What can we take away from this sequence marked by the announcement and the terms of allocation of the token ?
Honestly, we didn't expect things to go so badly. We were prepared for the operation to generate frustration among those who discovered their ineligibility, but we probably underestimated the importance of certain criteria. With hindsight, many of the complaints seem justified. I'm thinking in particular of the minimum threshold required in the wallet at the time of the snapshot and the exclusion of cash. These errors are our main faults.
However, not everything is negative. What are the positive points to remember?
We have been warmly welcomed by the Ethereum community, including some very influential members. Our project is the first to have allocated large quantities of tokens to its stakers. As Starknet is based on Ethereum, it seemed natural to reward the stakers, who are essential to the security of this blockchain. However, the complaints largely overshadowed the positive aspects. The perceived innovation in the design of the airdrop contrasted with the discontent expressed, making communication management particularly arduous.
Do you consider the launch of STRK a failure?
It's premature to draw any definitive conclusions. But at the moment I'm inclined to see the negative aspects more, because of my involvement in the post-announcement communication. I've been heavily criticised personally, and screenshots taken out of context have been circulated. Let's take a step back and assess the situation in a few weeks' time.
What was your role in the launch of the token?
I didn't have a specific role for the launch. I was mainly in charge of the ecosystem and relations with developers, assessing how they felt about the announcements. According to my observations, they were satisfied. For some, it even changed their lives, given the amount of money they received. I was touched by testimonials from Africa and Latin America. These allocations seem fair to me, given their commitment over the last two years. Nevertheless, the frustration of some may have been exacerbated by seeing these successes.
Developers have been particularly well looked after, why?
We are often perceived as an ecosystem made up solely of developers, but in reality we have no choice but to prioritise this aspect to increase our value in the long term. Unlike our competitors, who can reuse bits of existing infrastructure, we are not EVM (Ethereum's virtual machine) compatible and we have to build everything from scratch. That's what I think is a bit unfair about what we're being criticised for, because we're putting our resources and energy into what will bring value to Starknet in the future. We thought that to do otherwise would have been artificial.
Have Starknet forgotten that a blockchain is useless if no one uses it or adds liquidity to it?
A blockchain without a user is nothing, we've never disputed that. But as far as we're concerned, it's mainly a timeline thing: we initially chose to focus on developers building solid things around the protocol.
Some developers have received quite a few tokens for correcting a spelling mistake on Github, do you understand that this is frowned upon?
I do indeed notice that there are a few missteps of this kind, but I don't think there have been many. To give you some context, we used a benchmark data source (Electric Capital's Developer Report) without making any changes. Perhaps we should have adjusted it, but it's tricky when you're relying on such a source.
One person received 360,000 STRK with just three transactions, while others with hundreds received nothing. What is the philosophy behind this?
I know this case well and there was indeed someone who received 360,000 tokens. It was a mistake because he should have received half as much, so we contacted this person and he agreed to return half. To be transparent, STRK 180,000 was the maximum amount that early contributors could receive. I'd also like to add that the number of transactions was not taken into account in this allocation pocket. This was aimed exclusively at developers, while users were assessed according to other criteria.
Many users were disqualified because of the criterion of having 0.005 ETH in the wallet on the day of the snapshot, why was that?
In retrospect I think it wasn't a good idea. But this criterion had two functions; it wasn't chosen on the spur of the moment. Firstly, it serves to complicate the task of those who develop robots to make automatic transactions: by obliging to leave a minimum amount, we protect ourselves from cyber attacks. Secondly, without this minimum amount, 10 or 20 times as many addresses would have been eligible, but the amounts paid per wallet would automatically have been much lower. In the end, there would have been complaints about this too. But perhaps a small number of tokens distributed would have been better than none at all.
If you had to do it all over again, what else would you change?
I think we'd also include the cash, i.e. the LP tokens that are in the wallets. But this is a tricky choice because we wanted to remain neutral. When you take into account the LP tokens of certain projects, you have to make choices among them, and we didn't want to favour one Starknet application at the expense of others. Personally, I'm in favour of a clear separation between the protocol and its applications, but I agree that contributions that are objectively beneficial to the network should have been valued. And placing cash within it is part of that, as is holding stablecoins.
Were there any insiders in the light of what we were able to measure?
None. I was very touched by this criticism because we place integrity above all else. I'll give you an example: I wasn't authorised to receive the airdrop even though I was eligible on several criteria. No StarkWare employee was eligible and that's perfectly normal.
Yet it's always possible to set up a discreet address or give out a few tips...
Technically it could have been possible, but I assure you it didn't happen. Most of the people who received large allowances found out at the same time as the public. And everyone could participate in this special programme reserved for developers.
What is your strategy for attracting users and liquidity from now on?
The Starknet Foundation has mandated several committees, including the DeFi Committee and the Gaming Committee, which will work on these subjects and think about attracting users and capital. DeFi will be announced very shortly. This will come later for gaming, but it's a vertical that we believe in very much.
Will there be a second airdrop to respond to the critics?
There will be a second airdrop and perhaps others. In addition, tokens that have not been claimed by 20 June will also be redistributed.
So we can expect another Starknet airdrop from June?
It's possible.
What kind of criteria will be used?
We're going to learn from this week and change a lot of criteria. We can therefore assume that it will be simpler to qualify, but we can expect the amounts paid to each portfolio to be mechanically lower.
The total value tied up (TVL) in Starknet is only $55 million, so how do you explain this amount compared with Starknet's capitalisation after the token was issued (around $18 billion in pre-market terms)?
In November we suffered a leak on social networks claiming that the snapshot had taken place. We could have denied these allegations, but in our desire to be as transparent as possible we confirmed the information. As a result, activity on the network dropped by 80 to 90% in a few days.
How do you intend to increase this TVL?
Until now, we haven't actively worked on this subject. Perhaps we should have. Our challenge now is to create a favourable environment for value-generating applications. However, we need to remain cautious about certain TVLs announced by competing projects that include their own token in the calculation, artificially inflating the amounts. The important thing is to aim for organic and genuine growth.
It is regularly said that Starknet has the most advanced technology on the Ethereum layer 2 market. Yet history has shown that it's not necessarily the most brilliant technology that wins out. Do you fear this outcome?
It represents a real risk. Sometimes I wonder whether the market will recognise the quality of our infrastructure. The network effect will play a crucial role in our success, but the fact that Starknet is not EVM compatible is a handicap in the short term. On the positive side, our ecosystem is made up of very committed people who have invested time in learning a new programming language. We rely on the development of consumer applications offering an excellent user experience to set us apart. I can assure you that Starknet excels in this area.
What is your main competitor?
Arbitrum is our main competitor. Although this project is not exactly in our category, as it uses an optimistic rollup while Starknet is a ZK rollup, at the end of the day, we are both Ethereum layer 2s. Arbitrum has an excellent network effect, with a high TVL and number of users. I would then mention ZkSync as another notable competitor.
Despite the promises of efficiency from your Cairo programming language, Starknet's fees remain higher than most of its main competitors, such as ZkSync. Will there be a reduction in the future?
At present, 95% of layer 2 costs come from data availability, i.e. sending data over Ethereum. ZkSync is indeed more efficient than Starknet in this area. However, we are clearly better at the remaining 5%, which is crucial because the next Ethereum update (EIP-4844) will considerably optimise data availability. After its implementation, scheduled for March, Starknet will be able to demonstrate its full potential.
Would you be able to use solutions like Celestia to delegate data management?
Starknet recently entered into a partnership with Celestia (read our analysis of the project and its token) , but this relates to future layer 3s that will be built on our platform. For the time being, there are no plans to integrate Starknet with Celestia for data management. The community is divided on the question of whether you can be considered an Ethereum layer 2 if the data is not stored on its blockchain. We have chosen, for the time being, not to take this step, but that doesn't mean we never will.
How do you see the L2 war ? There are around thirty competing projects...
We can expect intense but healthy competition. The developments of each project contribute to strengthening the ecosystem as a whole. Although we are following different paths, we all share the same goal: to improve Ethereum's scalability. Currently, L2s are at the heart of innovation, because the main blockchain is so big that every change becomes extremely complex to implement.
The usefulness of layer 2 tokens is often questioned because of their centralisation. Do you have any idea when Starknet will be fully decentralised?
We intend to stand out on this point. Starknet will be the first L2 to allow the use of its token to pay transaction fees, a different practice from others that require the exclusive use of ether (ETH). This approach, seen by some as a misalignment with Ethereum, gives the STRK a more significant role than that of a simple governance tool. This option will be available from 20 February. Thereafter, STRKs can be staked to secure the network and strengthen its decentralisation. We are looking at at least 12 months before this becomes feasible.
The vesting period for the team and investors is particularly short (two months after launch). Why is that?
This is a point that's generating a lot of discussion at the moment, but we've always been transparent about the fact that the token's creation dates back to 2022. Although the vesting period may seem short compared to the launch, the STRK should have been introduced well before February 2024. In the original plan, this period would have seemed longer. It is important to remember that some investors have been supporting the project for a long time, as StarkWare was founded 6 years ago.
Do you understand that this could destabilise your ecosystem?
Absolutely, this is a legitimate concern. We're taking this matter very seriously and are working on mechanisms to reassure the community about a possible mass sale of the token. We will be communicating on this subject shortly.
What are the most promising applications in the Starknet ecosystem?
Many applications are starting to reach maturity. I could mention AVNU (a DEX aggregator), Ekubo (a DEX), Nostra (lending) and, in general, everything to do with gaming. The 'Influence' game is due for release shortly. We're also seeing good progress on the RWA (real-world assets) side with Carbonable, which is focusing on decarbonisation.
Why choose to build an application on Starknet rather than another blockchain?
Many advantages stand out, but the main one is that Starknet represents very new territory. In the world of EVM-enabled L2s, it's easy to duplicate existing models, which intensifies competition between players. Starknet encourages more innovation because it means starting from scratch. It also makes it easier to stand out from the crowd because there is still a lot to discover. What's more, our technological features make it easy to perform large calculations directly on the blockchain, a major advantage for gaming and artificial intelligence. We also anticipate the possibility of exploring new concepts in decentralised finance (DeFi).
Isn't it problematic that the large DeFi entities, such as Uniswap or Aave, are not present on Starknet?
We notice that integrating these large applications on Starknet is not a priority for these projects. This is understandable, as they are following a roadmap that encourages them to develop on as many networks as quickly as possible. Given the ease and low cost of deployment in the EVM ecosystem, Starknet is not a priority. However, we believe it is beneficial to encourage new teams that can bring a new dynamic. Nevertheless, I am continuing to talk to these blue chips in the hope that they will join us in the future.
What is your overall view of the airdrop strategies adopted by projects in recent months?
The current airdrop model is close to breaking down. There have been a lot of five- and six-figure airdrops, and people are starting to see this as the norm, establishing a kind of implicit contract between protocols and users. This creates a vicious circle, as projects feel the pressure to hand out large sums, while airdrop hunters move from one project to another. I understand everyone's interests, but this situation is not sustainable. In the end, the ICO model had its merits. Regulation has put an end to this type of funding, resulting in a more problematic system.
The Celestia project did, however, pull off a fine airdrop at the end of 2023, proving that it is possible to satisfy users and get them involved over the long term...
Without getting too far ahead of myself, I'd say it was one of the most successful airdrops of recent years, following the example of Arbitrum. The Celestia team have done a remarkable job, managing to satisfy a large majority of their community.
The points system seems to be becoming a new standard for airdrops, as we can see with EigenLayer (read our analysis of the project) or EtherFi. Why didn't you go for it?
The idea appeals because it offers a degree of transparency and gives users a sense of control over what they can expect to receive. However, it is possible that some projects adopting this system do not issue tokens, or offer an airdrop not linked to points. At the time our airdrop was designed, this trend was not yet in place. This will be a very interesting aspect to observe in the future.
What advice would you give to a project considering launching an airdrop in 2024?
First, learn from our mistakes (laughs) . I recommend always questioning choices, involving as many people as possible to model the community's expectations as closely as possible. It's tricky because an airdrop strategy needs to remain secret to be effective. The more people you inform, the higher the risk of leakage. In general, I don't think you should over-complicate things. I remember very elaborate discussions around this operation, perhaps we should have simplified things. I must also admit that we were probably out of touch with reality on certain points. We have the humility to acknowledge that today.
