danblick

28 ms·

1. 241.▲
   by danblick 10y ago

   This reminds me of arguments in David Mindell's book, "Our Robots, Our Selves", particularly what he calls "the myth of full autonomy": """ The final myth is the myth of full autonomy. Engineers and r
2. 242.▲
   by danblick 11y ago

   Has anyone been able to learn a function for the spiral (Swiss roll) data that's as good as a human-designed function would be?
3. 243.▲
   by danblick 11y ago

   Am I the only one who read "Type I and II decisions" and thought "System I and II decisions" (from Thinking, Fast and Slow?). I like the idea of distinguishing reversible from irreversible decisions, I just wish he&#x27
4. 244.▲
   by danblick 11y ago

   I agree. I think these lines are directly related to the NYT controversy: """ If it’s a distinctive culture, it will fit certain people like a custom-made glove. The reason cultures are so stable in time is because people sel
5. 245.▲
   by danblick 11y ago

   "the AlphaGo algorithm, this is something we’re going to try in the next few months — we think we could get rid of the supervised learning starting point and just do it completely from self-play, literally starting from nothing."
6. 246.▲
   by danblick 11y ago

   I think he's really missing the point about the importance of self-play in Alpha Go. Human play provided a seed for training the system, but the thing that made it work was the fact that the computer could play an unlimited number o
7. 247.▲
   by danblick 11y ago

   FWIW, when the Kindle Fire launched, the major "breakthrough" feature was that it would pipe all your web traffic through Amazon's servers in EC2 (for browsing acceleration)... http://techcrunch.com/2011/
8. 248.▲
   by danblick 11y ago

   Don't take my word for it, read the article... """ Also, the devices are mainly designed to deliver digital entertainment, instead of being used as communication tools. That suggests there’s less of a chance users will h
9. 249.▲
   by danblick 11y ago

   This whole story seems overblown. Amazon removed a feature that nobody would ever use, and people imagined some conspiracy. A Kindle Fire is not a desktop computer, it's somewhere between a television and a crappy web browser. Amazon
10. 250.▲
    by danblick 11y ago

    I don't buy it. Look at Facebook. It was completely tasteless; they made it big by showing up with the right dumb idea at the right time. The article talks about Zuck studying the past in 2014. That was a long time after he'd made
11. 251.▲
    by danblick 11y ago

    I think this is an interesting topic that hasn't been talked about enough yet. I wonder a little about inter-generational conflict. It seems to me like older people (who may already own homes) and younger people (who often do not) oft
12. 252.▲
    by danblick 11y ago

    https://en.m.wikipedia.org/wiki/Cohesion_(computer_science)
13. 253.▲
    by danblick 11y ago

    I think coupling and cohesion stand out as fundamental software design principles. (I wouldn't say it's "interfaces" as I can imagine interfaces with tight coupling or low cohesion.)
14. 254.▲
    by danblick 11y ago

    See "The intention economy" by Doc Searls. https://en.wikipedia.org/wiki/Intention_economy
15. 255.▲
    by danblick 11y ago

    Seriously? You don't get to be a senior vice president by picking background colors.
16. 256.▲
    by danblick 11y ago

    I guess most research ideas here will focus on technology. Just a thought: how about supporting research on economics? YC is in a great position to spawn a small "think tank" focused on technology startups. A couple ideas: how ab
17. 257.▲
    by danblick 11y ago

    It seems that maybe startup employees end up carrying a lot of risk because they aren't really involved in deal-making between management and investors. Aren't there crazy terms (vested share repurchase rights or other arrangement
18. 258.▲
    by danblick 11y ago

    My impression: here's an article from people publishing outside of their area of expertise, in a crap journal that has not much to do with their topic, with shady statistics and the implication that people should avoid vaccines (which
19. 259.▲
    by danblick 11y ago

    This article seems to suggest science is hopeless when we can't conduct randomized controlled trials. I don't think that's fair: the methods used in econometrics for learning from quasi-experiments (e.g. natural experiments,
20. 260.▲
    by danblick 11y ago

    "Really regulation is critical for the success of self-driving cars. Nobody is going to want to buy self-driving cars if there are ones that get into lots of crashes..." If I were a car manufacturer and nobody wanted to buy my car
21. 261.▲
    Naming and Binding of Objects (1978)(web.mit.edu)

    22 points by danblick 11y ago|0 comments
22. 262.▲
    by danblick 11y ago

    This idea: "Obviously the spread of computing power was a precondition for the rise of startups." seems closely related to the concept behind the book "Design Rules" (2000, isbn 978-0262024662). Here's a summary fr
23. 263.▲
    The Economist: Technology Quarterly: New Materials for Manufacturing(economist.com)

    1 points by danblick 11y ago|0 comments
24. 264.▲
    by danblick 11y ago

    Caplan is making the argument that opening borders would bring huge economic benefits. In addition to the economic argument, I think there is a compelling moral argument to opening borders. Why should we deny equal rights (the right to wor
25. 265.▲
    by danblick 11y ago

    Yes, there are implementations for many of the type systems in the book. :) https://www.cis.upenn.edu/~bcpierce/tapl/checkers/ The code is pretty elegant. I think it's intended to be clear enough for st
26. 266.▲
    by danblick 11y ago

    This video of punching the weeds looks promising, but just a thought: maybe you could focus sunlight to burn them? :)
27. 267.▲
    by danblick 11y ago

    For anybody wondering: Joe Hellerstein is a professor at UC Berkeley: http://db.cs.berkeley.edu/jmh/ I'd say he's best known for work on database systems and distributed systems. ( I haven't heard the t
28. 268.▲
    by danblick 11y ago

    It's a shame they lose so many good agents out in the field.
29. 269.▲
    by danblick 11y ago

    This is neat. I bet Facebook or OkCupid are sitting on all sorts of click data that could be used to develop tools for helping people make their photos look better. (Even if, personally, I can't wait for a cultural backlash against i
30. 270.▲
    by danblick 11y ago

    I thought küçük ("coo-chook") sounded small because it reminded me of baby cooing. I'm not really surprised by this result. Some words are imitative. Probably words that are more imitative are evolutionarily selected to rema