Page 38 of 38 FirstFirst ... 28363738
Results 1,481 to 1,497 of 1497

Thread: Inauguration Day, Inauguration Hooooooraaay!

  1. #1481
    ^^vv<><>BASTART
    Posts
    3,359
    Quote Originally Posted by Reverend Jones View Post
    One thing I noticed from those recordings is that Knuth only makes eye contact about 10% of the time. I guess he's pretty well re-purposed his visual and spatial faculties for abstract stuff and doesn't like those neural circuits to be thrown off by facial cues.

    Joke: How do you find the extroverted mathematician from the group?

    S/he will be looking at the other person's shoes.
    Yeah, he came across like a person mainly concerned with his own thoughts.

    Quote Originally Posted by Reverend Jones View Post
    You will do basic category theory anyway if you go far enough in algebra if only as a convenient generalized language to describe what is already clear from the concrete example from algebra.

    As for programming, you don't need to know too much about linear logic to take advantage of Rust's memory model, or know category theory to use ML's type system, or know a ton of logic to use prolog. But you probably won't be using math to come up with new such things if math isn't on your radar when you try to invent new **** in software from scratch, although to be fair most of that stuff was already done by the late 70s.

    If you want a job at valve and appreciate what they do, a simple guess says that you should probably learn how to program using linear algebra and computer graphics before expecting topology to pay off. Also, knowing probability, calculus, and combinatorics also is helpful pretty much in anything anywhere in life no matter what.
    I've seen some category theory, which is why I'm skeptical of it. As far as computer graphics, I've written some primitive 3D stuff, after a bit of linear algebra it's not hard to write code for 3D engines. Poorly optimized code, but it's conceptually not too difficult.

  2. #1482
    Ok, here's a non sequitur...

    https://www.nytimes.com/2017/04/22/opinion/sunday/is-it-time-to-break-up-google.html?action=click&pgtype=Homepage&clickSour ce=story-heading&module=opinion-c-col-top-region&region=opinion-c-col-top-region&WT.nav=opinion-c-col-top-region


    I’m under no delusion that, with libertarian tech moguls like Peter Thiel in President Trump’s inner circle, antitrust regulation of the internet monopolies will be a priority. Ultimately we may have to wait four years, at which time the monopolies will be so dominant that the only remedy will be to break them up. Force Google to sell DoubleClick. Force Facebook to sell WhatsApp and Instagram.
    It would be nice! The likelihood that Democrats will run on breaking up large tech companies in 2020 is literally about as great as the likelihood that Donald Trump will have a great record on climate change.
    Last edited by Eversor; 04-23-2017 at 05:00 AM.

  3. #1483
    ^^vv<><>BASTART
    Posts
    3,359
    Yeah, short of a grander economic crisis, it's a pipe dream to think anything will happen to Google.

  4. #1484
    Admiral of Awesome
    Posts
    13,826
    Quote Originally Posted by Reid View Post
    Yeah, short of a grander economic crisis, it's a pipe dream to think anything will happen to Google.
    Nothing will ever happen to Google short of the United States ceasing to exist.

    https://googletransparencyproject.or...olving-door-us

  5. #1485
    Quote Originally Posted by Krokodile View Post
    Oh man, I can't handle those big books full of words. I hope there's something like a booklet conveying most of this stuff in pictures.
    Dammit, the only reason he didn't do it is because I said he would! He knows what book I'm talking about! JAAAHHHHHHNNNNN!
    "I would rather claim to be an uneducated man than be mal-educated and claim to be otherwise." - Wookie 03:16


  6. #1486
    Quote Originally Posted by Reid View Post
    I've seen some category theory, which is why I'm skeptical of it.
    Category theory is just a language to describe very general relationships. Be skeptical of premature or misguided emphasis on expressing this generality, but algebraists and topologists use the language of category theory all the time to save ink and avoid writing proofs peculiar to the concrete setting they happen to be in.

    You can also use it to express intuitive conceptual relationships, and also to throw out a bunch of low level set theory boilerplate. In other words, it is a modern foundation for mathematical reasoning.

  7. #1487
    Admiral of Awesome
    Posts
    13,826
    Hey, look what hit HN this morning

    https://www.washingtonpost.com/poste...=.a84f2839cd62

  8. #1488
    TL;DR: category is most useful when you use it for what it was invented for

    https://mathoverflow.net/a/2284

  9. #1489
    It always feels kind of funny for me when I go back and forth between Jon`C's comments and reading through some of the Ayn Rand chaff on HN (in general, haven't read through the comments on this one) spreading FUD about stuff that questions their convenient entrepreneur morality.

  10. #1490
    I only glanced at it, but it looks like the top comment on the HN thread for that story is a No True Scotsman defense of 'free markets'.

  11. #1491
    Admiral of Awesome
    Posts
    13,826
    They're blaming regulation too.

  12. #1492
    Admiral of Awesome
    Posts
    13,826
    Tech startups basically fall into three camps:

    - Capital manufacturing (99.999%)

    - Criminal enterprise (Uber, AirBnB)

    - Actual company that aims to make **** and has to literally sue to gain the right to compete against incumbents (Tesla, SpaceX)

  13. #1493
    Admiral of Awesome
    Posts
    13,826
    "It's illegal to sell cars that aren't made by Ford" - Texas

  14. #1494
    ^^vv<><>BASTART
    Posts
    3,359
    Quote Originally Posted by Reverend Jones View Post
    Category theory is just a language to describe very general relationships. Be skeptical of premature or misguided emphasis on expressing this generality, but algebraists and topologists use the language of category theory all the time to save ink and avoid writing proofs peculiar to the concrete setting they happen to be in.

    You can also use it to express intuitive conceptual relationships, and also to throw out a bunch of low level set theory boilerplate. In other words, it is a modern foundation for mathematical reasoning.
    Well, it's generally of no help to analysts, and as far as I can tell, like you said, it's a good language, it gives you convenient ways to produce functions and/or objects in a category. I used hom and tensor functors a few times in the algebra course I just took (I'm attempting the qual this summer). But I've also been talking more with Baez, who does this stuff regularly, i.e. he's been doing stuff in applying categories to control theory and chemical processes, however it doesn't seem to be doing much? As in, it's a really cool universal language, but I don't think people in control theory or chemical engineering have a reason to care? So more what I mean is, I think I'm going to avoid that sort of category theory in itself, but seeing as I'm slowly getting shoehorned into the algebraist's way of life I'll have to know it.

  15. #1495
    ^^vv<><>BASTART
    Posts
    3,359
    Quote Originally Posted by Jon`C View Post
    Yeah, it's not a bad comparison to compare American corporate structure to an economic tumor. It's sucking the life out of the real economy and is going to kill it.

  16. #1496
    ^^vv<><>BASTART
    Posts
    3,359
    Quote Originally Posted by Reverend Jones View Post
    TL;DR: category is most useful when you use it for what it was invented for

    https://mathoverflow.net/a/2284
    Yeah, I know a bit about the influence of Grothendieck and the construction of algebraic geometry, but really, and alg geo is cool, but then again, jobs. One potential advisor I know does higher homotopy theory so that's a possibility.

  17. #1497
    As Alan Kay said to a HN poster asking about whether or not people at PARC were dropping acid in order to 'see' the future of personal computing: a little bit goes a long way.

    It's great that you've had a chance to talk to Baez about this stuff--his weekly finds archive are a goldmine for those looking for abstract connections between math and physics. But you have to understand that in my drug analogy, well let's just say that Mr. Baez is pretty much the Timothy Leary of category theory.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •