Sort:  

https://medium.com/openorchard/announcing-koinos-whitepaper-koin-mining-e2755f5be33f

It's the Steemit Inc devs trying to fix all the problems they ran into with Graphene basically.

Modular Upgradeability
Modular upgradeability enables us to rapidly add different features to the framework without impacting any other part of the blockchain. This is how we will be able to offer a testnet in just 3 months that delivers Ethereum feature parity, but with WASM smart contracts, fast block times, and fee-less transactions. While the testnet will feature proof of work, the Koinos blockchain framework makes it easy for us to swap that consensus algorithm out for the totally new and different algorithm that we’re developing for mainnet.

Thanks to modular upgradeability, individual upgrades can be pushed to the network much like an operating system patch but with an on-chain record of the entire path that the upgrade took. Instead of having to package critical upgrades into time consuming, political, and risky hardforks, individual fixes can be pushed to the network the moment they are ready enabling Koinos to rapidly improve itself. It is that capability that we believe makes Koinos the first blockchain in history capable of evolving.


Scalability
After upgradeability, the next major challenge facing existing blockchains is scalability. Blockchains hold on to every transaction and smart contract that has ever been added to them, regardless of whether they get used or not. That means that the cost of lugging around all of this expensive data gets baked into the cost of every new transaction which itself adds even more information to the chain. All of this old and unused data is responsible for the vast majority of the cost of running a blockchain. But what if we could remove all of that data, while retaining the ability to bring it back to life as if it had never left like MAGIC?
State-Paging

Well we have figured out how to do exactly that. We call it state-paging and Koinos will be the first blockchain ever to have it. Not only is this a genuine innovation that could be THE key to scaling blockchain to the masses, but existing blockchains like EOS and Ethereum cannot take advantage of this capability without totally rearchitecting their entire blockchain.

Damn that sounds pretty cool in some ways and really sketchy in others!

Doing things without needing a hard fork is cool, unless there’s Justin Sun types running around calling the shots.

Being able to off load things that take up memory, like someone’s stake and posts, seems a little iffy in the same scenario.

Other than that if it’s run in a good manner I think they will have some pretty great stuff there. I’ll have to keep an eye on it and try to research how to get involved.

Doing things without needing a hard fork is cool, unless there’s Justin Sun types running around calling the shots.

It's all about network consensus.
I imagine there is a very interesting model here behind the scenes.

Yeah true. I’ll definitely give this one a check and try to write a review of it. That type of stuff is how I am able to best learn and understand it, when I can try and explain it to others.

Thanks for the info man!