There is a new project that has some pretty big hitters on its team and they are building a new protocol to replace the current standard "Gossip Protocol"

marlin.pro

The system will be using a plethora of new networking solutions to create on-demand custom overlay networks. These include packets being forwarded even before they are received in entirety (cut-through switching), routing over low-latency multicast trees (to replace the epidemic flooding/gossip), randomized path selection before the diffusion process (a modified stemming+diffusion approach common in systems like Dandelion - for lightweight anonymity), ttransport layer improvements to improve over TCP head-of-line blocking and congestion control, and many more

This will be the project we hear a lot about in 2019 and few non-engineering people will really understand what it does.

A core dev from Zilliqa is heading it up. The project won a competition at Stanford earlier this year. With its mixed team from MIT, Stanford, IIT Mumbai with support and advise from City of Zion and other world class Research labs, the project is built for success.

Edit: There is now an official ANN discussion here

https://bitcointalk.org/index.php?topic=5074571.msg48342672#msg48342672

The concept of the project looks good. But how MARLIN PROTOCOL  works ? What is our assurance if we will adopt this kind of platform.?

The methodologies listed are all available today, but no one has added them into a universal protocol set. By having them as part of a decentralized relay framework, you can make them all available as modules with unlimited scalability and fault-tolerant uptime

The most important thing is to understand what a big affect this has on other blockchains. The network layer will always be a bottleneck for scalability and through-put if there aren't solutions like the marlin protocol available.

Few people actually understand that finality is more than just a consensus algorithm....

This looks interesting. I hope this project works out for the good. Thanks for the heads up. Will be keeping an eye on it. Hope to hear more about Marlin Protocol.

Very informative. What will be the approach of this start-up about blockchain? Will you create new blockchain or adopt any existing blockchain?

We are not fixated on building our own chain (def some advantages and have a design for it), but are also open to using chains like eth, neo, nem. Our testnet is being developed over Ethereum to allow modular testing.

good information for a new project, let alone the project won the competition, I think it's a good breakthrough to attract someone to join this project. that achievement will be an attraction

From what I had picture out in mind of this so called new project this will highly rely on network maker or masternode.
Overlay network layer is interesting when embodied with privacy. Is this some kind of disruption of some big content provider like netflix or youtube?

The disruption is at the network layer -- P2P communications have not been designed with the decentralized systems of today in mind. Netflix is a good example of one of the areas this disruption will be felt. Today, Netflix has a huge AWS bill. With Marlin, the same infrastructure can be replicated at a fraction of the cost

This project is quite interesting, but now is not a good time for the development of new blockchain platforms due to the deterioration of the total capitalization of cryptocurrencies and an increase in distrust of many projects due to scammers. It would be better to take (buy or forge cooperation) an already existing and practically operating blockchain project and make corresponding changes in it that significantly improve its attractiveness.