famil

Artillery with spotters accomplished the same task.

William Catton is another visonary. I consider his book, Overshoot to be better than limits to growth.

This lecture sums up his work well: https://www.youtube.com/watch?v=5WPB2u8EzL8

A discussion between Li and Fujiyama: https://youtube.com/watch?v=r0yIMs2qqb4

Dominico Lorsudo's Liberalism: A Counter History https://www.goodreads.com/book/show/9670246-liberalism is a a must read for the history of liberalism.

Eliminating suffering and preventing (re)birth? Sounds like Buddhism.

What is the use case of blockchain besides evasion of law enforcement and speculation?

I'm ready to accept that these phenomena might really be of Earth origin.

Arstechnica had a good piece on how patent reform was killed in Congress:

https://arstechnica.com/tech-policy/2014/05/how-the-patent-t...

Actually that's exactly what happened in the 1920s onward.

Harvard’s admissions committee began using the euphemistic criteria of “character and fitness” to limit Jewish enrollment.

From: https://www.nytimes.com/2014/11/25/opinion/is-harvard-unfair...

how they're "just middlemen"

As are Uber/AirBnb and internet advertising giants Google and Facebook.

Have you tried IQ testing instead of algorithm questions to measure intelligence?

I agree, it's bad for candidates but companies might like it. If everyone gives out take home assignments, it would be harder for candidates to negotiate by lining up multiple offers.

In my case, every company that gave me a take home assignment had on-site white board interviews as well.

#2 and #5 are issues with the company/hiring manager.

#1, #4 and #7 are issues with traditional interview as well.

#4 It's incredibly difficult to accurately assess potential during an interview.

#1 and #7: This holds true for white-board interviewing. Most successful candidates (including myself) invested time to practice coding interview questions. I did over 150 questions on LeetCode and it dramatically increased my interview skills. But I guess the upside is that these skills are transferable across companies who do algo/ds interviews so my prep time isn't "wasted".

I think #3 is the only thing thats inherently an issue with the project/assignment interview.

https://www.youtube.com/watch?v=r8RxkpUvxK0

In this video, a Google hiring committee were given their own anonymized hiring feedback and wouldn't even hire themselves.

What do you think about take home projects/homework interviews? I've noticed that they have become much more prevalent among smaller companies as a replacement for phone screens. As a candidate, interviewing is already a time consuming hoop jumping process, so if the project will take more than 3-4 hours, I probably won't be interested but most companies I have applied to are keeping these projects short enough.

Sci-Fi Short about telerobots controlled by gamers: https://www.youtube.com/watch?v=UXX0TRtg5Vk

Jürgen Schmidhuber had an interesting comment on the Strange Loop in his AMA: https://www.reddit.com/r/MachineLearning/comments/2xcyrl/i_a...

As we interact with the world to achieve goals, we are constructing internal models of the world, predicting and thus partially compressing the data history we are observing. If the predictor/compressor is a biological or artificial recurrent neural network (RNN), it will automatically create feature hierarchies, lower level neurons corresponding to simple feature detectors similar to those found in human brains, higher layer neurons typically corresponding to more abstract features, but fine-grained where necessary. Like any good compressor, the RNN will learn to identify shared regularities among different already existing internal data structures, and generate prototype encodings (across neuron populations) or symbols for frequently occurring observation sub-sequences, to shrink the storage space needed for the whole (we see this in our artificial RNNs all the time). Self-symbols may be viewed as a by-product of this, since there is one thing that is involved in all actions and sensory inputs of the agent, namely, the agent itself. To efficiently encode the entire data history through predictive coding, it will profit from creating some sort of internal prototype symbol or code (e. g. a neural activity pattern) representing itself [1,2]. Whenever this representation becomes activated above a certain threshold, say, by activating the corresponding neurons through new incoming sensory inputs or an internal ‘search light’ or otherwise, the agent could be called self-aware. No need to see this as a mysterious process — it is just a natural by-product of partially compressing the observation history by efficiently encoding frequent observations.