Interesting read. Reminded me of the Trinity 3D manifold visualization tool which (among other things) let's you explore the hyperspace of neural networks: https://github.com/trinity-xai/Trinity
Imho a „digital euro“ (ie retail CDBC) will remain an utopia: Knowing who transfers money on the ledger of the central bank is very important for the central bank to remain credible. KYC is the banks job in this, so the bank takes the hit in fraud cases, never the central bank.
If central banks loose their face, their currency/country takes the hit.
Agree. I also like the relative dates. E.g. on Github where it’s useful to get an intuition very fast of when files in folders/projects were last updated. I don‘t care about the actual dates in this moment, the fuzziness of the relative date is the approximation I‘m looking for.
And the central bank does the KYC? And it also decides who should get money and who not on every loan? Doesn‘t scale very well this model, no matter if DLT is used or not.
Same for me. Jumping into the code can be fun at times. But if the software is not modelled well, a lot of bugs will orginate from structural issues. When you see those, the bugs become anoying to fix because you basically know new ones will pop up eventually since the root cause is not adressed.
Applying concepts like DDD will be of very little value without a profound understanding of the domain at hand. This is mostly established by non-coding activities.
Plus one for C4 Models. Most important aspect of it: When you do lines and boxes in a diagram, put descriptions on the lines, not just the boxes. Sounds simple, but is often forgotten. The result is people seeing UML, making all sorts of assumptions.
Anyone thinking about earthquakes when going that deep?
In Switzerland they had to stop some geothermal projects after triggering earthquakes while drilling deep.
Fully agree. The current/legacy financial infrastructure is actually 10 times more decentralized than any DLT: DLT is the same software run on multiple nodes.
SWIFT in contrast is a protocol, that anyone can implement. So its hundreds of software teams, building and operating their part of the infrastructure and integrating with each other. The SWIFT protocol itself is maintained by the member organizations, anyone can bring in proposals.
The complexity of it does not come from technology, but the regulatory requirements which differ vastly across the world.
Big shoutout to all the wayland contributors: Thanks a ton for all the hours you spend on the project. Please keep up the good work and don‘t let the haters get to you!