What is Kaspa?

Going back to Kaspa, Kaspa’s protocol (Ghostdag) is a generalization of Nakamoto Consensus (NC). Generalization means identifying a situation (longest chain rule) as a particular case of another situation (Ghostdag).

To understand why generalisation of NC is a major breakthrough, let’s entertain a real life scenario where generalisation would matter:

Say we run an ice cream stand but have to avoid a bad neighbourhood – despite having many potential customers – because it’s not safe.

If we have a friend who lives in that bad neighborhood, we may find a solution to our safety problem by following this simple rule:

“Always call my friend to accompany me when I go to work in that bad neighborhood.” 

Let’s call this the call-my-friend rule (CMF).

CMF ensures we’re safe when we go to work in this bad neighborhood. However, if we rely on CMF, we can only go there when our friend is available to accompany us. If our friend is available only 3 days per month, we can also only go there 3 days per month.

This limits our ability to scale our business. A generalisation of the CMF rule would allow us to solve the problem by making our friend fungible for this particular purpose. In this case a generalisation would be the following rule:

“Never go alone (NGA) in a bad neighborhood”

Thanks to NGA now our business can scale without having to sacrifice security. In fact, once we realise it’s not our friend that makes us safer, but the fact that we are not alone that discourages potential attackers. So our friend becomes fungible aka replaceable by someone else.

CMF is like using NC in Bitcoin. Kaspa is like discovering the NGA rule (Ghostdag protocol). The CMA rule, i.e. longest chain, becomes a k-cluster rule in Kaspa.

We no longer look for the “longest chain”, but for a k-cluster, which becomes the longest chain rule when throughput is low (>>> 1 block per second). K-clusters are a bit too much to get into here, if you’d like to learn more about it I’d suggest reading the whitepaper on linked earlier int he article.

This has some market implications explained by what I call the trilemma curve paradigm: 

In every point in time so far in the history of crypto market expectations can be represented by a trilemma curve where smart money always follows coins that break the curve.

Let’s start by drawing a TPS-Security chart and plotting some coins there. In the TPS-Security chart below, TPS increases as we move upwards, and security increases as we move to the right.

If we plot cryptos in a TPS-security chart we can see that 2017 marks the first time the curve (line at the time) was broken as result of the big blocker forks BCH/BSV (circled in green) that combined many fold higher TPS with a security model that was almost identical to BTC’s.

This event underpinned the big blocker euphoria of 2017-2019, whose boiling point was market by the infamous hash war that led to BTC’s bottom.

Then, as crypto adoption advanced and demand for TPS soared, non-pow chains such as Solana came under spotlight because the only ones capable of offering the required tx throughput for an onchain economy. Their emergence stretched the boundaries of the trilemma curve further.

The dashed dark grey straight line can be seen as the trilemma curve for DLT performance market expectations BEFORE Solana. The market was already projecting TPS increases as we moved to the left, but not increases in the range of Solana’s TPS.

The continuous dark grey line that bends upwards is a visual representation of the adjustment in market expectations that took place AFTER the advent of Solana and Avalanche.

With Solana everyone realized that centralized DLT could handle exponentially more transactions than what was conceivable up to that point. This led to more euphoria as the Trilemma again seemed invalidated, although that was not the case.

The Trilemma remained valid, just the relationship between TPS and Security wasn’t as it was being priced in up to that moment. So the area marked in blue brush strokes ended up from being above/outside the curve (not priced in) to going under it (priced in).

The rally in Sol/Ava can be seen as a knee jerk reaction to this adjustment and ultimately as a noise move to be fully retraced, since the trilemma was not actually invalidated.

The most interesting part in the chart is the area in yellow. That’s the territory conquered by Kadena. The fact that $KDA emerged there explains the remarkable $KDA euphoria of Q3 2021 & its 50x price explosion.

In KDA the trilemma invalidation narrative was even stronger for this reason, because its advertised throughput was even higher than that of the best non pow coin, Solana.

However, again, the Trilemma law hadn’t been invalidated, the trilemma curve simply had to be re-adjsuted. With Kadena the area under the curve had increased even further and as the market priced this in, KDA’s market cap benefited.

The light grey curve marks the new curve as readjusted after Kadena. Today it is more likely for the Trilemma curve to be broken again & every time a new coin emerges there we should expect an euphoric “trilemma is invalidated” phase, followed by a correction.

The reason why these coins lead to euphoric price rallies that almost completely retrace (turning out to be noise) is that none of them invalidate(d) the trilemma law, which states that we cannot have higher throughput than BTC without sacrificing BTC’s security standard.

The grey lines marked with the green arrows below show potential future changes in the trilemma curve if we continue down this path. As long as the Trilemma law remains valid, even though the Trilemma curve may be displaced upwards, as its convexity remains.

Now you can start to see what makes Kaspa different: because it is a generalization of Nakamoto Consensus, Kaspa breaks with the current trajectory of crypto innovation where security is traded off for throughput.

To visualize this here I have added Kaspa to the TPS-security chart. As you can see, Kaspa plots right above Bitcoin. Not only does it break above the curve, but it completely invalidates it since it’s not displaced to the left.

In its current Golang iteration Kaspa supports 1 bps or 200 tps. That may not sound so fast, many chains out there promise much higher TPS. None of those chains however is a continuation of Bitcoin’s security model.

None of them inherits Bitcoin’s battle tested Nakamoto Consensus. Because it is a generalization of NC, Kaspa does. The fact that Kaspa is a generalization of NC means that there are no new attack vectors, as TPS increases the attack surface remains the same.

Currently a rust iteration is being worked on, which is expected to arrive in around 5 months time. It is expected to 10x the throughput. You can follow its development by following @MichaelSuttonIL on Twitter.