IBM

A Very Bad Estimator (with Donald Knuth) – Numberphile

A Very Bad Estimator (with Donald Knuth) – Numberphile Podcast

#Bad #Estimator #Donald #Knuth #Numberphile

“Numberphile2”

Donald Knuth is unquestionably a legend of computer science and mathematics – but he is bad at estimation and grew up with a “rhinoceros attitude”.

Don Knuth’s homepage –

The Art of Computer Programming (books) -…

source

 

To see the full content, share this page by clicking one of the buttons below

Related Articles

34 Comments

  1. in the 1970s I spent a long time implementing the normalizing divide . I could shift over up to five leading zeroes after a subtract iteration. I made five bit slice shifter subtractor in s 244 two input nand gate array.

  2. I have heard two uses of a "rhinocerous attitude".

    One definition is an attitude of charging in before knowing all the facts. It is a tendency to attack a problem first and deal with the details or consequences later. It is the character belonging to the first to jump to action, following the call to “do it first, apologize later” or “it is easier to ask for forgiveness than permission.”

    The other definition is an attitude of resiliency, a persistence and will to succeed in face of seemingly insurmountable challenges. It is a confident attitude that can not accept the phrase ”it cannot be done”.

    I have been accused of both of these rhinoceros attitudes.

    Nice to hear from Don Knuth again. Peace and grace, sir.

  3. I am not sure that I am the only one, but the old win screensaver (labyrinth) is making me sick. Literally, it triggers my motion sickness. If there is more people like that, can you avoid it in future posts, please?

  4. Hmm, on ChatGpt being better at writing than its training data… I recall a finding that a large number of estimates (how many peas in the jar style) make a quite good average, even if some guessed are terrible. This kind of effect doesn't always happen, but it does seem to in some cases.

    Early innoceent Twitter was understood as a flow of each message flawed and insignificant but with the emergent effect of insight (something like that),

    A common comment on at least some AI is that we really don't understand it fully.

    Something can, it seems, happen at a certain scale that isn't obvious beforehand, the philosophy of quantity vs quality is not settled.
    🤔

  5. As a person of faith, I would love to hear more interviews with smart people of faith and how it has impacted their academic career. His comments about feeling like he had to defend himself saying that he's not one of the crack pots on Christian TV was very interesting.

  6. As always, I love the content .. but the video background on this one triggers my "bad/dangerous video" reflex, and I am only moderately sensitive to such problems, so I'd request that you change it. Sorry to have to report that

Leave a Reply