3 ms·
Because idempotentcy can be cheap at the application layer, so why try to solve something we know can never truly be solved?
by inlined 9y ago
Because idempotentcy can be cheap at the application layer, so why try to solve something we know can never truly be solved?