kamkha

I get your frustration here, but keep in mind: your use of that word is not harming Stripe any more than alternatives you could use, but it does harm an unrelated and oppressed group.

It's not at all material to the point you're making, but to nitpick: Hans's first move was actually 1. d4 and thus the Wikipedia article you should be linking to is that of the Indian Defence [0], not Alekhine's Defence!

[0] https://en.wikipedia.org/wiki/Indian_Defence

In their example, you're not giving S as an input to K — instead, you're using K to write another program P which iterates over all possible strings and returns the first whose complexity is some value ("2 million" in their example).

That program P will certainly be longer than K (as it contains K), but not much longer — it's adding to K only the instructions needed to iterate over strings and define the threshold complexity ("2 million"). P will then produce that "2 million"-complexity output, but it didn't need any input and thus the complexity of its output is truly just the length of P (which is smaller than "2 million"). It eventually stumbles upon S by going through all possible strings, and didn't need S to be provided.

The main idea of the proof is very similar to the interesting-number paradox [0] or the Berry paradox [1].

[0] https://en.wikipedia.org/wiki/Interesting_number_paradox

[1] https://en.wikipedia.org/wiki/Berry_paradox

Yes; further reading: https://en.wikipedia.org/wiki/Solving_chess

Slang (https://slangapp.com) | Boston, MA & Bogotá, Colombia | Onsite or Remote

→ Software Engineers, Product Managers (PMs), Product Designers

Slang started as an MIT research project on using AI and NLP to make learning a new language as efficient as possible, and has grown into an adaptive English-learning platform that offers an unprecedented variety of specialized English courses and proficiency tests in 50+ industries like Law, Oil & Gas, and Aviation. With just a small team, we’ve already created the largest professional-English offering in the world, and now we’re scaling up to release 1,000 more courses over the next three years.

Learners use our learning apps to submit millions of language activities per week, managers use our LMS to visualize the progress of their students or employees, and our Content team uses our internal tools to provide expert curation and refine the output of our models.

We just closed a round and are scaling up our 6-person Product team to grow those applications. We're looking for engineers, product managers, and product designers. Our backend combines a Ruby/Rails API with a data-intensive Java stack, and our frontends include a React/TypeScript/MobX web UI and Cordova-based hybrid mobile apps.

I'm Kamran, the CTO — email me directly at kamran+hn@slangapp.com!

Their free tier has a 100GB/month bandwidth cap — surely the most trafficked websites in the world would very quickly hit that limit, no?

Your point stands nonetheless — they could certainly be losing money on certain large users of their free tier. Serving relatively static content has gotten extremely cheap across the board, though, and like the parent said, they might be benefiting from the pricing by more indirect means.

They're streaming a single live game right now! They explained that the research build of the game being used doesn't have an "observer" mode so it's not quite suitable for streaming live games — and, in fact, the games that they were showing were played over a month ago. (The current live game is being shown from the human player's point of view, and as such is a bit… dizzying.)

Close! You swapped an operation there: it'd be (2^8 button possibilities) ^ (5 minutes * 60 FPS), or about 10^43348.

A bit of a bigger space to search ;)

Its GVWR is 5 ton.

Check out this chess programming wiki article [1] for a (still fairly hand-wavy) rationale for similar piece values, and the Wikipedia article on chess piece relative values [2] for a more in-depth look at various weightings. Some modern approaches use analysis of a huge corpus of master games to come up with piece values, but many of the systems are based on intuition and empirical evidence.

[1] https://chessprogramming.wikispaces.com/Simplified+evaluatio... [2] https://en.wikipedia.org/wiki/Chess_piece_relative_value

Unfortunately, that's a bit of a false etymology! According to humorism, black bile [1] comes from the gallbladder, not the spleen. The spleen was supposed to produce yellow bile, an excess of which was thought to cause aggression and anger.

[1] Black bile is actually the source of the term "melancholy" — from the Greek melankholia, from melan- ("black") and kholē ("bile")!

Comma's main advantage over Tesla's Autopilot was, as I understand it, the use of cameras to collect video data. Hotz presented that as the advantage at TechCrunch Disrupt, at least, saying that he could beat Tesla's Autopilot because he had full video data from a small smartphone app that he had test users running as they drove while Tesla's vehicles had no video input.

I wonder if his choice to cancel Comma is related to Tesla's recent announcement that all new Tesla vehicles would ship with a full video camera outfit.

MobX is written in TypeScript: https://github.com/mobxjs/mobx

There is no viewfinder. It is mirrorless.

It's definitely not proven to be correct; we definitely don't know for sure that integer factorization is in NP-hard. If I remember correctly, it's generally suspected that factorization is not NP-complete (and thus outside of NP-hard).

Wikipedia has it first on their "list of problems that might be NP-intermediate", i.e. problems that are in NP but not in P or NP-hard. [0]

[0] https://en.wikipedia.org/wiki/NP-intermediate

What if a 3.5mm cable was a way for your phone to connect to various IoT devices to pass the basic setup data like your wifi SSID and password instead of this goofy process of broadcasting it's own wifi network that you have to connect to.

For what it's worth, the latest Canary (https://canary.is/) is configured via a 3.5mm cable connected to your phone, and the configuration process is relatively smooth compared to the goofy ad-hoc WiFi network approach.

spaCy's 96.8% accuracy is for the task of POS tagging while Google's reported 94% accuracy is for dependency parsing, a significantly harder problem.

It is, in the same way that MtGox's BTC was independent of Bitcoin towards the end. They keep their own ledger of balances.

You can deposit directly into a cold wallet; you can't withdraw directly from a cold wallet.

The situation is even more amazing than you think. That Las Vegas hotel you mention was designed by the same firm [1][2].

[1] http://www.rvapc.com/works/707-20-fenchurch-street [2] http://www.rvapc.com/works/index/691-vdara-hotel--spa-at-cit...

"Houses shouldn't have locks; you've got glass windows, don't you?! It's a false sense of security!"

Chrome's threat model [1] paints a different analogy. In their view—which I do think is reasonable—the OS is the house and Chrome is simply a room in the house. You don't put locks on all of the interior doors of your house to keep out burglars who might've bypassed your front door, do you?

Their threat model makes sense for many users, and it's flawed for many others. One valid criticism that I do support of Chrome's model of handling passwords is the obscurity of it all. Users as a whole should certainly have a much better idea of how their passwords are being handled.

[1] http://www.chromium.org/Home/chromium-security/security-faq#...

There's a fairly well-established notation similar to what you made up: ABC notation (http://en.wikipedia.org/wiki/ABC_notation).

I would guess that the service does something more along the lines of `pngcrush -brute` which is significantly more taxing.

From http://www.chromium.org/Home/chromium-security/security-faq:

Why aren't physically-local attacks in Chrome's threat model?

People sometimes report that they can compromise Chrome by installing a malicious DLL on a computer in a place where Chrome will find it and load it. (See https://code.google.com/p/chromium/issues/detail?id=130284 for one example.) People also sometimes report password disclosure using the Inspect Element feature (see e.g. https://code.google.com/p/chromium/issues/detail?id=126398).

We consider these attacks outside Chrome's threat model, because there is no way for Chrome (or any application) to defend against a malicious user who has managed to log into your computer as you, or who can run software with the privileges of your operating system user account. Such an attacker can modify executables and DLLs, change environment variables like PATH, change configuration files, read any data your user account owns, email it to themselves, and so on. Such an attacker has total control over your computer, and nothing Chrome can do would provide a serious guarantee of defense. This problem is not special to Chrome ­— all applications must trust the physically-local user.

That's explained in one of the threads linked: "I spawned two Yoshi's by hitting the block with Mario and the p-switch at the same time. I jumped on one Yoshi, got the p-switch in his mouth and let him die, so the second - invisible - Yoshi becomes visible and have a null sprite in his mouth."

My Bank of America account has a six digit PIN, and it has worked at every ATM (BofA or not) I've encountered.

I found myself asking the same question, and then found http://basecamp.com/breeze/compare. It seems that they're offering easier set-up in return for the $10.

Now I'm curious—what does "Private Browsing" do when enabled in Settings > Safari?

It is the exact same size as buttons on the iPhone if you are rendering the iPad 2 design on the iPhone as well, but in that case the button on the iPhone would be less than 40 points wide and also too small by Apple's guidelines. I am saying that it would be against Apple's guidelines on both the iPhone and the iPad 2.

That's why I mentioned an iPhone-specific design, as well. The iPhone's pixel density isn't a problem because it can be distinguished from the iPad 2. And, as you've shown, if the iPhone-specific design is displayed on the iPad Mini then there is no problem at all. The issue is that there is no way to distinguish between the iPad 2 and the iPad Mini.

Re-read my comment. If someone has an iPad 2 specific design which uses buttons which are 40 points wide on an iPad 2, those buttons will be less than 40 points wide when that design is inevitably served to an iPad Mini user.

If iPad Minis are served iPhone-specific designs, then you're right—it doesn't matter. And perhaps the solution is "you shouldn't be making iPad 2/iPhone-specific designs," but that's not what I'm getting at.