neuralk

Wittgenstein argues that there is no such thing as a private language.

Besides the course you are taking (is it an in-person class, or online perhaps?), are there any resources for learning Sanskrit that you'd recommend? I have the same long term goal as you! I'm looking to self-study the language since there aren't local courses where I live.

Beautifully presented! Thanks for sharing, these notes are inspiring, I especially like the simplicity of the presentation's design paired with interactive elements. Too often I find material with the latter to be extremely busy. As someone interested in instruction, I'm taking notes from this!

This feels like a business opportunity. Or, at least, an open source repo + a tutorial article that would attract donations.

I never (no longer, that is) see tweets that other people like. I think there is a setting to disable them from appearing now. At the very least, whenever one appears you can click the little arrow next to it and select "see less often" (or equivalent).

>Do you want an unsorted list of all websites that contain your keywords?

Yes, because then I could apply my own ranking. More realistically, it would be useful as various web services could select from among various forms of ranking as desired by their users. Right now, the internet as a collective is at the whim of the particular ranking Google elects to use. If they don't like you, too bad, you lose.

What's more, I would even settle for a sorted list of all websites that contain my keywords, but Google provides no such thing. Google's algorithm has changed significantly in just the past few years, making it very difficult to do research of unusual topics, no matter how broad or specific you make your search. For instance, just today I was looking up information related to an event in the life of Andre Gide. After you get through a few pages (maybe 150 or so useful links from the grab bag of what appears), Google makes it such that you can explore no further. Entering a new term with a slight variation can yield a set of another ~150 results with no overlap with the previous search's results, making any kind of exhaustive research quite frustrating.

No, it is not just interpolating. The underlying algorithm uses machine learning by applying a trained deep neural network. So there is value added besides a mere upscale.

You're ultimately right, though, and that a true HD is only going to come from the raw film content. What the neural network gives us are essentially plausible higher-res hallucinations.

Edit: as per the other comment, if the original exists only on video and not film, perhaps this is the best we're going to get.

The article is paywalled, unless you have access through a university or other library.

The high price you'd pay for paywalled papers goes in large part to the publisher for no real tangible benefit. Most research, and likely the linked article, are funded either directly or indirectly through public funds like taxes. So it is especially obnoxious to see an interesting article like the one linked be locked down.

As for why it's bad etiquette on the front page of HN, the free-and-open ethos common among hackers (at least, old school ones...) would usually see someone link directly to a pdf, instead of this.