Technically speaking, CanonChain is a decentralized supercomputer formed by the idea of blockchains. This supercomputer runs on any network device with CanonChain protocols and other DeOS (such as Ethereum, EOS, etc.). In the supercomputer, terminal devices provide input and output, data is processed in the foggy network controlled by CanonChain. CanonChain drives current Internet of Things (IoT) to the era of Fog of Things (FoT). Although various devices in IoT are distributed in different areas, the interaction between them and all the operations are performed through the cloud server. Essentially, IoT is still a centralized architecture with many bottlenecks in large-scale expansion: soaring costs, system congestion, reduced reliability, and server vulnerability.
Hardforks occur in the Bitcoin and Ethereum community when a proposal is updated or a security problem is to be fixed. In CanonChain, the governance power belongs to block producers elected through community consensus. They must represent the common interests of all accounts in CanonChain community, and community members can elect to replace block producers if they abuse power or refuse to vote on changes in compliance with community interests.
Following the standard chain of fog networking protocols, any underlying blockchain operating system such as Bitcoin, Ethereum, EOS can run on the CanonChain. This type of platform application is called “DOS.”
Interactive applications are the ones with low latency and high reliability requirements, including vehicular network, mesh network, medical diagnosis, security alarm, etc. Services for such applications need to be acknowledged within a short time period (possibly millisecond) that is much less than the block production time of current blockchain technology.
In CanonChain, any peripheral devices can participate in the vehicular network to process information about the surrounding vehicles. For example, with CanonChain enabled, the street light can be a beacon to help vehicle positioning; all roadside restaurant cashier stations can be the computing center for road scene calculations; pedestrian handsets can send alerts to vehicle collision avoidance systems.
CanonChain provides an economically efficient solution for implementing deep learning networks. By assigning the layers in the deep learning network appropriately into different nodes, only necessary parameters between layers are exchanged among nodes to speed up the learning process and improve the learning quality. In CanonChain, the data in the deep learning network is reasonably allocated according to the computing power of the device, and their corresponding rewards are provided.
There has been little to nothing happening to CanonChain ever since ICO. They’re still working on developing their Fog Network which is set to launch at some point in 2019. The team is consisted from a Chinese-based programmers and developers but no one is particularly well known or experienced. Their vision also seems a bit too utopian.
Overall, CanonChain sounds like a legit project, however they still do not have a working product. DAG protocols are a very interesting niche, and we’ll certainly be waiting to hear more from the CanonChain project.
