Slide 17
Slide 17 text
クラウドシステムの評価軸
Two scorecards. One output.
MARK’S RUBRIC
SCOTT’S RUBRIC
Correctness
Clarity
Does it actually do what it claims, under stress?
Could a tired human read this and trust it?
負荷がかかっても、主張どおりに動くか
疲れた人が読んでも、理解し信頼できるか
Performance
Confidence
How does it behave at p99 and at scale?
Do I understand WHY it works, not just THAT it works?
p99 やスケール時に、どう振る舞うか
なぜ動くのかまで理解しているか
Architecture
Recoverability
Will this hold up when the next person changes it?
When this breaks at 2am, can I fix it?
次の人が変更しても、持ちこたえるか
深夜2時に壊れたとき、直せるか
Same answer. Two scorecards. The gap between them is where engineering lives now.
同じ答えでも、評価軸は2つ。その間にこそ、いまのエンジニアリングがある