Thread by @ChainLinkGod

1/ After yesterday's discussion with the author of the recent @Delphi_Digital oracle report, who attempted to provide justifications as to why the report had inaccurate information regarding #Chainlink, it appears the doubling down is simply continuing

Bold move to say the least

https://twitter.com/mediodelphi/status/1404831363145080835

2/ DONs are not standalone ledgers (like was claimed), but are explicitly anchored to a mainchain, that is a fact that is backed up by numerous statement in the 2.0 whitepaper

DONs use layer 2 syncing mechanisms, a mainchain is required for this to work

https://twitter.com/mediodelphi/status/1404831366190075904?s=20

3/ This screenshot does not say what you think it does, it says you can leverage the nodes of an existing PoS chain to bootstrap the node set of a DON

But that DON is still entirely anchored to the mainchain, it is not a standalone distinct chain

https://twitter.com/mediodelphi/status/1404831370313089027?s=20

4/ This is a minor point (wasn't even a part of the three main issues I laid out in the original thread so interesting this is the first point), but any reader of the whitepaper would understand that DONs are not standalone chains & do not compete with blockchains (more like L2s)

5/ You did claim LinkPool stopped running their node on BSC, why claim you didn't?

Notice the hedging words like "mostly" and "likely" used for plausible deniability

In reality, LinkPool is running more feeds on BSC than ever, data.chain.link/bsc/mainnet/cr…

https://twitter.com/mediodelphi/status/1404831373265969152?s=20

6/ So again, the screenshots above show that you did exactly claim that, just in way to claim innocence later with more usage of hedging words here

Notice how their main investment Pyth is not mentioned here which also lacks off-chain aggregation

https://twitter.com/mediodelphi/status/1404831376478818309?s=20

7/ market.link is just one community run resource, and shouldn't be the single source of truth when you can check the on-chain data yourself with bscscan.com, which would have shown in seconds that your claim was untrue

https://twitter.com/mediodelphi/status/1404831380400377864?s=20

8/ And again, why didn't you check the on-chain data from bscscan.com or another explorer to verify?

You can see Linkpool's BSC node right here (bscscan.com/address/0x37fc…)

Why didn't you reach out to either the Chainlink or LinkPool teams?

https://twitter.com/mediodelphi/status/1404831383613288448?s=20

9/ In the thread, I stated the market.link UI had an issue and you should have checked other resources (this is what researchers are supposed to do)

Fair point on data.chain.link, but again why didn't you check bscscan.com?

https://twitter.com/mediodelphi/status/1404831387040075782?s=20

10/ The reality is that you had multiple ways you could have verified if LinkPool shut down their BSC node (check bscscan.com and/or contact the Chainlink or LinkPool team for context)

But ironically, for a report on oracles, you relied on a single source of truth

11/ The "context" in the report was either incorrect information on Chainlink, a shill for Pyth / API3, or had nothing to do with this specific point of trust-minimization

These rebuttals to the broad points I made are incredibly weak

https://twitter.com/mediodelphi/status/1404831390294740992?s=20

12/ 1. Explicit staking model does not have the weaknesses you tried to lay out in the report, you linked to Eric Wall's inaccurate post (claims the second tier has no economic incentive to be honest which is factually incorrect) without any counterbalance (a common theme here)

13/ 2. "Implicit staking is dumb" is weak

You do realize that Bitcoin and Ethereum are also secured by implicit staking? Nodes are paid in a native token, whose value is derived from the health of the network, so nodes are honest to not devalue the token
open.spotify.com/episode/5T7l9h…

14/ DONs can be made up of any number of oracle nodes, so why focus on only one end of the spectrum? You're comparing Chainlink in its current state to oracle networks in their future state, an unfair biased comparison

Not to mention the decentralization of data sources

15/ All Chainlink nodes being used in-production to power the data feeds that secure tens of billions of dollars are 100% public entities

Regardless, even anonymous Chainlink nodes have an on-chain reputation, you're conflating on-chain reputation with off-chain reputation

16/ You briefly mention DECO (but not other ZKPs supported or data source signing) after trying to justify the claim "One of their stated goals in the paper is trust minimization but our view after reading it is that most of their suggestions actually require greater trust."

17/ At the very end of the thread is the primary point I was making, Delphi Digital did not disclose that Chainlink supports data provider run oracle nodes

The fact is, you failed to disclose this info to pitch API3 and Pyth as the superior solution

https://twitter.com/mediodelphi/status/1404831393687932940?s=20

18/ Chainlink features the largest collection of data provider run oracle nodes out of any other oracle solution to date, using a percentage of the total nodes does not make sense here at all

Either Chainlink supports data provider nodes or it doesn't, it doesn't matter how many

19/ You didn't just "not focus on them", you didn't disclose the fact they exist or are supported at all!

And yet while there are zero data provider run nodes on Pyth and API3 out of a total zero live nodes, you do disclose that these projects do support such nodes

20/ Why didn't you hold Pyth and API3 to the same standards as you did Chainlink? If the adoption and percentage of data provider run nodes compared to total matters in terms of disclosure, you wouldn't have mentioned Pyth's and API3's support for them

Inconsistent logic

21/ If you try to use the argument "we compare oracle projects in their final state where API3 and Pyth have nodes", why not make it fair and take into account that the Chainlink network will also have more data provider run nodes (more than it already does today)?

22/ This chart perfectly encapsulates everything wrong with the report -> lying through omission to make your companies investments appear competitive

The repeated justifications, instead of a simple "yeah it could have been more accurate my bad", speaks on your intention here

23/ Eric Wall's post has been debunked, repeatedly in-fact, his points on super-linear staking and bribery are entirely incorrect as has already been pointed out

CO:

https://twitter.com/Crypto___Oracle/status/1389483937630621696?s=20

CLG:

https://twitter.com/ChainLinkGod/status/1366931120722309122?s=20

CLG2:

https://twitter.com/ChainLinkGod/status/1403880374988537859?s=20

https://twitter.com/mediodelphi/status/1404831397576167427?s=20

24/ This last point is just laughable

Of-course I'm biased, I will call out any BS posted about Chainlink, but I also don't claim to be a "independent research firm" who publishes comparison reports while having financial interest in two of the projects

https://twitter.com/mediodelphi/status/1404831399337775110?s=20

25/ If you were actually after the truth as claimed, you wouldn't be trying to justify why you published falsified information

I know you're trying to protect your work as a researcher, which is understandable, but you're digging yourself a deeper hole here

26/ Lastly, just as Eric Wall tried to do, when you can't make any valid points in a written form, the next step is to get them onto another medium to try the same exact tactic

There is nothing more to discuss here, you were proven wrong then and now, last thread on this topic