Data Availability ≠ Data Storage

What's the difference?

🥫 Data storage is like canned food

🍱 Data availability is an all-you-can-eat buffet

Data storage is a solution to the problem of needing to save data so that you can retrieve it later.

This is like canning food. You jar it up and store it away so that it is preserved to be eaten later.

Data availability is a solution to the problem of needing to make data available for anyone on the internet to download.

This is like an all-you-can-eat buffet. You want to put all the food out for anyone to come and eat as much as they want.

Data availability protocols like

@CelestiaOrg

make it efficient to verify that data is available to download with data availability sampling.

This is like checking that all the dishes in the buffet are available to eat by sampling a bite from each dish

A data storage solution isn't suitable for this purpose because you'd have to eat the entirety of each dish in order to verify that it's available.

Most people can't each that much... just like most nodes couldn't handle that much data.

But food at a buffet spoils quickly because it's left out, so the food might not be edible later.

Likewise, data availability layers don't guarantee the data will always be there to download, just for a limited time.

(Note: one full storage node fixes this.)

Meanwhile canning food does give you a guarantee that the food will be edible later.

And likewise storage solutions guarantee that data will be there to download at some point in the future.

To summarize:

Canning food is not the same as serving it at a buffet. Storing data is not the same as making it available.

@CelestiaOrg is a data availability buffet.

Protocols like @ArweaveEco and @Filecoin on the other hand are data canning factories.

Our sovereignty over our data may be the same as becoming a chef.

We can choose what we serve from Canned food.