Good for crafting in Minecraft

Common vulnerabilities on L1 are overconsumption of gas, multiple calls to counterparties' payments without verification, price pumping in a 15-second interval, various loops and long waits. Merkle trees are according to any L2, but we can also use our memory. And return values through Oracle (that’s what they call call-back sockets)

(Because there are so many complexities in distributed contracts that without formal verification, you can create something without errors only in such a visual language.)

The above technologies are good for crafting in Minecraft, but bad for real production.

In automated production, for objective reasons (due to the large number of products), they use the most inexpensive and easy-to-manufacture chip. It should only have a couple of functions. Supporting a critical channel to the Internet and performing business functions on bare metal.

The Internet will be at frequencies close to 10 Megahertz. Which imposes physical limitations on chip technology.

By the way, when it was necessary to make a non-changeable program, it was simply stitched onto the switch matrix. It was called microcode. In the future, such firmware will be based on ternary logic on a memristor matrix on the Galia Arsenide. They will be stitched with a laser. A prototype of a chip with a RISK-V core has already been released in Israel last year.

Of the shortcomings. It is worth mentioning races (ambiguity in time) and confusion with the owners (ambiguity in space). To avoid these problems, the tree must be built according to a clear rule, where the levels of coordination in time alternated strictly with the levels of coordination in space. And their combination was strictly responsible for one type of environment (or resources).

Well, from a hardware point of view, it’s not the right choice for a computing machine. I recommend the FPGA (62 Euro with case) and silicon ABC compiler for a complete product. There is a cheaper option 6502 (38 Euro with case).

An alternative is to create your own implementation based on the 6502 assembler. Because it requires minimal hardware. You can implement the entire project in the ABC flint compiler and save the tree of operations in the form of a Merkle tree or in a chain. Then our assembler programs will be evidence. We won't need to seal them. In addition, it will be possible to certify specialists.

Disadvantages - this is a rather costly event in terms of money and time.

Advantages - our contracts will not be vulnerable and will not be limited in terms of computing features.

Summary - The current state of affairs around contracts is a fuss of mice. We haven't focused on these technologies for 10 years. Perhaps it's time to clean up the mess.