the agent and tell your application to listen on localhost. no ports to forward or anything.
>decentralized DNS system, i.e. searx.loki
>easily transfer service to another server, just copy keyfile
Is there ANY downsides here?
This shit is awesome
Parker Reed
Other urls found in this thread:
docs.oxen.io
oxensn.com
docs.oxen.io
docs.oxen.io
loki.network
oxensn.com
twitter.com
Eli Johnson
>tell your application to listen on localhost. no ports to forward or anything.
Is that true?
Brandon Morales
Yes. Depending on your setup you may want to specify the Lokinet adapter to prevent clearnet connections. If the port is closed though localhost works fine. See
docs.oxen.io
Joshua Mitchell
Who is paying the node operators? How many nodes are there? How new is this?
Anyone trying to create an onion routing network separate from TOR are glowniggers to me, until proven otherwise.
Gavin Martin
go away jeff
Daniel Jenkins
The nodes are the service nodes for Oxen. Rewards come from staking a minimum of 15,000 Oxen (around $3.5k usd). Right now there is 1629 active nodes:
oxensn.com
Bad actors are automatically deregistered along with a steep financial requirement to spin up even one node. Low liquidity as well so if some entity were to try to abuse it, the price would skyrocket making it unfeasible.
Lokinet has been around for a bit now, since 2018 I think but only recently usable. I'd recommend reading through their network infrastructure doc for their flagship app Session here:
docs.oxen.io
More info on how the network works on the same site.
Sebastian Harris
>Rewards come from staking a minimum of 15,000 Oxen
Is there a use for the OXEN tokens as more than just a staking reward? Do I need to pay OXEN to send data through nodes? If the price for the tokens is fully determined by node operators buying them for staking rewards, then it's extremely unstable.
Julian Carter
No cost for sending data through the network. OXEN is burnt when registering a custom ONS name or custom Session ID. OXEN is originally a fork of Monero so it works as a private currency as well. Similar emission to Monero, etc. Node operators don't receive direct payment, more Proof of Stake consistent rewards for securing the blockchain while also acting as "Service Nodes" that allow for onion routing, name service, etc
Evan Campbell
>No cost for sending data through the network
That seems unsustainable to me, but I'll keep an eye on it and see if it goes anywhere.
William Young
TOR also has no cost as well, right? Along with being hosted by volunteers that you have to hope aren't attempting to deanonymize you. Having an incentive structure along with a barrier to starting a node sounds like a good idea to me. I would agree if you want it "proven" first though.
Connor Walker
Forgot to mention that being able to route and TCP/UDP connection through Lokinet seems to be unique. I could be wrong but is there any other mixnet that allows for that? Along with the ease of setup and being able to use human rememberable domains. Seems like these features are pretty big.
Ryan Perry
>TOR also has no cost as well, right?
It's also slow and a lot of the nodes are unironically run by government agencies.
The problem with lokinet is that it's trying to 'be a better tor', but it doesn't actually offer anything sustainable. If node operators are in it for profit, and the only token payment is for registering domain names or session IDs, then I would expect it to fail once the hype runs out because the price of the token is not based on the real work done by node operators, it's based on an artifical limitation that you just have to hope will cover the operating costs and profits of the entire network.
Robert Cook
you have to pay $3500 to even start a node? all this means is it raises the barrier of entry to non-glowniggers. if the NSA became interested the price to entry is nothing
Andrew Kelly
Operating costs for nodes is pretty low:
docs.oxen.io
Any $10/month shitter VPS can run one according to their requirements. Quick look at the tokenomics doc they appear to be banking on most node runners being in for the long run:
loki.network
I need sleep but will actually read the entire thing tomorrow. I personally think the network itself offers a few key advantages that make it pretty cool. If the token dies I would imagine the devs would either remove the tokenomics shit and go similar to TOR, or there would be a fork that would. Again we'll have to see if people end up caring.
Very low liquidity market. The price would quickly become astronomical if glowies could even find enough OXEN on exchanges to open more than 10 nodes.
Ryan Hughes
>Any $10/month shitter VPS can run one
You can run the software on a shitty VPS, but if you want to quickly route traffic, you need gigabit/10gig and a decent amount of CPU power.
If there's no financial difference for me running a node with 1mb/s and a raspberry pi versus a 10gb epyc/xeon server, then I'll use the raspberry pi, because it's more profitable.
>Very low liquidity market. The price would quickly become astronomical if glowies could even find enough OXEN on exchanges to open more than 10 nodes.
The other side of this coin is that the network is already fairly centralized.
Lincoln Cox
I believe node that fall behind/are too slow are automatically deregistered. As for network req, nearly every provider offers a minimum of a shared 1G link. I've gone through ~10 different providers and even the shittiest would consistently get me 600mbps up/down any time of the day.
Other side of the coin is true to a degree. There seems to be a pretty even spread of contributions according to oxensn.com
I would think the low liquidity in this case keeps the network from becoming more centralized from large buy-ins assuming these nodes are actually run by different people. Network is pretty young though so this is to be expected to some extent until more people become interested.
Henry Roberts
>make running a node expensive
>only feds can afford to run nodes now
kek, genius
Nolan Smith
Sybil resistance isn't that great of an advantage
Tor has had problems with sybil in the past (KAX17, glowies,etc.) but as far as i know no tor user or hidden service has ever been compromised because of sybil. its all because of opsec failures, timing correlations, etc.
With the addition of congestion control just recently and the upcoming spec on conflux traffic splitting tor stands to gain huge increases in speed too which makes their low latency point kind of moot, although still relevant right now
The only advantage I really see right now is the DNS system
Finally, this platform hasn't been vetted as much as tor has. Tell me once the first darknet market pops up on it
Jaxson Martin
>As for network req, nearly every provider offers a minimum of a shared 1G link.
So then I'd only run 1G instead of 10G, the principle is still the same. The value of the token is not related to the work done by the network.
>There seems to be a pretty even spread of contributions according to
And what if a big wave of random people got interested in it and wanted to buy into it? If there's not enough liquidity for 10 nodes, then the people running the nodes now will be the only ones running it for a long time.
Eli Morris
A good way to actually have what lokinet is trying to provide is to develop a system on top of XMR to pay providers to use nodes to onion route traffic through, and just pay by the traffic you use.
Charles Parker
fuck off pedo cuck