Re: 🍀🍀 whirlwind.money | Review Campaign | Write a review & earn up to $100! 🍀🍀
Thank you for a very detailed response! I addressed what I could below 
One thing I will address is about a singular fast input and output. Sure, the output may be from other coins. However, since the whirlwind hot wallet address is known, the same output and value would allow an outsider to easily deduce that the user deposited via Whirlwind, and received an older input. Hence, the level of privacy is lower/the chance of compromised privacy is higher. This is assuming a "not normal" threat level I suppose.
This is kind of irrelevant - Don't take it as a part of the discussion about the service itself, though I think it's worth noting. One could also argue that one sending Bitcoin between multiple wallets means that there is never 100% certainty that it is the same person, with each hop potentially becoming a new owner. However certainty is not always a primary factor, deduction of possibility is more so a primary factor. In this example, it would not be 100% certain that the hops are the same person, though it would be easy to deduce that it is.
It's not a disadvantage unless whirlwind is sanctioned like TornadoCash was. While right now it is tagged the same way as other apps, a sanction would mean that every coin involved in the service would be on high watch. Any usage where coins can be held from that point forward, would be a sure thing. Sanctions are possible if the app, its service and its coins can not be seized. If ChipMixer was operated as well as Whirlwind seems to be, I am quite sure that they would have been sanctioned rather than seized.
Until then, you are right that in this point in time, it's not really a huge disadvantage. It's more of a potential pitfall in the future if anything unfortunate like what I described were to happen for some reason.
I fully understand and am in agreement with where you are coming from in regards to breaking the trace via the improved methods that you offer (Notes, combining notes, time delay etc) in comparison to receiving "clean coins from centralized exchanges".
One thing that I did in my review was directed it toward the audience (potential users). For me to feel comfortable posting the review, I should aware anyone who is not so savvy of the risk of using centralized services after using a mixer. The principle would usually apply to all mixers. If you were to make a development that would seek to reduce AML score risk as well, definitely work on it on a fork of the current version that way the current version is not disrupted. If you did take that route and it actually worked 100% of the time, it might not increase privacy any more than the existing service but it would definitely increase usability. In fact, I would say that it would be a breakthrough, as this would essentially beat analysis companies.
Regarding the other points you made:
-It's impossible to connect your inputs and outputs with 100% certainty thanks to the way the system works. Even if you are using Fast mode with only one output, that output could have originated from any deposit since the start of the service. Obviously you know which ones your transactions are, but for an outside observer it's impossible to know for sure since Fast and Note outputs are identical and you can't know which option any user chose when he deposited.
-It's impossible to connect your inputs and outputs with 100% certainty thanks to the way the system works. Even if you are using Fast mode with only one output, that output could have originated from any deposit since the start of the service. Obviously you know which ones your transactions are, but for an outside observer it's impossible to know for sure since Fast and Note outputs are identical and you can't know which option any user chose when he deposited.
One thing I will address is about a singular fast input and output. Sure, the output may be from other coins. However, since the whirlwind hot wallet address is known, the same output and value would allow an outsider to easily deduce that the user deposited via Whirlwind, and received an older input. Hence, the level of privacy is lower/the chance of compromised privacy is higher. This is assuming a "not normal" threat level I suppose.
This is kind of irrelevant - Don't take it as a part of the discussion about the service itself, though I think it's worth noting. One could also argue that one sending Bitcoin between multiple wallets means that there is never 100% certainty that it is the same person, with each hop potentially becoming a new owner. However certainty is not always a primary factor, deduction of possibility is more so a primary factor. In this example, it would not be 100% certain that the hops are the same person, though it would be easy to deduce that it is.
-The fact that it's easy to notice an output comes from Whirlwind is by no means a disadvantage in my opinion, all other mixers have their clusters tagged by analytics firms.
It's not a disadvantage unless whirlwind is sanctioned like TornadoCash was. While right now it is tagged the same way as other apps, a sanction would mean that every coin involved in the service would be on high watch. Any usage where coins can be held from that point forward, would be a sure thing. Sanctions are possible if the app, its service and its coins can not be seized. If ChipMixer was operated as well as Whirlwind seems to be, I am quite sure that they would have been sanctioned rather than seized.
Until then, you are right that in this point in time, it's not really a huge disadvantage. It's more of a potential pitfall in the future if anything unfortunate like what I described were to happen for some reason.
At first glance that may sound "better" since you get coins with low AML risk, but you need to ask yourself what do you value more and why are you paying for a service of this type in the first place?
Getting low AML risk score coins, but them being easily traceable back to you? Or getting potentially higher AML risk scores with Whirlwind, but achieving real privacy?
I would choose the latter any day, and Whirlwind was built with that in mind. It's all about numbers
Getting low AML risk score coins, but them being easily traceable back to you? Or getting potentially higher AML risk scores with Whirlwind, but achieving real privacy?
I would choose the latter any day, and Whirlwind was built with that in mind. It's all about numbers
I fully understand and am in agreement with where you are coming from in regards to breaking the trace via the improved methods that you offer (Notes, combining notes, time delay etc) in comparison to receiving "clean coins from centralized exchanges".
If low AML risk score is really that important for users then I could offer this option too, without compromising the privacy. It would add more costs and complexity to the whole operation and that's why I'm not very keen on doing it, besides the fact that I don't believe it does anything better for your privacy.
One thing that I did in my review was directed it toward the audience (potential users). For me to feel comfortable posting the review, I should aware anyone who is not so savvy of the risk of using centralized services after using a mixer. The principle would usually apply to all mixers. If you were to make a development that would seek to reduce AML score risk as well, definitely work on it on a fork of the current version that way the current version is not disrupted. If you did take that route and it actually worked 100% of the time, it might not increase privacy any more than the existing service but it would definitely increase usability. In fact, I would say that it would be a breakthrough, as this would essentially beat analysis companies.