Assembly vs Zig

One of the best ways to learn how a computer actually works is to work your way up from lower level abstractions. Assembly, being only a level above machine code, is low enough level to make it clear how the computer is actually performing a computation, including code flow and loops, but high enough level to not be excessively tedious for the type of small projects that a student would do at the beginning of their first programming class. Use of an assembler with macros can stretch this even a bit further.

There is no other way to understand a processor, so dig in.

Because of its natural syntax and low-level nature, assembly language programs are typically really small and fast.

Unlike other programming languages, in assembly language it is really hard to create a slow and over-bloated program.

A curriculum that involves an embedded component, such as an Arduino or a Raspberry Pi, can encourage students by allowing them to immediately connect their work with 'real systems'. Assembly is the ideal language for getting started with and understanding these devices, and since Assembly can be called from C, the code will still be useful if students move on to C later in the program.

Assembly language syntax is relatively uniform, and so there's less room for a student to get confused by obscure characters, or miss some meaning implied by structure, such as with scoping rules, or call-by-name/value/reference semantics. While there may be a lot of mnemonics to look up, most work involves only a very small subset of them.

It's safer than C, at least.

Zig is pretty fast, in some cases even faster than C.

Zig programs can import C libraries and export header files to be used in C programs.

No operator overloading, property methods, runtime dispatch, macros, etc.

Zig can execute code at compile-time, allowing for more performant and readable programs.

The Zig compiler can build artifacts for any Tier 3 Supported platform without additional downloads.

Produces binary smaller than C.