Hacker Newsnew | comments | show | ask | jobs | submit | Jun8's comments login

A fiction version of these robots was described in Mr. Penumbra's 24-Hour Bookstore (http://en.wikipedia.org/wiki/Mr._Penumbra%27s_24-Hour_Bookst...), I suggest you check it out if you haven't already done so, nice, entertaining read.

reply


Borges at least made some interesting points (most of which are spot on, e.g. the film being too grand, etc.) but Sartre's review is populistic:

"Kane might have been interesting for the Americans, [but] it is completely passé for us, because the whole film is based on a misconception of what cinema is all about."

What does that even mean, assuming "us" means the French and not film connoisseurs anywhere in the world? Note that Truffaut, who was himself heavily influenced by Hitchock, later heavily criticizes general tendencies of French film of this era in a famous essay on (PDF: https://soma.sbcc.edu/users/davega/FILMST_113/Filmst113_ExFi...)

This xkcd strip seemed to be relevant in this case: https://xkcd.com/1112/

reply


Remember that Sartre was an existentialist. He therefore defines cinema to be an existentialist thing, and then criticizes Kane for not fitting into Sartre's definition.

But to paraphrase Shakespeare: "There are more things in cinema, O Sartre, than are dreamt of in your philosophy."

reply


Welles' F for Fake is practically a treatise on phenomenology, and it's nearly all in the present tense.

reply


as per the xkcd.. interestingly someone at the top of the chess game agreed with cueball:

http://en.wikipedia.org/wiki/Chess960

reply


It seems from the quote that Sartre simply finds flashbacks uncinematic.

reply


Zebra Technologies - Chicago, IL Data Scientist

We are looking to hire a full-time experienced data scientist for our team, which is working on next generation enterprise applications and research.

If interested contact me using the email in my profile.

reply


Mini Ask HN: Do you find it rude to crack one's knuckles in business meetings? I sometimes do this and wonder if it's bad form.

reply


I know a woman that cannot stand it. Doing it in internal meetings will get you a beating. If you were to do it in a business meeting where she was the client, she would probably never speak to you again and take the business elsewhere.

reply


If you do it once or twice then no big deal, but if you do it repeatedly, or a bunch of fingers one right after the other then it's rude.

reply


Some people hate it, so you might want to be aware of that.

Some people find it distracting, like tapping a pen, so you might want to be aware of that too.

reply


It's kind of rude when talking to someone or in a meeting, otherwise not a big deal.

reply


"Rude" things are 100% superficial, pretty much by definition of rude. (As soon as something becomes an issue in terms of actual consequences, we stop using the word "rude" and starting saying other things: unacceptable, not right, fucked up)

reply


I am wondering what value people associate with this functionality. One reason that I love HN is the long tail, e.g. low point, submissions. In the asymptotic limit of high point values (perhaps >500 or so) this is equivalent to reading cnn.com.

reply


I use http://hnapp.com/rss?q=score%3E10 to keep an RSS feed of HN submissions with score > 10.

The issue for me is purely one of volume. As much as I tend to like many of the "long tail" submissions on HN, even with the filtered 10+ feed I ended up getting over 70 items in the last 24 hours period, which makes for quite a bit of reading already. Any more and I'd be completely overwhelmed (I'll probably have to adjust the filter criteria in the future as HN inevitably becomes more popular).

reply


It doesn't have to be "at will", our bodies do a lot of processing without us being consciously aware of it. I think it is well within reason that the body may detect an abnormal embryo and respond to it.

reply


Interesting parallel to the flowing rivers is mercury in the best known Chinese tomb:

http://www.rsc.org/chemistryworld/2015/01/flowing-rivers-mer...

Speaking of using robots to investigate mysterious tunnels in tombs, one of the most publicized one must be the discovery of doors in a tunnel within the Great Pyramid: http://www.sciencealert.com/robot-captures-first-images-of-g...

reply


Detailed discussion:

http://physics.stackexchange.com/questions/2605/how-does-mas...

I was suprised this was not linked in the post.

reply


It doesn't always look like King practices the brevity he preaches in his own work, this is one of the points that he's often criticized (in a snickering sort of way in literary circles, see http://en.wikipedia.org/wiki/Stephen_King#Critical_response). As an example I thought the first two volumes of the Dark Tower series, especially the first one was great, but the volumes got thicker and thicker later. However, just now I found a comparison of approx. word counts of the series with other epic series and his books do not seem to be excessively long (http://cesspit.net/drupal/node/1869/).

What I found really neat about King's advice is its practicality. This, I think, makes his points applicable to young writers as well as young entrepreneurs, esp. (1) and (7) can be combined to pg's motto of "Make smt that people want."

-----


One difficulty with the hypothesis that babies are born with pre-existing knowledge about the physical world is the question about where this information comes from and how it is coded in the brain. Could it be that the initial physical rule s are random, similar to the Uncarved Block Hacker Koan, inspired by an exchange between Sussman and Minsky (), http://en.wikipedia.org/wiki/Hacker_koan? Or is the information somehow coded in the connection of neurons in the brain through evolution?

It seems to me that the role of evolution-derived innate world knowledge is relatively small to the mechanism of a powerful rule generation and update scheme, based on probabilistic analysis done in the brain's network. It could be that the rules are pseudo-randomly generated and this is innate knowledge that we use.

Here's the interesting thing: As more researchers see the power of such rule-update learning (e.g. see also this 2002 paper by Renee Baillergeon from UIC, PDF file: http://fitelson.org/woodward/baillargeon.pdf) I think the weight of Chomskian hypothesis of (in simple terms) "innate language faculty since language is too hard to learn" is decreasing. I found this review of Daniel L. Everett's book Language: The Cultural Tool by John McWhorter explaining this point vey well (http://www.nytimes.com/2012/04/08/books/review/language-the-...).

You might also find this TED Talk by Alison Gopnik on how babies think interesting (http://www.ted.com/talks/alison_gopnik_what_do_babies_think/...)

-----


My personal favourite hypothesis for language learning can be reused for this:

There /is/ some pre-coded information about these problem domains in the brain. It just /isn't/ coded explicitly, but implicitly as a bias of the corresponding learning faculty.

Even hard things can be easy to learn, if the learning mechanism itself is properly biased.

-----


"Those who contend that knowledge results wholly from the experiences of the individual, ignoring as they do the mental evolution which accompanies the autogenous development of the nervous system, fall into an error as great as if they were to ascribe all bodily growth and structure to exercise, forgetting the innate tendency to assume the adult form. Were the infant born with a full-sized and completely-constructed brain, their position would be less untenable. But, as the case stands, the gradually-increasing intelligence displayed throughout childhood and youth is more attributable to the completion of the cerebral organization than to the individual experiences -- a truth proved by the fact that in adult life there is sometimes displayed a high endowment of some faculty which, during education, was never brought into play.

"Doubtless, experiences received by the individual furnish the concrete materials for all thought. Doubtless, the organized and semiorganized arrangements existing among the cerebral nerves can give no knowledge until there has been a presentation of the external relations to which they correspond. And doubtless the child's daily observations and reasonings aid the formation of those involved nervous connections that are in process of spontaneous evolution; just as its daily gambols aid the development of its limbs. But saying this is quite a different thing from saying that its intelligence is wholly produced by its experiences. That is an utterly inadmissible doctrine -- a doctrine which makes the presence of a brain meaningless -- a doctrine which makes idiocy unaccountable."

- William James, The Principles of Psychology, 1890 (though he is quoting Herbert Spencer here)

I wonder if anyone has improved on this view in the century-plus since it was published?

-----


This has actually been studied with, for example, people who were born blind but somehow through some medical intervention regained their eyesight later.

Apparently, people who are born blind do not imagine objects getting smaller as they move further away. That is something that is taught to us. (I think I actually found the link for that here, so maybe someone else remembers the website)

Similarly, when regaining eyesight it takes a while before the brain can visually understand concepts such as being in front/behind something takes time to learn, or the ability to distinguish overlapping shapes (the whole gestalt thing).

See also this TED Talk by Michael Merzig, where he talks about how important it is that a baby has a fully functioning set of senses in the first year of its life, as that is the time that the brain learns to make sense (hah) of the senses. http://www.ted.com/talks/michael_merzenich_on_the_elastic_br...

Finally, you may be interested in this TED talk by Hod Lipson, about how they somehow got a robot to create a "mental" model of itself from scratch: http://www.ted.com/talks/hod_lipson_builds_self_aware_robots...

-----

More

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: