Terms specific to the Dat software.
A dat, or Dat archive, is a set of files and dat metadata (see SLEEP). A dat folder can contain files of any type, which can be synced to other users. A dat has a Dat link used to share with other people.
When you create a dat, you're creating a
.dat folder to hold the metadata and the dat keys (a public and secret key).
Dat links are the public part of a key pair. Users that have the secret key are able to write updates to a dat.
With the Dat CLI and Desktop application, secret keys are stored in a dat folder in your home directory,
~/.dat/secret_keys. It is important to back these up if you get a new computer.
User who can write to a Dat archive. This user has the secret key, allowing them to write data. Currently, dats are single-writer.
User who are granted read access to a Dat archive by the owner. A collaborator can access a Dat archive if the owner or another collaborator sends the the Dat link.
In the future, users will be able to grant collaborators write access to the Dat archive, allowing them to modify and create files in the archive.
A group of peers that want or have downloaded data for a Dat archive and are connected to each other over the Distributed Web.
In a Distributed Web (P2P) model, those who are downloading the data also provide bandwidth and storage to share it. Instead of one server, we have many. The more people or organizations that are involved in the Distributed Web, the more redundant, safe, and fast it will become.
Currently, the Web is centralized: if someone controls the hardware or the communication line, then they control all the uses of that website. Read more here.
A P2P software program searches for other connected computers on a P2P network to locate the desired content. The peers of such networks are end-user computer systems that are interconnected via the Internet.
In Dat, peers only connect if they both have the same Dat link.
The Beaker Browser is an experimental p2p browser that can view and publish dats alongside traditional web browsing.
SLEEP is the the on-disk format that Dat produces and uses. It is a set of 9 files that hold all of the metadata needed to list the contents of a Dat repository and verify the integrity of the data you receive.
The acronym SLEEP is a slumber related pun on REST and stands for Syncable Lightweight Event Emitting Persistence. The Event Emitting part refers to how SLEEP files are append-only in nature, meaning they grow over time and new updates can be subscribed to as a realtime feed of events through the Dat protocol.
Read the full SLEEP specification in the dat whitepaper.
A 32-bit hash that uniquely represents a feed. All feeds have keys. The metadata key is used as the public key for an archive.
The public key is the key that is shared in the Dat Link. Messages are signed using the secret key, allowing write access to feeds. The owner is the only user with the secret key.
The discovery key is a hashed public key. The discovery key is used to find peers on the public key without exposing the original public key to network.
A feed is a term we use interchangeably with the term "append-only log". It’s the lowest level component of Dat. For each Dat, there are two feeds - the metadata and the content.
Feeds are created with hypercore.
Like an HTTP header, the metadata contains a pointer to the contents of Dat and the file list.
The metadata is a hypercore feed. The first entry in the metadata feed is the key for the content feed.
The content feed is a hypercore feed containing the file contents for a Dat archive. The content feed together with a metadata feed make a Dat archive.
Hyperdrive is peer to peer directories. We built hyperdrive to efficiently share scientific data in real time between research institutions. Hyperdrive handles the distribution of files while encrypting data transfer and ensuring content integrity. Hyperdrive creates append-only logs for file changes allow users to download partial datasets and to create versioned data. Hyperdrive is built on hypercore.
Archives created with hyperdrive are made with two feeds, one for the metadata and one for the content.
Hypercore is a protocol and network for distributing and replicating feeds of binary data. This creates an efficient gossip network where latency is reduced to a minimum. Hypercore is an eventually consistent, highly available, partition tolerant system.