The software from the affiliated laboratories of C4DT are in three categories: showcase, fresh from the lab – incubator, being worked on by C4DT – market, stable and supported software.
The showcase offers an unfiltered insight into the totality of the laboratories software. Most of the projects are in active development by the laboratories, but by no means ready to be used by external users. https://showcase.c4dt.org gives an overview of the projects that the affiliated labs are working on.
1-2 times a year we choose a new selection of projects from the showcase that we actively work on. The choice is based on the maturity of the projects in the labs, the interest from our partners, and our estimation of how much this project is interesting to the general public.
Once we chose a project, the Factory team works on it for 6-9 months. During this time we refine the code base and create a proof-of-concept together with one of our partners. Depending on the interest by the partners and the lab, the project will continue as open source software, as project in the lab, or as a start-up.
We’re currently working on the following projects in the incubator:
- D-voting – a decentralized voting application using the blockchain from Bryan Ford’s lab
- Federated Predictive Maintenance – together with Martin Jaggi‘s lab and one of our partners we’re working on applying federated machine learning to the problem of predictive maintenance
Participation in Open Source libraries
As part of our work we like to participate in Open Source libraries that we use. This is to give back some of our time to the libraries that enable our work:
- Angr, a symbolic execution library used in formal verification apps
- Arti, the new library and app for the tor network
- MicroG, allowing to run Android without link to the Google empire
- Strictyaml, to make sure your yaml files are correct
- Tokio / TLS, a rust library for web services
These were the projects we worked on in the past:
- 2022/12 – Disco – a web-based frontend to implement distributed machine learning
- 2022/10 – Klint – Formal verification of network function binaries
- 2022/08 – Tandem / Monero – together with Kudelski security we evaluated the usefulness of applying a privacy-preserving private key protection to the Monero blockchain
- 2021/06 – AT2 blockchain – a very fast transaction-only blockchain
- 2021/04 – Garfield – Garfield allows one to write distributed Byzantine fault tolerant ML applications
- 2021/04 – LibreChovid – run the SwissCovid app on phones without google app
- 2021/04 – Stainless – formally verifying smart contracts
- 2021/01 – CrowdNotifier – privacy-preserving notifying of visitors of venues and events
- 2020/10 – SwissCovid – A privacy preserving exposure notification protocol to inform people about possible exposure to the COVID-19 virus
- 2020/10 – Spindle – Scalable, privacy-preserving, distributed learning on distributed databases
- 2020/06 – Drynx / MedCo – privacy-preserving medical-data sharing
For some project which are interesting to several of our partners we offer it in the Factory Market. This is a place for Open Source projects that we support for 2 years.
- LightArti-rest is a library for mobile apps that allows to use the tor network for anonymous statistics without using too much bandwidth
- DRand was developed in Bryan Fords lab and is now continued as part of Protocol Labs. We’re running a node of the network and participate in the randomness
These projects were in the Market in the past but are not currently supported by the C4DT Factory anymore.
- Omniledger / Byzcoin is a permissioned high-performance blockchain developed at the laboratory of professor Bryan Ford. Here is some more information:
- Factory Article – general information on the project
- Omniledger – the main project
- Calypso – enables a writer to encrypt a document and to store the list of certified readers on a blockchain
- Odyssey – it enables a data owner to enter their dataset into the system, together with the access rules that need to apply