4 ms·

A few things: 1) tx settlement time is ~3 seconds on mobilecoin, p99 latency right now with single block finality. Eth and Btc are great but they aren’t that f

A few things: 1) tx settlement time is ~3 seconds on mobilecoin, p99 latency right now with single block finality. Eth and Btc are great but they aren’t that fast (for payments speed really matters). 2) with respect to privacy, the key innovation of MobileCoin is that when all of the systems are operational, there is no transaction graph stored in the ledger. The links between transactions are known only to the counter parties to those transactions. In the event of a failure of the Secure Enclave, links between transactions degrade to probabilistic links between transactions (and forward secrecy can be restored upon recovery of the enclave). The effect is a payment system that is both fast and privacy-protecting with no central authority, a quality not present in any other payments system I am aware of today. Does that answer your question? Oh, last and perhaps most important, because of our consensus design, we don’t use a ton of energy like btc and eth.

jpxw 6y ago
Why not use an existing currency which solves the problems you highlight, like Nano?
- josh2600 6y ago
  Nano doesn’t solve privacy to the degree we were excited about. Again, fast + privacy protecting is really hard to achieve, particularly if you care about fast tx recovery on a mobile device.
  - vocatus_gate 6y ago
    If you actually cared about privacy, you'd have just used Monero, and saved yourself 4 years effort building YAS (Yet Another Shitcoin). And no, speed is not the most important. As long as the user can see the payment incoming, it's trivial UX to say "Payment received. Will be confirmed and available for use in 3 minutes." edit: it appears MobileCoin is (allegedly) built on a combination of XMR + the Stellar consensus protocol? If true that's a slightly better scenario than I previously thought
    - josh2600 6y ago
      Monero isn't fast enough and doesn't support transaction recovery (it also has probabilistic linkage which MobileCoin doesn't due to our use of secure enclaves). We spent almost 18 months building MobileCoin Fog to solve the second problem (https://github.com/mobilecoinfoundation/fog https://github.com/mobilecoinfoundation/fog). It's a non-trivial stack of code to allow users to recover strings from servers they don't control without the operators of those servers being able to learn what strings are being recovered. Don't get me wrong, we stand on the shoulders of giants, but there's a lot of new tech here.
dbrgn 6y ago
I have not yet read in detail how you use SGX. But setting up SGX requires complicated processes and signing contracts and other paperwork with Intel. (Correct me if this is wrong.) Given that setting up the "systems" requires a huge effort, I assume that the architecture assumes a single central entity is running all these core systems, right? If yes, does the system rely on these core components to be up? If yes, how does it not rely on a central authority? Another aspect I don't yet understand: Traditional cryptocurrencies solve the distributed consensus problem through mechanisms like proof-of-work or proof-of-stake. What does MobileCoin use as a consensus mechanism?
- josh2600 6y ago
  https://github.com/UkoeHB/Mechanics-of-MobileCoin/blob/master/Mechanics-of-MobileCoin-v0-0-39-preview-10-11.pdf https://github.com/UkoeHB/Mechanics-of-MobileCoin/blob/maste... << This document has an extensive explanation of the consensus mechanisms and the attestation/enclave mechanisms in their respective chapters.
tomputer 6y ago
Have other existing privacy cryptocurrencies been considered? Grin is a lightweight privacy cryptocurrency using MimbleWimble. It uses a fair distribution (no pre-mine), with an emission of 1 GRIN per second: https://grin.mw/ https://grin.mw/
- josh2600 6y ago
  Yes the problem with Grin is that it allows perfect input/output linkage (see: https://github.com/mimblewimble/docs/wiki/Grin-Privacy-Primer https://github.com/mimblewimble/docs/wiki/Grin-Privacy-Prime... under information leakage). Grin also doesn't meet our standard of <5 second blocktime.

jpxw 6y ago

Why not use an existing currency which solves the problems you highlight, like Nano?

josh2600 6y ago
Nano doesn’t solve privacy to the degree we were excited about. Again, fast + privacy protecting is really hard to achieve, particularly if you care about fast tx recovery on a mobile device.
- vocatus_gate 6y ago
  If you actually cared about privacy, you'd have just used Monero, and saved yourself 4 years effort building YAS (Yet Another Shitcoin). And no, speed is not the most important. As long as the user can see the payment incoming, it's trivial UX to say "Payment received. Will be confirmed and available for use in 3 minutes." edit: it appears MobileCoin is (allegedly) built on a combination of XMR + the Stellar consensus protocol? If true that's a slightly better scenario than I previously thought
  - josh2600 6y ago
    Monero isn't fast enough and doesn't support transaction recovery (it also has probabilistic linkage which MobileCoin doesn't due to our use of secure enclaves). We spent almost 18 months building MobileCoin Fog to solve the second problem (https://github.com/mobilecoinfoundation/fog https://github.com/mobilecoinfoundation/fog). It's a non-trivial stack of code to allow users to recover strings from servers they don't control without the operators of those servers being able to learn what strings are being recovered. Don't get me wrong, we stand on the shoulders of giants, but there's a lot of new tech here.

dbrgn 6y ago

I have not yet read in detail how you use SGX. But setting up SGX requires complicated processes and signing contracts and other paperwork with Intel. (Correct me if this is wrong.) Given that setting up the "systems" requires a huge effort, I assume that the architecture assumes a single central entity is running all these core systems, right? If yes, does the system rely on these core components to be up? If yes, how does it not rely on a central authority? Another aspect I don't yet understand: Traditional cryptocurrencies solve the distributed consensus problem through mechanisms like proof-of-work or proof-of-stake. What does MobileCoin use as a consensus mechanism?

josh2600 6y ago
https://github.com/UkoeHB/Mechanics-of-MobileCoin/blob/master/Mechanics-of-MobileCoin-v0-0-39-preview-10-11.pdf https://github.com/UkoeHB/Mechanics-of-MobileCoin/blob/maste... << This document has an extensive explanation of the consensus mechanisms and the attestation/enclave mechanisms in their respective chapters.

tomputer 6y ago

Have other existing privacy cryptocurrencies been considered? Grin is a lightweight privacy cryptocurrency using MimbleWimble. It uses a fair distribution (no pre-mine), with an emission of 1 GRIN per second: https://grin.mw/ https://grin.mw/

josh2600 6y ago
Yes the problem with Grin is that it allows perfect input/output linkage (see: https://github.com/mimblewimble/docs/wiki/Grin-Privacy-Primer https://github.com/mimblewimble/docs/wiki/Grin-Privacy-Prime... under information leakage). Grin also doesn't meet our standard of <5 second blocktime.