honoredb

Typo near the beginning, "press a petal," gave me a fascinating mental image for a second.

I'm sure it'll be informative to learn exactly why every theoretical avenue for faster-than-light travel doesn't work, but I feel like we can infer from the size of the universe that these effects don't exist; if they did, they'd dominate our observations of the universe and we'd never have thought the speed of light was a fundamental limit in the first place.

This is a nice explanation and a cool solution. I amused myself a bit ago by trying to come up with a solution for these two constraints without reading about the crypto work, and ended up with something different. In my scheme, a voter casts a vote for a candidate and gets a receipt with the UUID of the vote, which is simply mapped to the candidate so that they can verify it online later. However, the voter can also cast any number of additional ballots, which are constrained by the system to be pairs of votes and anti-votes, each for the same candidate, and get receipts for each. For example, I support Harker, but I've been paid to vote for Dracula. I go into the voting booth and cast my vote for Harker, getting Receipt 1. Then I press an extra button to cast a fake vote for Dracula, and get Receipts 2 and 3, with 3 being a special negative ballot. I can show Receipt 2 to my briber, who can verify that it corresponds to a vote for Dracula. But secretly, I can use Receipt 1 to check that my vote for Harker was counted correctly, and Receipt 3 to see that a negative vote was also cast for Dracula, cancelling out my bribed vote.

You need to allow each voter to cast multiple fake votes, otherwise the briber/coercer could simply demand receipts for a fake vote in addition to the real ones. Could get a bit unwieldy. But the big advantage is that there's no extra complexity for the average voter, since they don't need to cast any fake votes.

Naturally, you can generate awful continuations in the same style via GPT-2.

Model prompt >>> As they sprinted together down the echoing, looping ramp of the deserted Guggenheim Museum, closely pursued by three swarthy members of the resolutely vicious Cannelloni gang, square-jawed British Royal Marine art historian/world's deadliest sniper John Savage and his voluptuous young modern art critic/Navajo linguist Samantha Silver cursed architect/interior designer/writer/educator Frank Lloyd Wright for designing such a circuitous route out of the building.

>> ======================================== SAMPLE 1 ======================================== The fact that the entire way was over a five-pointed star of the "Stars of David," which both men had tattooed on their arms, was beside the point. Neither of them would have been deterred if their pursuers had been armed with nuclear weapons, which would undoubtedly obliterate the entire museum, as well as any of the many high-rise buildings surrounding it.

At the heart of the building is an enormous, ornate winged monolith bearing the words "

======================================== SAMPLE 2 ========================================

"Just the way he built this place," Savage snarled, his hands clenched, "he'd be able to make any building disappear any time he wanted."

A couple of hours later, back at work, Silver and I were finally able to catch up with fellow explorer/writer/reporter Charles Fenton, who had been in the Museum galleries earlier in the day and stayed behind to observe our conversation. "That was quite a day!" he beamed at me.

======================================== SAMPLE 3 ========================================

As they approached the main lobby, Savage turned and grinned, showing off his impressive, biceps-bound physique. There was a reason the gang of thugs had been lured here, he explained to Silver. There were two other entrances out to street level. "The most efficient way of crossing this place is by helicopter." He paused, eyes scanning the distance. "And that would require an army or navy to pull it off." He raised his gun and laughed. A few yards away, Savage and

This. Technical verification solutions help, but none of them defend very well against your CEO apparently ordering you to bypass the verification step.

The story of a bug that caused the AI to optimize for maximally disturbing text that went unchecked because the only people authorized to stop it were asleep is a great illustration of how easy it is for an AI to "go evil" when you're not worrying about safety.

Oh, I get it, it's a defense against plagiarism detection. The article is largely plagiarized from other sources, but since they didn't contain the homographs you can't just google the unique sentence "This is a fair and legitimate question, but not the first question you want to ask." (Original source, possibly: https://www.risesmart.com/blog/6-key-questions-ask-during-sa...)

They happen to look significantly different in the font in the article the way my browser is rendering it. They're using the Cyrillic character that looks like q, https://en.wikipedia.org/wiki/Qa_(Cyrillic).

I'm super distracted by the use of ԛ instead of q in this article, what's that about? Plagiarism detection? Seems bad for accessibility/SEO.

...who invented PIE https://en.wikipedia.org/wiki/Proto-Indo-European_language

I love this title and abstract; we need more of this kind of plain speaking and rigor when dealing with mysticism-adjacent areas.

So now we have to consider the specter of someone writing an overly verbose review of something that reduces to "tl;dr". Someone might summarize the review by writing "tl;dr: tl;dr". Or they might express the same thing, but overly verbosely, so someone might summarize their response by writing "tl;dr: tl;dr: tl;dr". In general, if N "tl;dr"s is a valid sentiment, and any valid sentiment can be expressed overly verbosely, then N+1 "tl;dr"s are a valid sentiment, so by induction any positive integer repetition of "tl;dr"s is valid. However, there exists some number L such that L repetitions of "tl;dr" is itself too long to express the necessary point, and someone who didn't read it would want a summary, the most succinct of which would presumably be "tl;dr: tl;dr nested L times". This meta "tl;dr" itself might be expressed overly verbosely, of course...

Nitpick: "…had 13 not been prime" is not meaningless whimsy--a prime number of images is harder to lay out on the page. Mentioning this in the caption without explaining it is whimsical, but like a lot of GEB I'd argue it's whimsy with a purpose: training the reader to look for applications of abstract concepts. The reader ideally goes "wait, what? There's no connection between those two things! Unless..."

It says something that I'd disagree about who won in the contest in the epigraph. Fooling a flock of birds sounds much harder than fooling somebody of the same species, not because birds are better at perception but because they're too alien for the same tricks to work on them. A painting of grapes that would attract birds probably wouldn't look particularly representational to humans; it'd be a kind of art that required conceptual flexibility and a deep understanding of both birds and grapes. Something, indeed, that you'd expect to see only in a modern art museum.

To store 500 million hashes in 860MB, wouldn't you need to truncate them to 13 bits each? That'd give a false positive rate of 1000 in 1000, slightly worse than the Bloom filter's false positive rate of 1 in 1000.

But that's still 12G of memory, since you can't compress it. Is there a data structure that's better optimized for space and probabilistic membership tests?

It's really hard to avoid logging passwords when there's a parse error on structured data that includes a password, for example if there's malformed JSON in the authentication HTTP request. A weird error like that is exactly when verbose logs are essential, but it's also a time when your system doesn't know what it's logging. Amazon has a machine-learning based service to try to deal with this.

No, there fundamentally isn't, because you can't trust the client to actually be hashing a password. If all the server sees is a hash, the hash effectively is the password. If it's stolen, a hacker can alter their client to send the stolen hash to the server.

Grubhub/Seamless | NYC | Engineering | Java | Webservices

I'm looking for experienced engineers to work on making food delivery more secure, reliable, transparent, and maybe a little socially conscious? You can contact me at azinger+whoishiring@grubhub.com or browse jobs at https://social.icims.com/viewjob/pe15251916278411627e

Cool! I feel like you could go further and let the student control the transfer to text. Give all the built-in functions both icons and names, and let student-defined functions be icons they create if they want. The point where it's less cognitive load to use text is going to be different for different students, so let text-based coding be a tool they discover, not a requirement imposed on them.

Medidata | New York City (and elsewhere) | Full Time | Onsite

About half of all clinical trial data goes through our various products at some point. We're working to make medicine get to market faster, and be cheaper and safer when it gets there. My team writes mostly Ruby, frequently open source gems or platform web services.

Email me at the output of this befunge code: http://pastebin.com/hL6Hbv8z

...the new learnings

Picked up a little Indian English along the way, I see. :)

I liked the attention to UX. Deliberately breaking the user's normal workflow in order to emphasize that something vital is different is a pattern I'll need to remember.

Medidata, NYC, London and elsewhere: http://jobvite.com/m?3fEZvhw9 (all listings)

http://jobvite.com/m?3BDZvhwu (NYC Software Engineer)

Medidata is an established company working on optimizing every aspect of clinical trials, getting medicine to people faster, cheaper, and more reliably. We're building a heterogenous platform (okay, there's a lot of Rails, but other stuff too) based on API discovery, hypermedia descriptors, and other ways of making REST-based systems more loosely coupled. If you're willing to spend an hour talking about whether a given method should be a PUT or a POST, come give us a hand.

moc.losdm@reniewa

Medidata, New York and elsewhere http://rolp.co/RaB67

We're helping fix the clinical trial process, getting treatments to market faster and cheaper, and we're always hiring for a variety of positions.

East Coast, West Coast, UK, Japan: Medidata Solutions is always hiring for a variety of technical and non-technical positions. We're an established company with something like 50% market share in the clinical trial industry, helping to make drug trials faster, cheaper, and more reliable. All listings: http://jobvite.com/m?3EOtAgwg Or come write hypermedia microservices in Ruby with me in New York City: http://jobvite.com/m?3WRtAgwB

I feel compelled to defend the honor of the New York Times here: that misleading graph attributed to it is from the New York Post, a completely different (and nowadays much less reputable) paper.

Not particularly unethical as described, but I'd imagine this is technically a federal crime.

Fun! Level 6 failed to load any widgets, evil or otherwise, in Chrome; I had to switch to Firefox and redo the whole test. For my external script I used http://pastebin.com/raw.php?i=15S5qZs0, although I don't think the lack of a .js extension there was the problem.

As does Citibank. I imagine it's for telling-support-over-the-phone purposes, which isn't great.

I agree here, but I wonder if it'd be kosher to pair program on an open-source project...that my team totally intends to incorporate into our closed-source product?

Medidata (New York, NY and elsewhere) - Software Engineer, Software Engineer in Test

As the leading global provider of cloud-based clinical development solutions, we deliver an innovative platform designed to make clinical trials more efficient. We understand the significant challenges faced by life science organizations. And as the pressure increases to safely get new treatments to market faster and at a lower cost, so does our need for highly skilled and committed professionals.

If you love creating innovative ideas to address business challenges, you will thrive at Medidata. Join a community of highly motivated and talented professionals in an environment where you can grow, learn and have an impact.

The heart of our business is building technology solutions that help improve quality of life. We invite you to make a difference.

http://hire.jobvite.com/Jobvite/Jobs.aspx?b=nwQ1BlwO (all job listings)

http://jobvite.com/m?3gHU7gwJ (Deep link to Senior Software Engineer Level II, New York)