News for hackers. Stuff that matters.

metalens

I'd like to add that those moves we (assuming you're referring to the US) made 100 years earlier were not made with liberty intact. A lot of our progress was built on the back of slavery and racial discrimination at scale in housing, education, transportation, jobs, etc.
Re:Reddit Welcomes Michael Seibel to Board of Directors on Thursday June 11, 2020 @10:32AM (#23489306)
Absolute agreed with you here.
Re:Reddit Welcomes Michael Seibel to Board of Directors on Wednesday June 10, 2020 @08:00PM (#23483619)
I am not sorry for him. What is your definition of merit for a board position?

As far as I know, board positions are often filled based on explicitly subjective criteria, e.g. such subjective criteria might be: Who is currently on the board and what personality/skillset/experience does the board need in order to balance its membership? Who does the company & board need in order to: raise as much money as possible, be introduced to the desired go-to-market partners, look good to customers, [insert any other subjective criteria here] etc.

I am not aware of the concept of objective "merit" when it comes to board of directors appointments.

~corrected for bad grammar~

Re:Researchers: Are we on the cusp of an ‘AI winter’? on Friday January 17, 2020 @07:11PM (#22081089)
I used to do electromagnetic modeling using finite element methods (though now a product manager for AI software infra) and it would to take me on the order of hours to days or weeks to model wave interaction with real-world objects.

A machine learning model trained to understand Maxwell's Equations can in principle be used perform said simulations, resulting in probably an order or more of magnitude increase in simulation speed. Getting this to work well will reduce the time (and cost) it takes to design optical sensors, radar for autonomous vehicles, smartphone antennas, MRI machines, and more.

Having said that, it would require a lot of heaving lifting to pull this off to achieve near-physical accuracy for real-world physics problems.

A cursory search on Google for "arxiv deep learning electromagnetics" returns results of proofs of concept in this direction.

Legally questionable?
Re:Hard Part of Computer Science? Getting into Class on Thursday January 24, 2019 @07:00PM (#18994285)
Spot on.
>>90% percent of defendants plead guilty before trial because 90%+ of defendants are actually guilty of the crimes they've been charged with. In most cases, the defendants accept their guilt and are simply trying to get the best deal (i.e., least prison/jail time) as they can.

Except that a disproportionate percentage of poor defendants (who can't afford expensive lawyers) plead guilty before trial because their overworked and underpaid public defenders advise them to.

Logic is that pleading guilty before trial gets you a shorter sentence than going to trial. The alternative is going to trial and being incorrectly found guilty and getting a larger sentence for a crime you didn't commit (your public defender has limited time/resources to make a great case for you).

The American Justice system is particularly unjust to the poor.

Edited for clarity