kqr

12 ms·

331.
War of Dots(warofdots.net)
2 points by kqr 8mo ago|1 comments
332.
by kqr 8mo ago
We don't. We know with great certainty it's either random or needs non-local variables, which would let information travel faster than c, among other things. (This is a consistent result of the Bell test.) Most prefer to belie
333.
by kqr 8mo ago
> then you can't screw up the p-value test Bakker and Wicherts (2011) would like to disagree! Apparently 15 % screw up the calculation of the p-value.
334.
by kqr 8mo ago
The observation that it is impossible to ennumerate 300 bits of assignments with a 256-bit random generator is a very clever observation. Just like almost all assignments are balanced (300 vs. 295 bits), I wonder if there's a similar e
335.
by kqr 8mo ago
That's why I qualified with "seems" "to a foreigner"! I know tail-star (comet), ice-cupboard (fridge), fire-stick (match), etc. are real words, but since I read them as their literal translations into my language, t
336.
by kqr 8mo ago
The reason for this choice is explained in the article. It is a highly apt analogy.
337.
by kqr 8mo ago
Today I learned what a kenning is. I have encountered them many times before (to a foreigner, Finnish seems chock full of them) but I never knew they had a name. I love them!
338.
by kqr 8mo ago
The New Economics! Then Out of the Crisis. Then if you're really hardcore, Some Theory of Sampling.
339.
by kqr 8mo ago
...assuming people want to know that the change is in the expected range. That's often not the case. People's careers are built on phantom improvements and being able to say that regular process issues were one-time occurrences.
340.
by kqr 8mo ago
I disagree. Where I have worked, these important quantities have appeared thin-tailed: - Size of pull requests (due to feedback loops) - Effort required for bug fixes (the variation is large, but not a power law) - Developer-hours in a spri
341.
by kqr 8mo ago
None of those data points are outliers, since they are within the band of what's expected from the process. Yes, the variability of the thermostat is awful, and the SPC practitioner would care about that. But the key thing is that deal
342.
by kqr 8mo ago
You don't need expertise in statistics to draw control charts. You might need that expertise to teach people to draw control charts, but not to draw them. Line workers are the reflexes of the organisation. They can react to trouble b
343.
by kqr 8mo ago
One of my favourite techniques from that book is to remove the centralised backlog. People's ideas for improvement shouldn't be everyone's administrative burden. There are too many ideas for that. Instead, keep a central re
344.
by kqr 8mo ago
Sort of. The word to search the web for is Auftragstaktik. Here's the Wikipedia page on it: https://en.wikipedia.org/wiki/Mission-type_tactics
345.
by kqr 8mo ago
Read his books.
346.
by kqr 8mo ago
Reinertsen has borrowed more from queueing theory than from Deming. This is not unexpected -- Deming worked mainly with thin-tailed statistics, whereas Reinertsen applied his knowledge to the power laws that show up more in design and devel
347.
by kqr 8mo ago
To be clear, this is not a diss of the article. It is hard to create fake data that look realistic at first glance but are not under statistical process control. That's just how good control charts are at separating signal from noise.
348.
by kqr 8mo ago
I think you're confusing Deming with statistical process control. It is true that SPC works best for the non-chaotic parts of product development and manufacturing alike. There are parts of product development that are non-chaotic, and
349.
by kqr 8mo ago
Deming was a statistician first, yes, but he also had strong opinions in terms of management science/philosophy. These opinions came from a perspective of systems theory and understanding variation.
350.
by kqr 8mo ago
> for thermostat B there are many more outliers. We’d say that [...] thermostat B is not [under statistical process control]. (In practice, you’d draw a control chart to identify whether the system is under statistical control). I did dr
351.
FutureEval: Measuring the Forecasting Accuracy of AI(metaculus.com)
2 points by kqr 8mo ago|0 comments
352.
by kqr 8mo ago
How much of this reply is environmentalism baked into it with post-training? I don't have access to a good non-RLHF model that is not trained on output from an existing RLHF-improved model, but this seems like one of those reflexive &q
353.
by kqr 8mo ago
Do we know if these models are also trained on scripts for TV series and movies? People in the visual medias surprisingly often take their bikes for walks.
354.
by kqr 8mo ago
Worse! It's trained to output coherent reasoning, so by putting the assumption last there's a risk it massages the assumption slightly to fit the conclusions it has already drawn.
355.
by kqr 8mo ago
I would assume sling release is not done by timing but by proprioception.
356.
AI Outlook with JetBrains(metaculus.com)
2 points by kqr 8mo ago|1 comments
357.
by kqr 9mo ago
What is the Nomad like? What were your concerns when selecting an epaper notebook? Which alternatives did you rule out and why? Sorry, just long wanted to get one but the good ones are expensive and I don't want to experiment with that
358.
by kqr 9mo ago
Or put a large file on a regular size SD card.
359.
by kqr 9mo ago
No, you're correct. I would have expected the highlights to blow out much sooner (for digital) and the shadows to block up much sooner (for analogue). The simulation doesn't portray this accurately, but it gives the general idea!
360.
by kqr 9mo ago
The kind of noise also makes a huge difference. Chroma noise looks like ugly splotches of colour, whereas luma noise can add positively to the character of the image. Fortunately humans are less sensitive to chroma resolution so denoising c

More ›