See the response below:
1)I can promise you that passing any meaningful peer review is far more difficult than responding to your "severe criticism." For us, it's not that big of a deal. For instance, this is the first study in the scientific literature to produce actual transaction data on a public blockchain. Unfortunately, we wouldn't call your opinion more relevant than that of the journal editor and peer reviewers, given your abysmal h-index and lack of any blockchain-based published scientific literature. However, we welcome any input and are ready to discuss any concern regardless. INNBC DApp is a completely free-to-use tool and there is no hidden fee for the user, who is only required to pay the network fee to generate the transaction. It is offered to users in the same way it would be provided by a nonprofit organization. It was not created to generate a profit but to provide a free tool for research scientists to easily store their research data on the blockchain, generating a permanent, decentralized record of scientific data. For someone to assume that contract code is not open source, it would mean they lack even the basics of blockchain technology, given that the contract's source code is publicly available for anyone to view, audit, and use from the contract page, as reported in the paper. The very open nature of a blockchain ensures it. It would not make sense to make a direct comparison with an off-chain indexing database such as OpenAlex, given that its intended purpose is vastly different from that of INNBC DApp. So you totally miss the point of this work. With an off-chain database, you have substantially superior performance; data is updated instantaneously (no need to wait for block time), storage is cheaper, and so on. But that is not what a blockchain is about. Blokchain is about ensuring that once some data has been added to the database, it will be virtually impossible to alter the data. The same cannot be said for an off-chain database. It does not provide the same level of security as a blockchain. It would be orders of magnitude less complex to alter the data of an off-chain database than thousands of nodes on a blockchain. And we are looking for a way to protect scientific data from any possible alteration or falsification. The goal of INNBC DApp is to take advantage of the unique features that you can expect from a blockchain that are not available from off-chain solutions. Features such as immutability, decentralization, security, timestamping, and identity, with a specific application for biomedical data. Performance is not the priority; it is rather data integrity. So that we can provide research scientists with an easy way to upload their data on the blockchain to ensure permanent availability for the data and proof-of-existence, as well as to protect authorship. The advantage of using a blockchain with respect to off-chain solutions is that the data can never be modified or tampered with in any way. Having said that, we provide a very thorough analysis of the on-chain storage cost by storing on the blockchain a large number of different types of data, including many complete research articles, at a relatively low cost. See Tab. 1 in the paper. With regard to data privacy protection, we specifically address this issue in the paper. Now, it should be mentioned that the main use case of INNBC DApp is to upload data from sources that are intended to be accessible to the public, such as data published in scholarly journals. This represents an ideal application for posting data on a public blockchain. But there are also specific applications that require special attention, including when handling sensitive privacy information that can be found in healthcare data. In such cases, we propose AES data encryption as a privacy-preserving processing technique for complete data anonymization. In fact, the AES encryption algorithm has been proposed to secure medical health records in hospital systems that use cloud storage and need to protect sensitive data.
2)Science is not democratic. Anything that is not peer-reviewed would be considered “pseudoscience” and incompatible with the scientific method. Peer review evaluates a study's validity, importance, and originality in order to serve as a filter and guarantee that only high-caliber research is published, especially in reputable journals. In this case, the data, tools, and methods presented in this work have undergone peer-review. We are thereby creating the foundation for legitimate decentralized science (
#DeSci). As a professional scientist who is also a member of the scientific community, I find it extremely unlikely that any credible scientist would accept any work that has not received peer-review and has not, therefore, been certified by the scientific community as valid and scientifically sound. And we would strongly advise against any projects that would define themselves as “DeSci” and lack peer review. Furthermore, I would highly advise against any project that considers itself to be "DeSci" and does not undergo peer review. We would be discussing pseudoscience, not science, in the absence of peer review. Users would never be able to trust a clinical application project if it refused to go through peer review. Peer review is crucial for DeSci, especially considering the crypto industry's already poor standing and lack of credibility.
3)This is totally incorrect. We developed a specific smart contract that can generate a transaction and store the desired data as “transaction input data," which is additional data that can be attached to a transaction. Specifically, a transaction has an “input data field” that can include two types of data: function call data and arbitrary data. Arbitrary data are not executed by the Ethereum Virtual Machine (EVM) and translate into human-readable text in UTF-8 format. Arbitrary data is where we include the data that we post on the blockchain with the INNBC DApp. As such, our design choice was to store the data not directly in the smart contract but within the transaction itself. We believe this is a clever way of storing biomedical data on a blockchain. There is no error generated due to special characters in the uploaded text. At worst, they would not be displayed correctly, but no error would be generated. The strategy of using Base64 encoding is implemented to allow for the storage of a very large number of different file formats (e.g., documents, images, audio, and video). In fact, we initially even stored the complete raw text of one of our previous research articles. Then, we uploaded the data as files for the successive experiments. Please refer to the paper.
“The data isn't encrypted so their tool can only be used to create references for existing data in the public domain - which defeats their stated purpose for building the app”
Clearly, you are not understanding the proposed use case of the DApp. First off, the main use case of this application is to upload data that is intended to be public, such as the data generated by scientific studies included in scholarly publications, for which there is no need to be encrypted. Let’s imagine a scenario where we have a research scientist who is performing groundbreaking research and, before sharing the data publicly with other scientists, wants to protect his data from being stolen by other people who could claim it as theirs. Normally, it can be challenging to produce such evidence, particularly in cases where work is carried out in large teams, which can result in disagreements over authorship. On the other hand, if the data, including authorship information, are recorded on a blockchain, the timestamp can establish that the data were created on a certain day and time, giving each author a clear and unambiguous reference. Specifically, recording data on a blockchain provides us with a temporal and permanent track of their creation, allowing us to identify subsequent falsifications and revisions. Additionally, rather than just creating references to the existing data, we are fully storing the data itself on the blockchain. In fact, in most blockchain tools proposed by other authors, the actual data is stored off-chain, with only a hash of the data stored on the blockchain. They use what is defined as a “blockchain anchor” instead, a unique digital fingerprint of given data generated by hashing the original data with a consistent hashing function such as SHA256. This representation of the data is stored on-chain, while the original data is stored off-chain. This approach can be used for proof of validity and proof of existence, but it cannot ensure the permanent availability of the raw data. By contrast, we propose the strategy of storing the raw data on-chain on a permissionless public network such as BNB Smart Chain, presenting a truly decentralized storage solution that takes full advantage of blockchain technology. Secondly, we completely address in the paper the scenario in which the researchers would want to upload sensitive data on the chain, as in the case of healthcare data containing sensitive information. To protect patient privacy, if this application is used for such data, it must be used in compliance with current regulations, such as the GDPR. In the paper, we provide a real-world example of how the privacy issue could be solved by encrypting the data with the AES encryption algorithm, making the information unreadable to the public without the encryption key. A whole section of the paper is dedicated to the protection of data privacy.
“It doesn't make economic/business sense to store data on BNB, an EVM chain with limited horizontal scaling. They should have at least enabled the opBNB L2 to leverage a rollup and mitigate cost data volume settled on L1. It would cost nearly $80 to save the word "test" using their dApp (screenshot #1). No wonder they didn't run a benchmarking trial.”
Easy cowboy. This is 100% inaccurate, but thank you for pointing this out. This happens because you have no balance in your BNB wallet. We measured chain storage costs as low as 14 USD per megabyte. Actually, to save the word "test,” it would cost $0.01 (see screenshot #1) and to prove this, we just performed this transaction (
bscscan.com/tx/0x50ad2358f55…) in which we stored the word “test” for a 0.000024075 BNB ($0.01) fee (see screenshot #2).
Secondly, we address in the paper the rationale for using L1 with respect to L2 for biomedical data:
”In this regard, we believe that layer-1 is more suitable for our DApp because it offers the greatest decentralization and data immutability, whereas layer-2 provides the benefit of extremely low gas costs but at the expense of decentralization. Given that we are proposing to store data on-chain, such as documents and images, we also need to consider the possibility that malevolent actors could flood the network with random data. Actually, layer-2's incredibly low fees might be a double-edged sword because they would make spamming attacks much less expensive as well, making the chain much more vulnerable to an influx of “garbage” data. By contrast, layer-1 would require significantly higher fees to generate the same amount of spam transactions, effectively discouraging such attacks. In order to allow for reasonable storage costs, transaction fees must be low enough, but not so low as to encourage malicious traffic attacks. For this reason, we believe the BNB Smart Chain layer-1 network is currently the best option for INNBC DApp.”
“Alternate Solutions:
@oceanprotocol,
@eas_eth,
@HyperfilesOrg, Hyperfiles Deets”
None of these solutions store the raw data on chain. As such, we must rely on storage solutions external to the blockchain for the availability of the documents without taking advantage of its immutability for the raw data. By contrast, with INNBC DApp, we take full advantage of immutability by storing all data on-chain, with a true decentralized storage solution.
“Comparatively, the INNBC claim to be "the first 100% on chain decentralized storage Dapp for biomedical and general purpose data" is frankly, complete BS and highlights just how flawed & ineffective peer-review is when managed by exploitative publishers. Their dApp is not 100% on-chain and it's definitely not the first to do what it does.”
This is totally inaccurate and given the very open nature of blockchain technology, anyone can verify that INNBC DApp stores the raw data on chain by simply browsing some of the many sample data we stored on chain. For reference, this is an image stored on the blockchain, 100% on chain, as Base64-encoded text:
bscscan.com/tx/0x6831da5deb6…
You can decode the text from Base64 back to a file and obtain the original image file (see screenshot #3). Anyone can perform such a procedure and verify that our claim is 100% correct.