https://facebookresearch.github.io/nougat/

But, I don't think your claim is in tension with the article's claim that being able to type fast lowers a barrier (or provides a benefit) that's more than simply the "time it takes to literally type".

There's more there there.

It's mentioned in the article, but I mostly think of it as: the faster one types, the shorter the iteration time. When you can type roughly as fast as you'd normally speak, it's a totally different experience than t-y-p-i-n-g each word of a sentence.

A simple test would be to see what happens if you try to, today, try to watch something on Netflix. Does it allow you to start watching shows? Or does it ask you to do something like restart a subscription? What do you see today?

Comments URL: https://news.ycombinator.com/item?id=34584781

Points: 1

# Comments: 0

But you are saying here: "If you leave git diffs in your files, whether Jupyter notebooks or otherwise, and run/compile them... They will break."

Have you changed your mind in this thread? Or what's your objection?

You could think of it as purchasing an option, which businesses definitely do.

R1 is the rating (top tier) research university. See a list here, https://www.brandeis.edu/diversity/dei-recruitment-hiring/ou...

Take an idea like Newton's Law of Gravity, and let's separate the words that Newton wrote, and the content of the ideas behind those words. Many people, over decades, have considered how to best to present and teach the content of the idea in the best way possible. The same happens with Einstein's General Relativity.

Current practicing astrophysicists, even those whose career is making measurements of General Relativity, have very likely never read neither the original words of Einstein nor Newton's words on Gravitation. This is not a scandal because the physicist's attitude is: who cares about the words used in the original description, when what matters is the content of the idea?

Further, if smart and pedagogically minded people can work and re-work the actual content into a much more understandable distillation of the actual idea, why not use that? Everyone still realizes that it's Newton's idea.

Contrast this with a philosopher working on some modern topic that built on some ideas of Nietzsche -- saying they have never read the original words would be seen as an admission of professional neglect. But even a 1st year undergraduate, what's taken as an important thing is to read the actual words of Nietzsche (perhaps in translation) and not someone's presentation of the ideas in their best possible form.

So, let's circle back to the original distinction: the idea and the words originally used to express that idea. In physics, you attempt go to the best exposition of the idea (which is rarely the original words). In philosophy, there is a primacy given to the words themselves that is bizarre from the physicist's perspective.

Finally to address what you wrote here:

> The "replacement" for Nietzsche's books, if that's a goal, must also address WHY Neitzsche's arguments are weak or insufficient.

This isn't what is being asked for. Easier perhaps to shift into the physics example.

When learning of Newtonian Gravity you learn about the concept in the best form possible (again, not the original words used by Newton but a modern presentation of the content of the idea). That's the equivalent of what's being asked for w/ Nietzsche.

You may later learn how Newton's ideas are incorrect just as you may later learn ways in which Nietzsche's arguments could be improved, but that's for later.

The missing idea in philosophy is having the best presentation of the content of the idea.

For a bet of $X, the expected value is: (1.5 * X) * 0.5 + (0.6 * X) * 0.5 => 1.05 * X. The ensemble average per round is positive (1.05) and over multiple rounds smoothly tends to infinity with the number of bets. (Definition here: https://en.wikipedia.org/wiki/Expected_value).

The time average for any specific person betting in this game is 0.95 * X (for the reasons you mention) and tends to zero with the number of bets.

So let's go through a few specifics of your comment:

> He actually picked bad numbers. That is a losing bet even on average.

The point of this article is that "on average" is trickier than people tend to assume. There are different ways of taking averages. If you do the expected value calculation and get a positive number, you might (as other comments have said explicitly) expect that a participant repeatedly engaging such a bet would have his wealth trend toward infinity. But, they are wrong (as shown in the article).

> This happens to not show after only 100 trials just because some tiny number of people get really lucky and draw up the ensemble average, but if you keep going, somewhere between 200 and 500 trials, the ensemble average pretty quickly drops below the starting average wealth and stays there, asymptotically approaching 0.

The ensemble average is positive and monotonically increases w/ the number of rounds of betting.

This is wrong. The bet as described has a positive EV and the time average for a single player tends to zero as the bet is repeated.

> There is one assymmetry at the zero point (assuming people can’t recover from bankruptcy by borrowing another dollar) ...

The result is not due to zero being an absorbing value. In the setup you can go arbitrarily small and come back without issue. The result is the same.