Jeff,
Lots of good points. And a lot of good work by the group. I am just a bystander lurking, and offer the following devoid of participation. I may be wrong but the following are based on some of my operating premises here.
Blockchain uses hashes, it is fully legible, immutable yes but not secret (e.g. encrypted), I don’t think it is correct to say it uses encryption and not put the overall crypto into context.
Not all distributed ledgers use blockchain, see https://arxiv.org/pdf/1607.01341.pdf where the author suggests alternatives. I think it is important to note that there will be a next wave of distributed ledgers.
Also it is the incentive structure, subsequent mining and distribution that provides the trust in bitcoin, this is not inherent to a blockchain but is part of bitcoin and can be part of other implementation.
Sal
From: dg-bsc-bounces@kantarainitiative.org [mailto:dg-bsc-bounces@kantarainitiative.org] On Behalf Of j stollman
Sent: Monday, February 06, 2017 4:52 PM
To: Eve Maler
Cc: dg-bsc@kantarainitiative.org
Subject: Re: [DG-BSC] Notes from BSC telecon Tuesday, February 2
I have some serious concerns about the draft report that I am reviewing.
I should probably start by apologizing for my late attention to this matter. But I have only just come our of a sequestered environment that did not afford me the flexibility to actively participate in the DG for the past 3+ months.
That constraint now lifted, I sense a fundamental definition problem with the report.
First the report is from the Blockchain and Smart Contract Discussion Group, but it appears to have a strong focus on a completely separate topic: Empowering the unempowered user in transactions with empowered parties. I don't know if it is the intention of the group to have this one issue dominate the report, but it does. It's a valid and important issue. But I think it is extremely limited as a theme. And the document addresses much broader context, but without a clear theme. We need a statement of purpose for the document.
Second, I disagree with the report's description of what "blockchain" is. The report seems to fall into the trap into which so many others have fallen: equating Bitcoin with Blockchain. It may be the case that all Distributed Ledgers use Blockchain, but there is no reason for a Blockchain solution to use a Distributed Ledger. I think it is important to more crisply define what blockchain is and avoid confusing the issue (as so many others are doing daily) by conflating blockchain with distributed ledgers. Blockchain is a genus; distributed ledgers are a species.
Last time I checked, I was not God, but in my view, "blockchain technology" refers only to an encryption technique that can be used to create an immutable ("difficult to alter?") record of transactions and the order in which they occurred. That's it.
Every other characteristic that we seem to want to conflate with "blockchain" is not inherent in blockchain. And I believe that it is a widespread mistake to confuse distributed ledger, distributed governance, and distributed validation with blockchain. Some blockchains may use the distributed ledger technique to increase the difficulty of modifying the blockchain by creating a large number of blockchains the majority of which will have to be modified in the same way. Some blockchains implementations may use distributed governance -- allowing any member of the blockchain an equal say in the rules under which the blockchain is run. And some blockchains may allow anyone with an interest to become a miner and participate in the validation of transaction blocks as they are added to the chain.
But blockchain encryption solutions can also be composed and useful that use a centralized model for both governance and validation. Distributed ledgers have a value in increasing the difficulty of altering the blockchain record. Distributed governance is valuable where there is little trust among blockchain members and no one wants to cede control to any one entity. And distributed validation is useful if I believe that some parties in the blockchain have an explicit inventive to create a corrupt record. Bitcoin provides the canonical use case for these. But some blockchains may be operated by an entity trusted by all members who maintains only a single copy (or multiple copies under the same ownership). For example, as long as there were independent audit provisions, I would be happy to let my state government operate a blockchain that tracked property ownership and rights. In such as instance, the state agency assigned to operate the blockchain may not even distribute the transaction validation (i.e., mining) outside the department.
This distinction is important to the report because the report's conflation of blockchain with these distributed features is a thread that runs through all of what I have read of the report. Distributed ledgers, governance, and transaction validation were implemented effectively in Bitcoin, along with blockchain. But teasing these concepts apart affords us more accurate insight into the problems we want to solve and greater flexibility to solve them.
It may be the case that to have distributed governance AND distributed transaction validation, the only technology currently available is blockchain. But blockchain has value of its own -- independent of these additional features.
Other points:
Trust and Distrust
In the discussion of trust and distrust, we define what trust isn't, but we don't say what trust is. We need to take a stab at this or drop the subject in the definitions.
It seems that the topic of how blockchain can be used to empower users who currently lack equal power has hijacked the emphasis of the report. I think that this is a valuable topic. But it might be best treated separately or included as a distinct section rather than running through the entire report. There report treats on many other topics that are useful broadly and are not limited to use cases of unequal power. Such use cases are only a subset of what blockchain can be used for and, therefore, they do not warrant center stage in this broad report. Perhaps they should be treated as a separate appendix explaining how blockchain -- along with other features implemented in Bitcoin such as distributed .." can fill a current need by disintermediating vendors who use their hegemony over a market to exploit their customers.
Economic Trust – Expectation that markets and currency valuation will operate in accordance with prevailing norms of supply and demand, and without non-compliant manipulation.
I would change this definition to:Expectation that markets will operate in accordance with prevailing norms of supply and demand.
I removed "currency valuation" because that is subsumed in "markets." I removed the end of this definition because I don't know what "non-compliant manipulation" means. Perhaps just removing "non-compliant" is enough.
PROVENANCE
Regarding the definition of Provenence:
Provenance – Information about entities, activities, and people involved in producing a piece of data or thing, which can be used to form assessments about its quality, reliability or trustworthiness. Metadata relevant or pertaining to the RP’s ability to evaluate the source of the attribute’s value.
Perhaps we could state this more simply as:
Information about the prior instantiation of a thing (e.g., person, asset, ownership of an asset) which may include evidence related the prior instantiation to the current instantiation, as well as substantiation of the credibility of the information..
Blockchain Types
I am not sure that there is a good reason to separate general purpose and special purpose blockchains. I recognize the distinction, but I am not sure that we make describe the value such a distinction has to the reader.
Smart Contracts
Smart Contracts are a technology that can be supported by a blockchain that records them. But I believe that the topics of blockchain and smart contracts are distinct. I disagree with claiming that smart contracts are a form of blockchain. They are merely a separate technology that can be supported by blockchain, in the same way the land records can be supported by a blockchain solution.
Smart contracts have been given a giant launching platform by blockchain because I don't know of any other technology that empowers them as effectively as blockchain. But I see no reason to try to conflate them as we do in calling them a type of blockchain.
The subsequent discussion of smart contracts is deeply intertwined with blockchain which exacerbates the perception that the two can be conflated.
I would describe "smart contracts" as "if-then logic that links inputs to outcomes". For example, a smart contract between a parent and college-age child might include the logic, "If that balance on my Bitcoin account #16 used by my child drops below $500, stop accepting new charges."
I am not sure that we need to say more about "what" they are. I don't think we need the technical details provided. I DO think it is very valuable to include the current discussion about the lack of legal enforceability such contracts may have.
"Human Involved"
I think that the broader concept that we are trying to express here is "ambiguous" whether or not the ambiguity is a human factor.
Legal vs. Smart Contracts
I don't see support for the claim, "Thus, the electronic nature of smart contracts is unlikely to be problematic moving forward."
That the UN is driving attempts to unify certain legal issues is a far cry from having universal enforceability across jurisdictions. One of the biggest problems we face in internet security is that certain countries allow internet crime to thrive. It is not illegal if it doesn't impact domestic citizens. I don't see such national/regional issues going away any time soon.
IPFS, Certificate Transparency, Verifiable Claims, Opal/Enigma, User-submitted terms, IAM
I am unclear what context prompts these topics to be discussed. I do not see any link in the discussions to know if they are relevant to blockchain (potential use cases?) or examples of smart-contract use cases. Some link need to be established to help readers understand the context that causes them to be discussed here.
There is some mention of blockchain w.r.t. Verifiable Claims, but I still don't understand why they are discussed in this report.
Common Accord
We provide a lot of interesting information about Common Accord. But we never marry this information with the theme of the paper. Why does Common Accord warrant space in this report? I am not claiming that it doesn't. But until we establish the connection, it is orthogonal to the theme of the report.
Consent Receipts
It appears that Consent Receipts are a Use Case for blockchain. Perhaps the discussion should be moved to "Use Cases."
Conflict Diamonds
This is a very weak example of how blockchain could solve a provenance problem. We spend two paragraphs describing the issue and two sentence describing how blockchain could help. But the discussion doesn't get down to the nitty-gritty of verifying the initial mining location. If this is not done properly the blockchain will only propagate the lie.
If we want an example of provenance, a better source might be a brand-name luxury good or high-value pharmaceutical. Both can be serialized using random numbers and their manufacture can be verified by the manufacturer themselves who have little incentive to lie. Blockchain then provides a method to track subsequent transaction through the supply chain to ensure that the serial-numbered product was never owned by two parties at the same time to ensure that the serial number at the end of the supply chain can be linked to the original manufacturer.
IAM
I believe that one important reportable outcome emanating form the DG is that recognition that blockchain does not appear to overcome some of the biggest problems with identity. As with conflict diamonds, it can track a claim (e.g., "this diamond was mined legitimately in South Africa" or "the account charged with the purchase of a $35,000 car was paid off in full." But does not provide a mechanism to ensure that the person claiming to be the owner of that account is, indeed, the legitimate owner who paid off the balance in a timely manner. Identity proofing and the linking of a credential or a claim to a person are issues not solved by blockchain technology.
Thank you for your indulgence.
Jeff
---------------------------------
Jeff Stollman
stollman.j@gmail.com
+1 202.683.8699
Truth never triumphs — its opponents just die out.
Science advances one funeral at a time.
Max Planck
On Thu, Feb 2, 2017 at 3:06 PM, Eve Maler