Technology

I just saw an ad from Amazon to pre-order a new Samsung 4K Ultra HDTV. What is a “4K Ultra HDTV”? It’s like an HD TV, but with 4 times as many pixels. HD TVs have 1920X1080 = 2,073,600 total pixels. The new 4K Ultra HDTVs have 3840X2160 = 8,294,400 total pixels. So we’re going from about 2 million to about 8million, so I guess that explains the 4. I don’t know what explains the K. (As an aside: most people who actually talk about resolution in their day-to-day lives would consider this a doubling because only the rows are counter, which is why 1920X1080 is referred to as “1080p”, but that’s neither here nor there.)

The 65″ model is going to set you back about $7,000, but–aside from the facts that I don’t have $7,000 to spend on a TV and that even if I did my wife would kill me for it–there are a couple of problems. First: the 4K Ultra HDTV is only one of two standards under the “Ultra HDTV” moniker. The other is 8K Ultra HDTV (I’ll let you figure how many pixels that one has). So if you bought the 4K, would you end up sort of like those folks who bought the 720p HDTVs before the 1080p HDTVs became standard? Even more importantly: what are you going to watch on this monstrous display? There’s not a lot of content available at the resolution. The only thing I found was TimeScapes which, while cool, doesn’t seem to justify a brand new computer.

But what I’m really curious about is when TVs will finally move to a display standard that makes sense, like the Apple Retina Display. The idea there is that it’s not the number of pixels that matter, it’s the size of the pixels relative to the distance you are from the screen. Once you can no longer visually recognize pixels at a normal viewing distance, adding more pixels gets silly. It’s just a nice, simple number that marketing guys can emphasize to sell their product, although a lot of other things (like color fidelity, brightness, and contrast ratio) matter a lot for making a screen look good.

Wikipedia has more, if you’re curious.

This is gorgeous.

It seems more and more true every day: private space exploration companies have all the best toys.

CNN has a cool article with great pictures showcasing the size and position (relative to their own stars) of some possibly life-sustaining planets.

There’s no direct evidence that the planets support life. They just happen to be somewhat close to the size of Earth (these particular ones are all larger) and also exist within the “habitable zone”, which is the distance from their own stars where temperatures would possibly support life similar to what we know (e.g. the possibility of liquid water). One of the cool things about the exoplanet research so far is that rocky planets are being found frequently–and often larger than the Earth–while gas giants are being found in much closer orbits than scientists had expected.

These are exciting times!

Using a special camera that captures images in 1-40,000th of a second (vs. about 1-200th for normal cameras), scientists have captured images of snowflakes as they fall, revealing fascinating images quite unlike what we’re used to seeing.

My favorite thing about these images is that you can see the three-dimensional structure of some of the flakes. I caught the story from a short article on ScienceMag.org, but the University of Utah also has a gallery.

So I guess everyone’s making lists on the Internets now. I’m not sure when it became a thing, but it is certainly a thing. And Business Insider (?!) wants in on the action, so here they are with The 12 Most Controversial Facts in Mathematics. Leaving aside the silliness of the name, some of these are actually really good mind-benders. In fact, they led off with my favorite: The Monty Hall Problem. What’s best, however, is that this was the first example I’ve ever read that actually made some intuitive sense to me!

Believe it or not, that image will make sense after you check out the article!

The rest of the 12 are less exciting, in my mind, but the illustrations are pretty cool and the explanations are good so it’s worth a read. (Estimating pi by throwing darts at a dartboard was also pretty cool…)

So this is a master playing the retro-game centipede:

It’s obvious watching the game that this guy is good, but it’s also weird to think of a “good” player of this game since, sort of like tick-tack-toe, the optimal strategy is easy to compute. If you just crawl your centipede across the screen from top-to-bottom (or left-to-right) getting every single row (and never deviating to snag one of the prizes) you will eventually fill the entire screen and “win”. But that would be rather boring, and that’s not what this player does. Instead, he or she often takes risks to get the prizes faster and only uses the sort of careful-nesting technique when absolutely necessary.

That indicates the player is playing a much more complex game: not just how to get the maximal length (which is easy), but how to do it quickly, which is trickier. I did a quick Google search and I couldn’t find any contests to design AI to do it. The game is still simple enough that I think it might be solveable, but if speed matters as well as length, then it’s much more interesting to think about.

This great website provides a really cool illustration of the distance between Earth and Mars, to scale.

You get to check out the distance of GPS satellite orbits and the moon as well. One thing that the illustration doesn’t mention, however, is that the shortest point between the Earth and Mars isn’t a realistic metric. Because the travel time takes so long (months), you can’t aim for Mars where it is at the time of launch. You have to aim for where it will be when you get there, which of course depends on the route you take. I can’t quite wrap my head around what a “shortest route” would look like, and that’s why I think I need to study some orbital mechanics before I write much more science fiction.

Anyway, enough orbital mechanics talk. Go watch the cool visualization!

This idea is simple to explain, but the implications are much more complex and murky. Take a young, 20-something college graduate and use the kind of information they would put on a résumé (college degree, GPA, etc.) to predict their future earnings over the next 10 years. Then, offer them a lump-sum payment now in exchange for a percent of those earnings.

It sounds a bit like science fiction (very good science fiction, in fact), but it’s actuallya real-live business named Upstart (that’s them in the photo) that is already earning candidates an average of $50,000 now in return for a slice of their future successes. On the one hand, it’s a great solution to a frustrating problem. Smart young grads have a ton of potential, but they are also liquidity-constrained. Borrowing from their future to kickstart their careers makes a lot of sense. It makes sense to investors, too, with a targeted rate of return in the range of 8-9%.

But it also looks a little bit like indentured servitude. Of course it isn’t, not today anyway. For one thing, there are safeguards: no more than 7% of income can be taxed, and there’s a cap on how much you have to pay out if you strike it rich and sell a business for $100m. For another, the entire dynamic is inverted: instead of down-and-out, vulnerable people the service is targeted towards those with the  best degrees and the most options.

Here’s a video from Upstart, the name of the company, explaining what they do.

This is one of those articles that I think catches a real glimpse of ways the future might be totally different from what we’re used to, but not based on what you normally think of when you think of futuristic inventions.

The Roku 3 has a lot of improvements over previous generations, like the headphone jack integrated into the WiFi-based remote, but for me there is one improvement that stands out above all the rest: the Roku will let you search for a show across all your “channels” (Netflix, Hulu, etc.) That’s a subtle but incredibly important feature that really changes the nature of the device. Check out the review at The Verge for more.