Koinos and Hive: What's the difference and why?

This post is partially in response to @edicted 's recent post here: https://peakd.com/hive-167922/@edicted/koinos-bandwidth-derivative

While I agree with some of his points, I disagree with others and some are incorrect so I thought I'd take the time to write a post from my own perspective.

For those who don’t know, Koinos is a free-to-use general purpose smart contracts block chain that was developed by most members of the same team who originally built Steem/Steemit (and therefore also Hive). Koinos main net went live on November 5th 2022 (just a few months ago).

Intro

First off, I was previously a Steemit Inc employee but I am not a Koinos Group employee, however, I am currently building things on Koinos (one of which is the upcoming Kollection). I know the Koinos team and worked with them for years before they decided to build Koinos. My opinions are my own and may not be that of Koinos Group or its employees.

Second, I'd like to say that this is not a battle - there's absolutely no reason why Hive and Koinos can't co-exist. I love the Hive community, even if Hive does have some flaws (most of which have carried over from Steem) - but really it's all about the community. After all, I am here writing this blog post on Hive and I'm not currently building another blogging platform myself.

Why am I not writing this on Koinos? Because it's not a blogging platform, it's a general purpose smart contracts chain. Could a blogging platform exist implemented on Koinos? Of course, but, that doesn't exist yet and it may never (but I wouldn't be surprised if someone makes an attempt at that at some point).

DPOS vs Proof Of Burn

@edicted mentions throughout his post that Koinos is DPOS - that is not correct and is one of the big differences between the two chains. Koinos uses the Proof of Burn (PoB) consensus mechanism. It's not necessarily a new idea as it was first proposed for Bitcoin back in 2011 by Iain Stewart, but, I don't believe any chains actually implemented the idea until now. It's kind of a hybrid between PoS and PoW - it takes some of the best pieces of both of these systems. It's honestly a lot more like PoW than PoS. Think of PoW but instead of paying for mining equipment and electricity you are burning your stake in exchange for virtual hash power. There is no voting in witnesses or block producers and in fact there is no stake weighted voting at all. I'll let you read further on that elsewhere (like in the Koinos unified whitepaper that @edicted mentioned he read in his post)

The Value of Free Transactions

So next let's get this out of the way - one of the things that I do agree with him on is that eventually the free-to-use element of the chain called Mana will likely have some monetary value by way of renting it for people who don't want to permanently hold the underlying Koin token in order to get perpetually renewable free transactions (phew, that was a mouthful). Even though Mana is not a tradable token and is itself a property of Koin it absolutely does have some value in that it allows you to perform transactions. We're probably a long way off from "Mana rental" even making sense to anyone, but it could certainly happen one day. I disagree with him that this isn't well known or is something being "hidden". The Koinos community has long talked about this aspect and even the Koinos Group themselves is talking about offering a "Mana Fountain" product to go with an eventual paid API service for businesses.

One big part where these two chains differ is that Hive Power not only allows you to perform free transactions perpetually but that Hive Power also influences your ability to distribute inflation through stake weighted voting. Koinos does not have this by design as there is no stake weighted voting and no "rewards pool". It was super obvious that bid bots and Hive Power delegations would eventually become a thing - people are essentially leasing their ability to get a piece of the rewards pool. Whether this is "good or bad" is beyond the scope of this blog post but the one thing I'll say is that if code allows people to do something and it's monetarily advantageous for them to do so they will absolutely do it. No question. I will say that Steem/Hive did come up with some pretty good mitigations to bring balance to the system - it's not perfect, but it often mostly works.

Staking

So speaking of Hive Power, here's another big difference with Koinos: you do not pre-stake your Koin. There is no "Koin Power". Koin by default is always in liquid state. As you use Mana to perform transactions your Koin becomes temporarily locked, but only what you actually needed to perform that transaction. It automatically reverts back to liquid state over the course of 5 days. On Hive it's necessary to have a long power down time for a variety of reasons, but, it's mostly reasons relating to the stake weighted voting aspect. With no stake weighted voting on Koinos, there's no need for a "power down time".

Delegations

Coming from Steem/Hive, the word delegation has quite a different meaning. I now believe they should have chosen a different word for this on Koinos. I've talked with both the community and the team about this and most agree that in hindsight a different word would be more appropriate - it is still riddled throughout the original whitepapers though as @edicted mentions in his post. What Koinos has is basically payer/payee semantics, it could probably be called "mana borrowing". You can authorize other accounts to spend a certain portion of your Mana to perform transactions, but it's only spent as it's used - it never leaves your account and you can authorize or unauthorize this spending at any time. It's not a "temporary rental" of "voting power" (remember, there is not stake weighted voting on Koinos).

Why build another chain?

So this is a bit of a contentious topic but it's right there in the beginning of @edicted 's post: Why would previous Steemit Inc employees build a new chain instead of working on Hive? Well, I am not them, so I can't say their reasons but I can tell you from my own perspective why I think they would do that. @edicted mentions that it might be because devs love to "abandon spaghetti code and start fresh". As a developer, I can say that's always a tempting thing to do with any project but I firmly believe their reasons were quite different.

They set out to build a chain that differed in these ways:

1. Replace DPOS with a different consensus mechanism that is immune to the exchange attack.

2. Ability to perform chain upgrades without hard forks

3. No unnecessary politics required for the chain's ability to upgrade

4. General purpose smart contracts instead of being an application specific chain

5. Free accounts (bitcoin style) that don't have to be subsidized by anyone to be able to use.

If they proposed these changes on Hive two years ago it is likely (in my opinion) that people here would still be voting on whether they need to form a committee that can further vote to decide if they can start working on these changes. It's just political. It's sometimes difficult to get things done quickly. It's also very unlikely that everyone with large amounts of stake would agree that all five of these things need to change. AND that's ok! Hive can be Hive, and Koinos can be Koinos. It is what it is. The Koinos team instead spent the last two years developing Koinos and Koinos mainnet went live November 5th 2022.

There's also another elephant in the room here that's not often discussed. I am unsure if Hive's proposal system could reliably fund a full time dev team. Maybe it could - I could be wrong. It’s still a very cool DAO-like system that can fund stuff. Regardless, it's a bit iffy trying to run a professional team that way because of the nature of stake weighted voting and politics. Funding that you have today may be gone tomorrow. Regardless, they built Koinos without any funding other than their own (that I know of) and certainly not any initial VC funding. There was no ICO, no dev team allocation, no ninja mine, just everyone for themselves. A true decentralized fair launch. Their business model is not selling tokens - it's offering services to other businesses who wish to build on the chain. Imagine that - a cryptocurrency company who isn't trying to make it on "number goes up". What a concept.

So let's talk about the five points I mentioned above. I don't think I need to discuss the "exchange attack" and moving away from DPOS because most Hive user's here already know the backstory behind that. I'm also well aware that Hive came up with their own solution (which works) but it would require everyone paying attention to power ups before witness voting could happen and then subsequently performing a hard fork to fix the problem. That brings us to point number two: hard forks.

Koinos's governance, consensus mechanism, block production, and pretty much everything else are all implemented as smart contracts themselves. That means all of these pieces can be upgraded without even updating your nodes. It's full on in-band upgradability without the political aspect of trying to decide what changes can go into a hard fork. There's zero downtime for exchanges. It's just a downright pleasant experience for everyone involved in comparison to what we've all already been through in the past. The only type of things that you'd need to upgrade your node software for are performance upgrades or potentially security related fixes. As far as those performance fixes, Koinos is built as a series of microservices so it's much easier to replace certain pieces than with a monolithic codebase. In fact, Koinos is written using multiple languages already for different microservices (C++, Go, TypeScript). Sometimes it's just best to choose the proper tool for the job.

So, what about general purpose smart contracts? Are those really needed or can we do it all with layer-2 solutions on Hive? I'm not gonna lie. There are some amazing businesses built on Hive using layer-2 solutions, it does work and is often a great solution. However, there is one really important difference in having actual on-chain smart contracts: all dApps built are all on the same base layer and are exposed to each other in an entirely decentralized way. With layer-2 solutions you often need to have an entirely separate set of validators to maintain some level of decentralization and you are basically "off on your own" instead of being able to have your smart contracts directly interact with each other. It's just a really nice thing to have that I think will make more sense as to the "why" in the future once people see dApps interacting with each other and people building businesses on top of other people's businesses. Maybe think of it like stacking legos. You can plug your software into their software on-chain. Very cool stuff. Aside from these reasons, dApp developers can easily upgrade their smart contracts without waiting to see if the features they need can be included in the next big hard fork.

Free accounts. This one I think is extra important and something I think Hive should consider tackling at some point too. As Steemit was trying to scale up due to periods of heavy interest it became increasingly difficult to subsidize account creation. It's a pain point that still exists on Hive. For Steemit it required rate limiting in the form of collecting some pieces of user data (phone, email, etc). At times there was even an approval queue. This is what the "free account credits" band-aid was born out of. I think Hive is probably doing ok with this right now but I also think it would be pretty difficult to scale up quickly if tons of people tried to sign up. Beyond that, the overall landscape of cryptocurrency outside of Steem / Hive has changed quite a lot since Steemit first launched. People expect a different experience today - they are used to just creating an address in Metamask and immediately using it. No KYC, no asking a friend to create an account for you, just create an account and start using it. It does make a difference in user experience. Human readable account names are very cool, but those can still be implemented as smart contracts (which someone is already tackling on Koinos as well). To be fair, maybe Hive already has plans for this and I'm not aware of it yet.

Conclusion

The differences between these two chains are not limited to the things I talked about in this post and are actually quite vast. They are not in competition with each other. It's two things that happened to be originally developed by many of the same people and teams. I fully believe that Hive and Koinos can all be "on the same team" and co-exist.

Anyone have any cool ideas for Hive / Koinos collaboration projects? What can we build together?