Understanding Tempo – Methods to Protect and Attack Decentralized Networks

Introduction

In short, tempo is a set of methods for defending and attacking next-generation decentralized and/or distributed networks (hereafter DAGs) with consensus, like blockchain solutions or DAGs based on the use of temporal anomalies.

Maybe not everything is obvious even after the definition: and that’s ok, because the pace is only three or four years old, and the methods and techniques that underlie it are not far behind: maybe we should start counting from the mid-1970s;

Examples of tempos.

Imagine the following situations to begin with:

• Zero time, that is, time that has stopped;
• Inverted time, that is, time that has gone backwards;
• parallel time, that is, one that somehow appeared outside of the main time;

Does it all seem like a fantasy? A fantasy game? No, it’s the reality of DRS;

Zero Time

Solana is famous for it.. This network, among other things, has a part of the consensus algorithm codenamed PoH (read more in the White paper).

It is important to us that Solana works with both standard, global, time and local. And the local, in turn, has more than once deviated from the global.

Example: “the problem (was) caused by an increase in the period it takes the validator to send the block to the network (slot time);

Solana has become notorious for taking more and more lunch time since 2021, a problem I addressed in a previous article.

But what happens in the meantime with time? With the very local time that not only Solana has, but also the Bitcoin network, Etherium, and many others like it? It stops..

Until when? As a rule, until a so-called social consensus is found;

So zero time is not a fantasy, but a reality.. Inside the DRS. It can be used either for good or for evil:

• For evil, for example, by combining it with any kind of Sybil attack: as long as the network is “offline” and the holders of the nodes negotiate among themselves, there is little standing between the average user and the spoofed transactions;
• For good, for example, if the zero (local) time of one network is the basis for creating a genesis block of another network;

I wouldn’t be surprised if in the near future there were networks that acted solely as temporal oracles. But let’s move on;

Inverted time

This time has several sub-states:

• Say, if we are talking about the Binance Hub hack, the time first stopped and therefore became zero within the system, and then it was restarted and went back several blocks. That is, there was a parallelization of the present through a return to a breakpoint in the past. Something similar happened to Harmony’s testnet after their bridge was hacked. That is, the inversion is that the blocks go in a straight sequence (the height is not broken), but the integrity of the chain is broken by removing a number of them to a certain value.
• The second approach of inverted timing is different: it is a characteristic of the Bitcoin network, for example: block X+1 appears in the main chain before block X. Why is this possible? Because the difference in timestamp between nodes can be up to two full hours and is inherited from the Satoshi era: “A timestamp is valid if it is greater than the median timestamp of the previous 11 blocks, and less than the network-adjusted time (+ 2 hours).. “Network-adjusted time is the median of the timestamps returned by all nodes connected to the network.”
• Finally, the third approach of time inversion. A blockchain is created with a block of, say, 1,000,000,000, and then there is a countdown. So then there will be a 999,999,999 block, then 999,999,999,998 and so on. Up to anti-genesis block: 0 (zero block);

The last case is very good for chronocaps. We’ll talk about them separately, but in short we can explain this is a multi-sig like Lighting Network, where the timing and other parameters are determined by the parties;

And here’s an example of a chronocapsule already in operation inside the Bitcoin network:

• Imagine that you sent a Bitcoin transaction into the future, meaning it becomes available on a certain block – a standard mechanism built into this DRS.
• This is how you set the first part of the time period. Then, knowing the main impact zones on the Bitcoin network – halving (210,000 blocks), recalculating complexity (2016 blocks), and so on – you create blockchain A, which counts down and creates blocks, say, every hour. Calculate that N blocks should be created in normal conditions during the period of sending the transaction. This is how you get the second number of the given range. For example, I made this test page on ChatGPT;
• Finally, you calculate the margin of error and assume that the transaction in blockchain X (let it be Etherium) will be made only if the absolute difference (modulo) between the block time in Bitcoin blockchain and blockchain A is no more than 48 hours. Otherwise, the transaction will wait until the next period;

Why would you want to do this? For example, so that no one can get certain assets before a certain time, whether it’s an inheritance or digital property belonging to a certain DAO.

In a stable network, right now, at the beginning of May 2023, transactions in the same Bitcoin are very expensive, and because of the congestion of the membrane, completion time can be days rather than hours;

But that’s not all;

Parallel time and the era of interoperability

There are three ways to achieve asset interoperability today:

1. Create a single environment of existence: whether it’s EVM blockchains or parachains inside Polkadot, hubs in Cosmos, subchains in Avalanche.
2. Create a unified toolkit that exists initially in any of the interacting DRSs, say, (w) NFT.
3. Create a system of oracles that can check the initial and final conditions of assets;

There is no contradiction in combining all the methods into one. And the best part: all three are direct confirmation of parallel time. Judge for yourself:

1. When you go to the first generation bridge and make an exchange, you actually freeze an asset in one network (let it be ETH in the Etherium network) and create its counterpart in another (let it be wETH in the Polygon network). Thus, it is important not only that the asset is frozen in the Ether network, but also that the time for this asset in the DRS with the allowed degree of detail of divergence is the same: otherwise how will we know that the money is back over the bridge? That’s right – no way;
2. In any multichain, by definition, localized and local time are not the same thing: localized time is global for the whole multichain, and local time is local only for a part of it.</nbsp

But what is the practical point of all this? Especially in the here and now?

MEV bots and their friends in time

Let’s start with an example once made public on Bits.media: “The Etherium alarm clock allows network users to plan future transactions in advance by determining the recipient address, amount, and desired time of the transaction.

Also, users must pay the gas fees in advance.”. So, wherever something appears with the word “in advance” or its synonyms – it’s tempi. But where does this happen all the time? That’s right: MEV bots;

What are they? MEV stands for Maximal Extractable Value: “In theory, validators should determine whether transactions are included in a block, but in practice it’s done by searchers, independent participants who look for potentially profitable transactions and launch their bots into them.

Validators still win a little bit, because scratchers put the phi gas above average, and the delta goes to the validators in return for the increased probability of including the transaction in the next block.

So in effect, MEVs work in accelerated time. And it’s not just arbitrage, but a fine-tuning inside the so-called “dark forest” of Ethereum, which is trying to be mastered both positively and negatively.

You can read more about attacks of this type at the link: if the topic is of interest, I will describe modern methods of defense and attack in a similar vein.

Especially since characters like jaredfromsubway.eth are becoming more and more popular;

But here’s what’s important to us:

1. MEV bots and their coping mechanisms – it’s a tempo practice that has seen a new development in 2020-2023 due to the growth of TVL and, as a result, projects that these very blocked funds want to divide among themselves;
2. Not only do MEV bots optimize transaction execution speed (and their behavior correlates with network commissions), but they also create a complex system of parallelized time:

1. Sharding and other scaling methods because they are initially targeted at parallel (local) time;
2. DAG Evolution. Solutions such as byteball, hashgraph, tangle and the like also support parallel (local) time in concept;
3. Multi- and/or crosschain solutions that are already facing all the conclusions of this paper in practice today;

Of course, these are not all the theses, but I think that’s enough to start with: write comments on the forum and in the telegram channel, share the article with your acquaintances and Do!