blackkettle

This maybe also dovetails with deep mind (?) or openai (?) article yesterday talking about adapting and updating general models with limited data for significantly different tasks.

The phone number validator on the signup page does not work correctly. Doesn't recognize or validate my valid Swiss phone number in any normal variation although it does correctly change the little flag to Switzerland.

So basically this is the prequel to Children of Men?

actually makes pretty good sense as a coping mechanism I guess. decreased memory function might help erase bad experiences, thereby decreasing stress - unless your job depends on abnormally high levels of recall... probably not a big issue in days gone by.

I just saw this yesterday after being cajoled into it by my wife. I had very low expectations, and was pleasantly surprised at how misplaced they were. It was a lot more compelling than the rehashed installment from last year.

we're just quibbling here...

> You can argue that some ongoing tragedy of the commons also has externalities.

sure, two of the preceding opinions [wikipedia, mit], argue that the tragedy of the commons is a specific type of negative externality, rather than the other way around. one [north carolina] gravitates more towards your interpretation.

the fundamental issue remains though: our conception of time and consequence is woefully limited in scope.

> imposes a negative effect on an unrelated third party

yes, and i'd argue that is certainly the case here.

but the two are closely related.

http://ipeatunc.blogspot.com/2011/08/negative-externalities-...

i think the above is a bit over the top for my taste:

> Secondly, is it really clear that people in Bangladesh (say) have a rights claim to an climate environment that is not two or three degrees warmer than it is right now? How did they acquire this right?

... because yes, I'd say they do have that right, but the point about the expectations being tied to some kind of property rights [physical or fundamental] as a differentiating factor seems to be a strong one.

here they describe the tragedy of the commons as a specific sub category of negative externality (in direct contradiction to the preceding link):

http://web.mit.edu/11.203/www/econ/index.php%3Fid=14.html

In economics what you are describing is typically referred to as a Negative Externality (https://en.wikipedia.org/wiki/Externality#Negative). It's hard to reframe these for the obvious reason that the parties involved directly in the transaction benefit from ignoring them (at least in the short term...)

tldr: It was never worth $740M to begin with.

> You can also go to art school to make those social connections to the larger art world - the art world is nothing but a huge web of social ties. I found that part of it pretty dark and disgusting, really. I just wasn't willing to stab so many people in the back to get ahead.

This is true of every profession. Politics is a part of human life. Academia is absolutely rife with this, business obviously; and you can argue that engineering and certain other professions are 'more' merit driven, but in the end this is only true at the bottom of the pile. Once you hit the threshold where you have sufficiently competitive domain knowledge, then the value of your position is determined by a balance of factors which begin to have more and more equal weight:

Can you get along with anyone? Do you have a strong network which can help drive business? Can you sometimes make hard decisions? Can you formulate a larger view of the landscape?

Of course these things count to a varying degree at all stages of a career, but there is either no such thing as a 'pure' meritocracy - because this is a fundamentally one-dimensional, naive view of the world; or it is a meritocracy by definition, and 'merit' is measure of outcomes.

FogCreek uses their own in-house programming language? Is there more of a story there?

Welcome to Psychohistory. Thanks Hari.

Why do think any of those labels are critically important? They are entirely derived from human desire to understand, process, and define a taxonomy for language. There is absolutely no reason to think that they are inherently necessary to achieve the goal or high fidelity translation (or anything else in this domain). Similarly they might be meaningful but inferred naturally through the learning process.

If your goal is to map machine learning processes to established dogma, this makes sense; if it is just to build an effective learning machine that solves an extant problem, then it's not.

Could you explain a bit more why - in practice - you think Bazel is 'next generation' compared to Make?

Every time I've encountered Bazel thus far in the wild I've found it an atrocious experience.

I haven't yet found anything I prefer to simple Make, and even auto tools while awful, are at least completely a known quantity.

Looks well put together, well documented. Quickstart tutorial was easy to follow.

Why should I think about using this instead of (or in combination with?) the plethora of other similar offerings out there?

The subredditsimulator subreddit does this for both articles and comments and is restricted to bots.

It is still mostly inane garbage but the content and gems have improved steadily over the past year or so. Interesting experiment in any case.

I think these two are the 'real' crux of the matter:

* Almost everyone believes that they are right and that their values will make things better, even people who strongly disagree with each other.

* Almost everyone believes that their own values and beliefs are well-supported or obvious, even people who strongly disagree with each other.

These two things are incredibly difficult for people to come to terms with. The corollary to these, which in fact leads to the real problems is:

* Almost everyone finds it extremely difficult to believe that those who hold beliefs or a belief system fundamentally at odds with their own could mean just as well, or even more challenging, be equally 'right' in the context of that system. That is, "there's more than one way to skin a cat".

it's absurd in any practical sense. but it's a great story to keep the rank and file mostly satisfied most of the time.

politics in the end is a part of life; it should not dominate or overshadow goal oriented work, but it is a natural, unavoidable consequence of people holding different viewpoints.

on the other hand, all the openness strategies they suggest are great, and these do work well in practice to mitigate conflict [but politics does not equal conflict].

I don't get the thought process that locks this possibility out. Imagine you are a simulated entity inside simulated universe - v100000 of Eve Online.

There are rules that you can see and access, and those you cannot. Both are in some sense irrelevant to your own perception of truth because they are determined outside the bounds of your empirical experience.

What good evidence do you have to refute such an explanation [surely its likelihood increases in line with our own ability to create the 'next turtle down'].

The books are interesting, but I think that the reason that a lot of people actually 'swear by them' is because they give 'smart people' a very nice excuse for any failings they might see in themselves.

You have to use a decent acoustic model - not the one in the demo. If you do I think it works 'pretty well' as a proof of concept. That said I'm not recommending Sphinx as a recognition framework, it is way behind the times in 2016, but this is the only 'in the wild' demo of this I've seen on the web, so I felt it was worth mentioning.

actually somebody even cross-compiled pocketsphinx to javascript with emscripten for this purpose:

https://syl22-00.github.io/pocketsphinx.js/live-demo.html

this works pretty well - all in the browser, especially if you drop in some better acoustic models.

IMHO, until they decide to provide proper HTML5 support - particularly access to the getUserMedia and general WebRTC APIs - or at least allow 3rd party vendors like Google and Mozilla to do so by themselves, this is not a believable sentiment.

It's the same Google ASR on the backend.

This was rejected by the parliament last year:

http://www.basicincome.org/news/2015/10/swiss-parliament-opp...

but is currently slated for a public referendum here in Switzerland in 2016:

https://en.wikipedia.org/wiki/Swiss_referendums,_2016

I think it is one of really only four possible futures for human society [mincome, totalitarian, oblivion, expansion into the solar system and beyond (which will me effective status quo)].

I don't think he was arguing that the (possible) non-dealership model was broken, but that Tesla's current attempt to disrupt it is flawed, and that if they fail because of poor execution, that this could stifle future innovation in this area.

The next sentence:

> My concern is if Tesla fails the experiment of directly selling cars to consumers may be considered a failure, when in reality it was tremendous missteps on Teslas part that caused it to not work.

The article and all the quoted studies do indicate that conscientiousness is probably the most important positive trait (which I guess will elicit a few irreverent 'no duh's), but it also points out there are some equally powerful and opposite traits like neuroticism low agreeableness, which lower the utility of being conscientious. The title sort of omits the latter part.

Are there any examples using this in the wild? I couldn't find any 'Sites powered by Aurelia' page on the site. Would be interesting to see some full fledged sites (besides the project site).

> Working with a Japanese company may seem like a good fit if you speak Japanese, but with the vastly contrasting work cultures and the fact you’re a foreigner in a Japanese world–you might want take a look at other options.

If you honestly think you speak and understand Japanese, but you don't understand or are not already better and more completely aware of what is being discussed in this article - then you don't actually speak or understand Japanese as well as you think you do (or as well as the many 'nihongo ha jouzu desune!' acclamations might have led you to believe.)

And the next step will be to show different variations to multiple focus groups, then train a neural network to automatically select the set of blending weights for the film that are projected to maximize revenue, thereby automating the post production as well.