Accounting is actually not hard, what's hard is the fact that accounting as a field didn't adapt to the existence of negative numbers so they still talk about the effects of "credits" and "debits" based on underlying account type.
Physics may have gotten the direction of current flow wrong, but at least they were _consistent_ and didn't start talking about "source voltage" and "load voltage" which would have muddled KVL into `sum{delta_sources} = sum{delta_loads}` compared to `sum{delta_voltage} = 0`.
The real issue is that besides the famous Kleppmann article, it's hard to find material on accounting that sheds this legacy baggage.
From my experience as a programmer who did accounting for a while: Accounting is the easy part, but the borders to taxes are nonexistent (at least here in Germany). So when you do accounting you always do taxes to a certain degree. And this is not something that can be understood with logic, it’s completely arbitrary.
On software that was written by accountants: look no further than DATEV. It’s an abomination I would not inflict on my worst enemy.
Plenty of articles of programmers talking about accounting to form one side of that opinion, where are the articles from accountants talking about programming to inform the other side?
Physics may have gotten the direction of current flow wrong, but at least they were _consistent_ and didn't start talking about "source voltage" and "load voltage" which would have muddled KVL into `sum{delta_sources} = sum{delta_loads}` compared to `sum{delta_voltage} = 0`.
The real issue is that besides the famous Kleppmann article, it's hard to find material on accounting that sheds this legacy baggage.