都要玩到幾hardcore先知有幾多
例如每個python run time都係一個c runtime
python有native function比你access個c runtime既data,又未至於anti pattern,自由到我可以access上一層call stack既 scope,甚至改到成個runtime 1+1 = 3 都得
oo應該比任何lang都flexible
pandas numpy 做table oriented programming
distributed computing 好易implement,我可以成個function send去個worker度run

,個worker開著就得,唔洗sync source code


