Inter-Planetary File System
= "an open source project that aims to enable peer-to-peer methods for storing and disseminating information on the web".
URL =
Description
By Chelsea Barabas, Neha Narula and Ethan Zuckerman:
"The Inter-Planetary File System (IPFS) is an open source project that aims to enable peer-to-peer methods for storing and disseminating information on the web. The goal of this project is to give users the ability to publish online without having to trust a single third party server to host their content. Instead, IPFS provides a verifiable means of retrieving content from a distributed network of storage providers. IPFS’s main insight is to use hashing functions to point to content, instead of using the IP address of a server where the content is housed. A hashing function is a function that can be used to map a data file of any size to an output of a fixed size, usually in the form of a series of random letters and numbers. The hash produced by a file can serve as a unique fingerprint of that information–for any given file, the hash generated will always produce the same unique value. Conversely, if any aspect of the file is modified, the hash value will change.
The IPFS system uses the hash of a file as its pointer, effectively decoupling the physical server that hosts the content from the address that points to where that content can be found. It does this by storing files in a distributed hash table (DHT). A distributed hash table is a distributed system in which the responsibility of maintaining the mapping from key to value is distributed over all nodes in the system. A DHT replicates data, and can tolerate nodes entering and leaving the system. Clients are assured of the integrity of the data they are receiving by checking the hash of the file they are looking for.
Therefore anyone can easily copy and serve content, making it harder to take that content down, and potentially improving latency by making files accessible in multiple places. IPFS stands in contrast to the way content is currently discovered online today, using URLs and HTTP links to identify a specific server host, where that content lives. If IPFS were to gain mainstream adoption, it would make content more resilient in contexts where Internet connectivity is weak, or censorship threats are high. IPFS is essentially a distributed file system with a simple protocol that enables easy finding, caching, and serving of files. If those files are appropriately replicated across the network, they could tolerate outages more easily than today’s web. The developers on the IPFS project imagine their system could serve as the backbone for a peer-to-peer file sharing network, whereby information is exchanged locally via a mesh-like network. In this way, important digital content could be disseminated, even if access to the Internet is cut off or platforms are pressured to take down specific content.
Not only might IPFS make content more resilient, but it could also enable a more competitive landscape for publishing platforms in the future. Agreeing upon a peer-to-peer protocol and a way for storing and retrieving content is one way we could enable more sharing between applications, as we saw with the web, and thus lower the barrier for a more diverse social media landscape. If combined with a shared common data format, IPFS might reduce switching costs between applications. Many different services could use the same data, thus eliminating the need for the user to replicate the same information, such as their social graph, photos, prior posts, and interaction history each time they sign up for a new service. If it’s easier for users to switch between related platforms, then it makes it easier for new services to bootstrap a network, ultimately providing more choices in the market." (http://dci.mit.edu/assets/papers/decentralized_web.pdf)
Status
- As of now, the project has two implementations, in Go and Javascript