Exploring Kaspa & a robust DAG-based approach to PoW (GhostDAG)

I really don’t want to be labelled a shiller here so I want to preface this a little bit and make this post purely informative rather than sensational. I have no connection with the development team or the community besides an interest. I only found out about this a couple of weeks ago but have been researching it quite a bit and want to share what I found for anyone who is interested.

I know every single day there’s a new coin that comes out that everyone is trying to hype and get investors. I’m not encouraging anyone here to buy Kaspa or anything like this – you can’t invest in it anyway it’s already been launched it was a fair launch either way. I’m only posting because I think what the team at Kaspa has come up with is genuinely super cool and for anyone interested, and that it’s worth exploring, doing your own research, getting to know the idea, the implementation, etc and see if it’s something you think it’s cool as well. For all I know it can tank tomorrow and I’m in no way claiming it’s going to make you a millionaire, so do not invest in it.

This is strictly informative.

I also want to say that although I’ve done a fair amount of research on this coin, it’s possible I might be getting some things wrong so please correct me so take everything I say with a grain of salt and confirm for yourself.

So Kaspa is essentially a community project (if you go on the discord you can cast your vote for the direction its going in), and has no business model. It was researched and primarily by Yonatan Sompolinsky (Postdoc CS at Harvard University), Shai Wyborski (Ph.D candidate at HUJI/BGU, quantum cryptography), and a few others.

It basically uses a DAG protocol, which stands for directed acyclic graph. You can think of a DAG as a block-web rather than a block-chain. The DAG has a central backbone that is chain-like, but blocks can branch out from the backbone, and can “depend” on multiple parent blocks – rather than just one. In a blockchain, each block has one parent. In a BlockDAG, each block can have multiple parents. As I understand, DAGs were generally considered a problem as they are almost a pictorial representation of what you don’t want. A web of blocks that may not be in agreement with each other. In fact with BTC and other Blockchains, the whole point is to remove any “web-like” structure (orphan blocks) to have just one solid backbone with nothing sticking out of it, making it a chain.

However, at the same time, BlockDAGs solve a lot of issues. They allow for super fast and cheap transactions, and scale with ease. For example, Kaspa for example is capable of securely doing one block per second right now, and can theoretically go higher. I’m not sure how this compares to other DAG coins, but this basically means instant (and cheap) transfers. Which is pretty neat.

There have been attempts to use BlockDAGs in the past, and these BlockDAGs have been vulnerable to attacks. Not all BlockDAGs are the same. The team at Kaspa have been working on a specific protocol they call GhostDAG for what I believe to be nearly a decade. In fact, I think one of Yonatan Sompolinsky’s related papers was cited in the original Ethereum whitepaper. Anyway, GhostDAG is a way of creating a secure way to order the blocks in such away that it protects against potential attacks and double spending. It uses a clever way of ranking parent blocks in the web in terms of their trustworthiness.

What I personally find intriguing is that the security of the protocol was actually mathematically proven by them, and you can read about this proof in the whitepaper. It boils down to a “k” value, where basically you choose how secure you want it at the expense of performance. Right now “k” is set to 18, which basically means 18 “parallel” blocks are allowed at any one time. I believe this equates to protection against a 47.5% attack, but don’t quote me on that. It gives ample room for fast transactions. In the future, they are working on something DAGKnight protocol, that would theoretically allow for “k” to scale dynamically with change in network latency times, basically auto-optimising what would otherwise be a constant parameter.

There are some other issues that they have ways of solving. For example DAGs can be quite heavy storage-wise, as they branch out, taking up lots of room on a hard drive. But the team have come up with a way of pruning old blocks to make it lighter. I’m not totally sure how that works as I haven’t looked into it too much, but this whole thing is explained in detail in the second half of this GhostDAG 101 video. This is a great video but it’s 2 hours and quite technical. If you like that stuff, please watch it.

What does this mean practically?

From what I understand, BTC offers a very high amount of security, but it’s expensive and takes ages to send money. This makes it impractical for quick transactions, which you might say is why pizza shops don’t accept BTC generally. But it’s trustworthy due to being proof of work. Nobody wants to spend an extra few dollars for a pizza and also wait 10 minutes – 3 hours for it to go through. So people generally use BTC to “hold value” and move it in large chunks when needed. When we are looking for quick transactions, we generally resort to level 2’s or a proof of stake, which is apparently less secure than proof of work.

The ideal is to have a fast, secure, a scalable level 1, and it seems Kaspa has come up with a contender for such a thing. Something that it is the same echelon in terms of security of BTC, but also something you can use easily in daily transactions. If you think of the “ideal” currency, it’s decentralised, scalable, cheap to use, fast, and secure. It would seem that before Kaspa, all the coins had some variation of all of these aspects but not all of them.

On that note, if anyone knows another coin that meets all these criteria, please post it below.

Again, don’t take my word for this. I’m not asking you to buy Kaspa or anything. If you find this interesting, just do a little research. I thought it was super cool to learn about this technology, and I hope it picks up in the future. If you look up Kaspa, you might find a lot of people saying “LFG!!!” or “to the moon!!!” because they are only looking to get rich off this. That’s not the point of this post, I’m looking to grow the awareness of a coin I think has real life utility in the future, and IMO has a solid team behind it with good intentions and good values.

Thanks for reading!