The only reason why I'm not fully sold on #nostr is that I don't see incentives for relays to store huge amounts of data.
If I were to build a YT-like application, how many of the relays will store the videos uploaded through my app? If I set up my own relay and mine is the only one that does the saving then it kinda defeats the purpose. If other relays actually save those videos, the costs would be enormous for them.
What am I missing?
#asknostr
I think you can reboadcast your posts to other relays from time to time. Biggest use case is close family and friends can use the relay and have your full history. Does everyone else really need to see all of your old posts? Anyone that cares enough can add the relay.
Data harvesting.
Analytics.
Sell data to corps and governments.
Use data for making decisions.
Offer related services based on data.
Good point
nostr:npub1xtscya34g58tk0z605fvr788k263gsu6cy9x0mhnm87echrgufzsevkk5s
Relays don’t store media, they only store small text blobs
Okay so the size of storage is smaller than I originally mentioned, but the core issue still stands: why would the protocol work if relays aren't incentivized to store data?
there are so many applications of relays I'm not sure it makes sense to say there are no incentives to run them.
In my mind this is the same question as “why will people run email or web servers if there is no incentive to do so”. They are pretty useful for communicating and coordinating things on the internet in a store-and-forward way with a data format that is simple and extensible. I would say thats more than enough reason to run them.
So what I understand is that, given enough user traction, eventually relays will have to transition to a paid version or some sort hybrid (offering tiers for example) in order to sustain themselves, thus making this the incentive to run them.
Relays can be sustained by the business model of the application that is using it as a backend.
Supply and demand. If anyone wants it to be stored or wants to access it in the future, they will pay.
The question should be - why is YouTube storing all the crap? Who pays it for them?
There's market for everything, and there are (will be) markets for data as well.
Forget about YT, it was just an example to point the size (which apparently can't be done because relays don't store media).
My difficulty comes from the fact that is stored off-chain, so what do you do with publishing rights for example? This is something we would would want to keep those and access them in the long run.
My thinking is in comparison to the immutable, secure, censorship resistance that on-chain data is offering.
You can run your own relay to keep a copy of your own notes available. I have one set up on an umbrel server.
I can imagine there will be relays where you would PAY in order to store your data. You want the data stored, you will pay (if it is useful, you could get it back on zaps). And because redundancy matters, you could pay several of them.
Just like in traditional world; you can get free webhosting, or your own server, or pay someone who stores your data for you.
I (still) think something like torrent could help with it.
Yes it is not 100% reliable nor fast every time. But you could have both big seeding silos that take almost anything and are run by some rich people that just want to scrape everything or some donation based org. Or if someone really want their content to be available, they will seed it themselves.
Buuuuut apple doesn't allow torrent apps in the store so an universal tight integration of torrent in nostr clients is not on the table rn.
Webtorrent?
