I am aware of that we need some principles and methods to live in this world.
Knowledge: base the analysis on life(using)
Qunataty: great large number => Working set, JIT, Knowledge >> life: Infinitive dissipation
Complexicty/granularity: centric high/large => abstaction, causality
Complexicty/granularity: well-distributed, low/small => practice
Causality or practice
Material: Character, sound
onset-nucleus pair: yes for kernel
procedure or descriptive knowledge bias: English and kernel are both procedure knowledge
* 死知识, Life-relevant:learn in real-life, no dead knowledge
What is science by Feyman, do you know brown-throated thrush?
知识的“局限性” 碎片化的知识无用, 碎片化知识切割人生.
低品质勤奋的人, stereotype prototype
- Plunge into things;
Overconfident subjective understanding. – on larning kernel
- Literal learner, 以为学习是纯粹的文字游戏, 没有把知识同实际生活关联, 死知识;
Rote, it’s the bigset part I used to do.机械的摄入, 使自己成为智障.
除了浸沉于书籍之海不做别的事情的学者……最后则是完全丧失了为自己而思考的才能。除非有一本书在手里，他就不能思考。当他思考，他只是反应一种刺激即他所看到的别人的思想，结果他所做的一切都是反应.–Why I Am So Clever Section8
All by themselves my eyes put an end to all bookwormery, otherwise known as philology: I was released from the ‘book’, and read nothing more for years—the greatest favour I have ever done myself!—That nethermost self, as if buried alive, as if made mute beneath the constant need to pay heed to other selves (—which is what reading is!) awoke slowly, shyly, hesitantly—but finally it spoke again. I have never been so happy with myself as in my life’s periods of greatest illness and pain: you need only take a look at Daybreak or The Wanderer and His Shadow to understand what this ‘return to myself ’ was: the highest kind of recuperation!… The other kind simply followed on from this. –Human, All Too Human With Two Continuations
- 缸中之脑, 将生活维持在大脑刺激当中, 而不是真正的生活, fake life.
Enlightened suffer’s increasing of wisdom
nonego->分析知识, 分析自身-> life-relevant 有生命的活知识->循序渐进-> Practiced professional
working set, 有时可能需要简短的书来入门,然后就可以自己思考了。
From question, not zero
Showing-off: in front of interviewer. 装逼谷歌表示，该中心将在北京拥有一个专家团队，谷歌已在北京有数百名从事研发和其他工作的员工。谷歌AI中国中心将由李飞飞和李佳领导。李飞飞是斯坦福大学人工智能实验室的负责人，领导着谷歌云业务的人工智能部门；李佳是谷歌云业务的人工智能部门研发主管。
I have never thought about questions that are not real ones—I have not squandered myself. – N
Willpower: An idea is like a seed; once planted in a person’s mind, it’s growth is determined by how well it is nourished. Without lasting relevance it might wither and die or become ousted by other ideas that occupy a larger space in the mindspace of the host. Some ideas though, once planted, strike roots and grow from a captivating notion into a fundamental ideology that overshadows and influences every thought that blossoms in it’s wake.” – Homo Ludens or: Kojima’s Evolution of Mankind (Death Stranding), RagnarRox
People cannot see something wihout being denominated
循序渐进, gradual improvement, working set, gradient, content feeling
Occam’s razor and the transpoints
The key is deliberative practice: not just doing it again and again, but challenging yourself with a task that is just beyond your current ability
进一寸有一寸的欢喜 – Hu Shi
事实 我们要实事求是，现在像贴贴标语，什么打倒田中义一等，都仅务虚名，像豆腐店里生意不好，看看“对我生财”泄闷一样。又像是以前的画符，一画符病就好的思想。贴了打倒帝国主义，帝国主义就真个打倒了么？这不对，我们应做切实的工作，奋力的做去。 – Hu Shi
life prefer way
Active learning: is the first criterian for measuring learning effect,拒绝填鸭,cramming!
How much we understand the world isn’t better than how well.
不要成为跑马场, 叔本华晚年整理的论文杂集《附录和补遗》里的《论思考》第4节 别人的跑马场. or in 叔本华美学随笔.
Learning by doing: 从动手中学习, 创造知识.
What did Richard Feynman mean when he said, “What I cannot create, I do not understand”?
我们必须流下热汗 才能真正拥有父亲留下的遗产 –Goethe
Take the programs, modify them and enhance them. –Stevens
practice/experiment/doing, abstruction, causlity
Causality is abyss. Watch out origins.Causal reasoning
JIT e.g. kindle narrtion, soramimi
Output is better than input
Sound is stornger than memory to remember things.
Feynman and Descartes’s methods
Automatic Thinking, 人脑自动关联, 演绎, 可以被忽略很多不必要的知识.
Critical thinking, feedback
nonego, out of knowledges, never experinced, jump out
Watch out vicariousnees!!quick understanding, no complex brain labors, it’s easy.But it also brings fake.
MECE (Mutually Exclusive and Collectively Exhaustive).
Failsibility – Popper
Reflexicity – Soros/Statistics-based method training. –Firo
Tactical error. strategic error: If stracegic is too large, it will lead tactical error.
Open, 不带有色眼光的, 看待事情知识本身. 别说什么你不感兴趣, 有些内容与计算机无关. 这都容易, 造成认知对象的消失, 从而偏见与盲目.