Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ogogmad
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
69 ms
·
901.
▲
by
ogogmad
4y ago
I'm not going to accept your bet because I don't understand its conditions (and what's the point?), but I accept your challenge. Also, I'm going to f*** around bigly with floating point. def f(x,a): return 1 + a
902.
▲
by
ogogmad
4y ago
I don't fully understand your bet (and I don't take kindly to being asked to silence myself). I don't have time at the moment to craft an example. I might play around a bit with the edges of floating point. Differentiation is
903.
▲
by
ogogmad
4y ago
Won't work on my example. Given any differentiable function f:R -> R, you can add a very small function to it that dramatically changes its derivative. But maybe this is too pathological to bother about. Over the complex numbers, th
904.
▲
by
ogogmad
4y ago
No. The dual numbers are very similar to the complex numbers and produce exact results. Ridiculous is true and fair. [edit] Also, I wouldn't count using Cauchy's Integral Formula as the complex-step method.
905.
▲
by
ogogmad
4y ago
Such methods are doomed. Consider the function f(x) = a sin(x / a) for very small a. Finite differencing methods would approximate f'(0) as 0 as f(x) is very close to 0, but the actual value of f'(0) is 1. Differentiat
906.
▲
by
ogogmad
4y ago
The central derivative is equal to Im(f(x + j h) / h) where j is the split-complex number which squares to 1, "Im" is the imaginary-part function, and h is a small real number. When j is replaced with the dual number epsilon,
907.
▲
by
ogogmad
4y ago
Maybe this is the wrong forum for this, but the Talmud says that A Sanhedrin that executes someone more than once every 70 years is bloody . If you think about it then, you could simply limit the frequency of executions, and scale it up or
908.
▲
by
ogogmad
4y ago
> Given this, it seems like MacBooks could easily end up becoming the laptop of choice for Linux developers in the years to come. What's the situation with traditional USB and HDMI ports? Dongles are easy to forget and easy to break
909.
▲
by
ogogmad
4y ago
> 1: it's wrong even for the guilty What about an example I gave above: The commandant of Auschwitz? On a purely emotional level, I find the point of view that a pitiless mass-murderer deserves infinitely more mercy than what he gra
910.
▲
by
ogogmad
4y ago
I'm saying that many would want to execute the commandant of Auschwitz, especially if the evidence for their guilt is overwhelming. I think a standard of proof high enough can sometimes be met, and the crime sufficiently deserving. On
911.
▲
by
ogogmad
4y ago
It depends on how rigorous the standard of proof is. The US spends a big amount of money proving that someone sentenced to execution is guilty of their crime.
912.
▲
by
ogogmad
4y ago
Could this be used as an argument against capital punishment? If it takes 20 years to exonerate someone, then while you can still compensate them financially if they're still alive, you can't make up for killing them.
913.
▲
Categorical Imperative
(en.wikipedia.org)
2 points
by
ogogmad
4y ago
|
0 comments
914.
▲
by
ogogmad
4y ago
What about both? Treatment and penance. I don't want to live in a world where people won't take responsibility for their actions, so penance is needed (and therefore prison). But equally, I want to see people helped and reintegrat
915.
▲
by
ogogmad
4y ago
Would it be enough to treat him for paranoid psychosis? What should be done with such people?
916.
▲
by
ogogmad
4y ago
> if there was ever an argument for privacy this should be it Sorry, what's the argument? The '00-'10 libertarians on Slashdot never gave arguments.
917.
▲
by
ogogmad
4y ago
Intuitionistic and linear logics are good examples of "weird" logics that have had some relevance to computing. Intuitionistic: Curry-Howard, Dependent Type Theory, formal methods. Linear logic: Rust.
918.
▲
Continental Philosophy from Hegel [pdf]
(scholar.harvard.edu)
2 points
by
ogogmad
4y ago
|
0 comments
919.
▲
by
ogogmad
4y ago
(I like mathematical logic). Continental philosophy eschews formal logic and focuses more on literature and the arts as the source of philosophical truth. Some insights from Emmanuel Kant are used to justify this ([1]). Can someone explain
920.
▲
by
ogogmad
4y ago
Question: Do fractals have projective duals? In the projective plane RP^2, the dual of a point is a line, and vice versa. Do fractals like "Burning Ship" and "Mandelbrot" have projective duals, where each point in the se
921.
▲
by
ogogmad
4y ago
A follow-up question, not easily answered using Xaos: The dual numbers can be pictured as lines in the Euclidean plane instead of points (details here: https://en.wikipedia.org/wiki/Laguerre_transformations ). What woul
922.
▲
by
ogogmad
4y ago
Here's the split-complex Mandelbrot (and Mandelbar for comparison): https://imgur.com/a/j2c7qkB Read my other reply.
923.
▲
by
ogogmad
4y ago
A holomorphic function over the split-complex numbers is the direct sum of two holomorphic functions over R. In that sense, it's trivial. You will be very disappointed once you see the split-complex Mandelbrot set. It will be an instru
924.
▲
by
ogogmad
4y ago
Here it is. Split-complex burning ship using Xaos: https://imgur.com/a/tBTuUDV This is my first time using that program. It doesn't look too bad. Same thing, but with the dual numbers: https://imgur.com
925.
▲
by
ogogmad
4y ago
If you replace the complex numbers with split-complex numbers, then the set of holomorphic functions becomes uninteresting. You therefore need to use * (meaning the split-complex analogue of complex conjugation) to get anything interesting.
926.
▲
by
ogogmad
4y ago
I've experimented with the "Mandelbar" (sometimes called the "tricorn") fractal over the split-complex numbers: https://imgur.com/PkzY7B6 The result isn't completely trivial, but isn't par
927.
▲
by
ogogmad
4y ago
Organisations sometimes make mistakes when responding to something novel. They don't necessarily screw up again, unless there's something deeply wrong with what they're doing. Is there something deeply wrong with the WHO?
928.
▲
by
ogogmad
4y ago
What's the difference between a standard and distributionally robust language model? The paper didn't say anywhere. [edit] After some quick reading: A language model is a model which (roughly speaking) tries to predict the next ch
929.
▲
by
ogogmad
4y ago
I consider it a good thing in this case. Germany could try rationing its natural gas to ensure that everyone gets at least a minimum amount. Could that work?
930.
▲
by
ogogmad
4y ago
I'm Jewish and I'm tired of this interminable grievance-mongering. He should've mentioned the Romani and Slavs too in this short essay, right? The UK fought Hitler.
More ›