They always were going to. Its how tcp/ip protocol works. And the internet is built using these as the method to talk. Cannot get away from that without building a whole new network, hardware and all.
I agree that early on especially that with just payments for uploading records (includes chunks) is not enough incentive. If the network was at a more mature state then the pricing algorithm would hopefully have the cost to upload a record at a high enough level to definitely compensate a node runner for their efforts. The incentive will lie in other areas, as others have suggested before. Some will even run nodes for no compensation.
How the node fills is along the lines of records being placed in the nodes closest to its address at the time of upload. Due to the address of a record being based on the hashing algorithm the address will end up approximating random distribution for all records, ie even spread. But as the network gains new nodes and loses others the closest nodes to a record will also change and the record gets moved to the closest nodes.
This means that today your node might receive 5 records but in a week/month’s time due to the changing layout of the network with new nodes joining and ones leaving those records will no longer belong in your node and thus not be stored there. This also approximates random moving since node addresses is also following a hashing algorithm that causes that.
So over time your node will have received payment for records that no long are stored in your node and your node in turn will be holding records it never received payment for. Every node has this happening over time so is relatively fair. Each node in theory should end up being paid approximately the same amount per period (week/month/year) that it is running.
Also the records it holds will be approximately the same as any other node due to the moving around of records on a double random system of addressing.
So from that you node will never fill up (unless network fills up) and will continue having new records uploaded to it and receive some payments.
As to how many new records (w/payment) each node gets is unknown at this time and is very much dependent on both number of nodes in the network and how much new data is being uploaded.
Thank you, I think I finally get the economics now, I hadn’t considered that my node could actually reduce what it holds over time as more nodes join! I guess this is kinda analogous to proof-of-stake crypto ‘mining’, I hadn’t thought of it this way before.
BTW I understand IPs are necessarily exposed in order to communicate, my point was only that because of this many people will still want to use VPNs when running nodes. I want to know how secure running nodes is. Not requiring a VPN is preferable (and necessary if they are going to implement geo/IP blocking), so I am wondering if they are going to implement sufficient onion routing to compensate. The security aspects of the network elude me.
For me the issue is not so much my IP address is known to other nodes, but more the ISP is seeing so much traffic to random IP addresses. Whatever a node can do with thousands or 10’s/100’s of thousands of IP addresses I am not sure. Track me, no good unless they expect to run such a large number of nodes. And then what would this tracking achieve, they’d know I run nodes and that is about it.
No matter the network if its just attacking IP addresses that connected to one of their nodes, then they might as well just randomly pick IP addresses anyhow. You know like port scanners do anyhow
The encryption ensures that anybody watching my traffic is unaware of what it is and its the same for other nodes, they do not know. They do not know what is in the records/chunks either that is stored in their nodes.
From all that i can determine from what is said the idea is not geo-blocking, but nodes deciding to have a good mix of geo-locations in their routing table. Not with the intent of blocking, but getting a good mix.
What that will do for mega node runners in a data centre is to mean that in an overall sense their nodes will have poor connectivity since instead of another node out there potentially having 5 or 10 of the mega nodes runner’s nodes in the routing table there might be only 1 or 2. When that happens across the network their nodes will on average be under represented in the network and as a result the network will favour the smaller node runner who may have a few hundred nodes. Having a few hundred nodes it would be extremely rare for any other node out there having more than one of those nodes in their routing table.
If you were a leaker/whistle blower you will still want your ip address hidden (plausible deniability). There are other legitimate use cases for wanting good anonymity online, however VPNs are useful to get around ISPs - and as you point out they could start blocking node runners with so many ip connections . But if VPNs mess up routing tables on the network what else is to be done?
Reading this @blvd, you may be correct that emissions are being suspended rather than coming to an end. You can read the earlier text the other way though so
The wording suggests this will be clarified on Feb 19th. If not it will be worth asking then.
At least we can be sure of a few months to test a network without emissions which will be useful IMO, and may help make a better distribution system going forward whatever the Feb 19th plan says.
agree with Josh but also Blvd I guess. Reading the announcement it definitely leans more towards it returning in 2.0, however, talking with the team on discord indicates they are looking to remove it completely but will probably want to see the effect and work out a new plan before locking themselves in.
Nodes from an ISP viewpoint would be something that talks to a lot of other computers in a seemingly coordinated way. They will be unlikely to know what they are for though.
No matter how the network was going to operate this could never be guaranteed. It was more that in the jungle of traffic it becomes extremely difficult to discern what is what and who did what,
As to the whistle blower, then who, which node, which uploader, uploaded what. Its all encrypted and no way to trace back to the computer that uploaded what when there is plenty of people uploading. Even if all the security agencies had your IP address being recorded they still do not know what you uploaded. They need to install a logger on your computer or other such surveillance device. BUT if there is only a few uploaders and the news hits then they bring in those people for questioning and examine their computers. That was always the case if the agencies had enough nodes of their own out there.
For a client (uploader) then they can use a VPN, or TOR as @TylerAbeoJordan says. Remember uploaders do not need to be running nodes as well.
And I read it as the method of incentive payments may be stopped permanently, OR emissions return in modified form, OR a new type of incentive is used to disperse the token (maybe some proof the node is storing the data then giving an incentive)
Because the more nodes you run the more chance of getting one of the payments for storing data. Because which nodes store the data is based on the hash of the data.
This is also another blow to privacy/anonymity if IP addresses are being monitored for geolocation (which will requiring blocking VPNs etc also - not easy).
I doubt they would track the IPs. There are ways you could enforce the uniqueness of an exit IP without actually tracking it somewhere.
