Hacker News new | past | comments | ask | show | jobs | submit | missingET's comments login

No, there’s an abstract algebra extension of real numbers to have an extra symbol h such that h^2=0. This is not a real number so you cannot apply the argument h^2=0 implies h=0, much like complex numbers don’t obey all properties of real numbers.

(For example for real numbers, x!=0 implies x^2>0 but i^2=-1)

https://en.m.wikipedia.org/wiki/Grassmann_number


What’s interesting is that there’s a net charm content: there are more charm quarks to be found than anti-charms.

Given that charm quarks are heavier than the proton, you’d expect to only find them in deep inelastic collisions when they are produced in pairs with an anti-quark, so it is surprising that there’s an asymmetry.


Wait, what? I read the article as saying that it's an open question whether there's a charm asymmetry.


It largely is. Many use AI and machine learning interchangeably these days. I’ve even seen a random forest model called “AI”…


Jax has a much nicer handling of higher order differentiation. PyTorch has functions to compute Hessians and there are libraries to keep differentiability through optimizers, but going out of their standard use-cases becomes tricky very fast. In contrast, JAX can compute nth-derivatives of things very easily.


Although I never find myself agreeing fully with them, ideological absolutists who refuse compromise are actually useful as some kind of anchors to the Overton window. They are not the ones ending up negotiating the actual realistic solutions but their activism helps shifting the equilibrium.

If your side of the argument is all compromise-makers trying to meet people halfway you often end up giving up more than you’re comfortable with. Both types of people play a useful part.


This is usually known as a radical flank, particularly useful in discussions about climate activism these days.

The FSF/GNU seems a bit too fringe and not too great at explaining the problem to everyday people instead just offering entirely uncompromising solutions that don't work for most people. More isolationist than activist. What would the less radical organization benefiting from the existence of FSF/GNU even be? The EFF?


Not when the power balance is this big. Against big tech who has the resources to wear any conciliating opposition down, you think you're just compromising like a "mature adult", until the next time, where you're forced to compromise on your compromise and so on.

Give them an inch and they'll take (or more precisely, they took) a mile.


You can’t really negotiate with corporations on what they do.

On this kind of topic negotiations are at the political level and whether you like it or not they will take the unsavory position of large corporations into account. A total refusal to compromise can end up leaving your side out of the discussion and end up with worse regulation. Uncompromising idealists just don’t have the leverage to make impactful threats of leaving the negotiation table.

That doesn’t mean you should not draw lines and do your best to hold them. But if the line is already crossed you can still do good negotiating, as unpleasant as it is.

If there are no “soft mature adults” there to be heard, no one is heard. But your point that giving an inch ends up losing you a mile absolutely stands: this is what I was saying about absolutists being anchors, they allow the soft negotiators to start from a stricter position before the negotiation starts. A group made up entirely of “reasonable compromising people”, however, is terrible because they start from a weakened position that they view as a reasonable compromise.

However, there will be negotiations and you will end up giving away some inches unless you have a lot of power. I absolutely value your un-negotiable position, but don’t underestimate what moderates can do representing it in a politically-acceptable, watered-down way.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: