Condensation is a open-source general-purpose distributed data system. Its implementation provides conflict-free synchronization, data integrity, and end-to-end security by design. It is a lightweight solution representing around 15000 lines of code, and composed of three modules:
- 📝 A client module : the core module allowing to structure the data from a local mutable database into an immutable blockchain. It can be installed in web, mobile, or IoT client application.
- 💾 A storage module : this module can be installed on a cloud, on a private server, or on personal device in order to store the data.
- 🔐 An encryption module : the data is end-to-end encrypted with industry-proven algorithms from existing libraries.
Condensation remains simple to learn as it offers an API in the form of a document database. No specific background in blockchain or related technologies is required to be able to build applications with Condensation.
1. For the user
- Offline-first
- Versioning
- Digital certification
- Collaborative editing
- Reduced loading time
2. For the infrastructure
- End-to-end encryption
- Live back-up
- Parallelized computation
- Unconstrained storage location
- Peer-to-peer synchronisation
While both database and file systems have greatly evolved over time, their main structure has hardly changed. Database systems are based on tables with mutable records (rows), while file systems use a hierarchy of folders with mutable files inside. In both systems, data can be modified with little effort, and this at any time. Data synchronization, however, is notoriously difficult and error-prone. To perform such operations data has always been collected, processed and merged in servers, today organized in clouds, in a centralized manner. With Condensation we believe that we can change this paradigm by providing a solution to share and synchronize data in a distributed manner.
Distributing our processes is advantageous for several reasons:
- Ownership : Delegating the storage and the processing of our data to a third-party brings some contraints (we can do only what the third party allows us to do), and security risks (for example our data can be read by the third party).
- Speed : Transporting data is always slower than processing it where it is generated. Moreover, our solution has the clear advantage of circumventing intermediate parties.
- Costs : Storage, transportation and data processing have a non negligeable costs. If it can be done by the device itself it is always beneficial.
Today, we lack tools to develop fully distributed applications
Most applications work with data systems that are centralized in clouds, which means that all the data involved flows through "someone else's computers". It is very difficult today for many companies to keep the ownership of their data. These companies need to trust a third party, like cloud companies, and adapt to their constraints and prices. With this project we aim to provide a tool allowing the development of modern and highly scalable applications without needing to trust a third party.
Condensation allows to use clouds as just data storage. In this set up, data remains fully encrypted when in the cloud company's hands. All the processing and more importantly data synchronization in between the different users can be done outside from the cloud, on the users's devices, or on the data owner's servers.
On one side, even if they are great for source code management, current revision control systems are not suited as general purpose data systems. In order to benefit from such tools, the user needs to have a minimal understanding of branches, merging and conflict resolution, which is far beyond the knowledge of an average computer user.
With Condensation users do not need to share entire documents to synchronize, but only the last changes. They can easily work offline because synchronization can be done in peer-to-peer, there is no need of a central ground-truth that is used to synchronize everybody. The merge algorithm is Condensation's secret sauce. It is highly efficient and allows to avoid latency and conflict issues.
Condensation Java implementation is stable but we are still packaging it, preparing instructions for the installation, and building the first tutorials. The most advanced developers can already get into the code with the following link. Source code
Do not hesitate to contact us for a POC at [email protected]. We would be very happy to collaborate with you.
We need your help for :
-
Raising awareness and educating about Condensation. Firstly, we would like to raise awareness on the innovative contributions that Condensation brings to the development world. Because this new kind of database bridges the gap between mutable and immutable structures, producing educational material is critical to put Condensation's project forward. We are open to many type of contributions, like articles, use cases, diagrams, tutorials, advices and testimonials from your experience. Today, we are building this community in order to support your content, we will always do our best to foster your positive impact on the project.
-
Developing the core of Condensation. Another priority is now to port the code to other languages (e.g. Javascript, Swift) since Condensation is mostly developed on Java as of today. For the immutable part of Condensation, specifications are now exhaustively described and we are available to guide you into each step. If you are interested in contributing, we will give you more details on the next priorities and connect you with our pioneering community.
We are actively developing Condensation, if you want to invest, to contribute or to develop a pilot please reach out at [email protected], and join our discord CDS - Condensation Data System.
Condensation author :
Thomas Lochmatter, PhD, Teacher at EPFL meet him
Early contributors :
- Alex Mouradian : Project coordination
- Thomas Souksouvanh : Responsible of application developement
- Lou Locci : Education and content management
- Zafar Ansari : Infrastructure and development lead