I work with Indigenous tribes & Indigenous activist groups throughout the west and midwest, everyday. They can be confrontational, no doubt. But, I also don't actually fault these brats. They just remind me of all of the jock ass ignorant privileged scum I grew up with in PA being confronted with something they aren't prepared for.

Not to mention the fact that we are now fully in a digital age of Google algorithms that basically give a person exactly what they are looking for and nothing else (an internet search used to be research in the 90's, siphoning through a lot of info). People are growing up in very insular bubbles; and ignorance breeds from that. One should also fault the educational system; but in a place like Kentucky, I'd imagine the social media bubble is fueling these kids' education more than their public schools are; there's always college (hopefully).

"Only 3% of people who use google have actually read the terms and conditions that they agreed to. "

3%?? I would have been amazed if it was as high as 0.3%.
3% would be (conservatively) over 10 million people. I doubt it's anywhere close to that.

I am not sure that privacy as a concept is even possible in a world with machine learning algorithms and big data. That's not a value judgment; I don't think privacy is worthless, I just find it increasingly untenable.

Machine learning has gotten so good, that even if you anonymise data, it's now pretty easy to tell a lot about you. Your digital fingerprint is there and an AI will be 99% correct about your age, gender, politics, sexual orientation, etc, even without you giving up that data.

I mean..., with the right algorithm, we can do anything!

As hamsteralliance says, ContentAware uses proximity matching and relative area matching. If you tried to fill in the white space with ContentAware, it'd be full of everything except eyes. They nVidia folks used thousands of images to train the neural net (ie generate the model using training data) which has more discrete sequential or spatial relationships between features (ie. eyes go to either side of the nose, below the eyebrows, level, interpupilary distance etc etc). The neural approach ALWAYS needs training data sets - it doesn't appear to (from reading the paper) any adaptive or learning algorithm outside of the neural framework (so, it's not AI in the sense that it learns from any environmental stimulus and alters its response... that I can see anyway. The paper doesn't get into the minutiae). But I'd still date her, if only she'd have me.

"I've always wondered myself about keyless tech safety for this exact reason. How can the signal not just be copied and replayed?"

Well, I can't say for certain, but if I was designing it, the signal wouldn't be the same every time. Basically, you would have an algorithm that generates a signal (essentially a large number encoded as a binary stream) based on a seed and the current time.

The seed is unique to the car and the key.

So when you press the button, the key does something like

entryCode = SomeComplexAlgorithm(seed, time())

so the car would do something like

entryRequest = GetSignal()
checkedRequest = SomeComplexAlgorithm(sameSeedAsKey, time())
if (checkedRequest == entryRequest) Unlock()

That's obviously a vast oversimplification (not sure how they'd get around the time sync), but you get the idea.

What surprises me here is not that the car starts, but that it doesn't cut out once it gets out of range of the key. Even a strong relay would only have a short range (1-2km at most?).

I can just imagine the poor machine learning algorithm to detect dicks.

Some poor bastard will have to spend hours looking at dick photos to help the system learn.

By far the best class I took while getting my Computer Science degree was "Software Engineering Project". We got assigned a project and divided up into teams including CS, CIS, and MIS majors. The MIS people managed and divided up assignments, and the CS people handled different aspects of the programming, like data structures / algorithms / UI / whatever.

This looks like an extremely fun and interesting extension of that. There's CS guys programming, EE guys doing hardware / sensors / haptic panels, full-on Engineering guys doing fluid dynamics, etc.

Destin seems like a great "jack of all trades" type that can get in there and ask really smart questions off the cuff. All the guys geeking out and being impressed with his intuitions and yet hesitant to confirm anything is hilarious to watch.

If it's a learning algorithm, then we can teach it not to hurt and kill. Think of it as a dog. I wonder if eventually we'll have to program in some equivalent of "treat" and "newspaper" so that it can be trainable.

So here’s the thing. I’m willing to bet large amounts of money that the robot is using some kind of machine learning algorithm.

Which means, that all we can do is set it a task, not tell it HOW to achieve the desired result. As a corollary to this, we also don’t understand the robots “reasoning”.

Where am I going with this?

Well, we’ve told the robot to go through the door. It’s figured out how to do this. And now, we’re fucking with it to force it to come up with new solutions. Well, the most obvious solution is to eliminate the pesky human preventing it from achieving its goal.

I’m not actually joking here.

*quality

Mildly terrifying too, but this is the future people.

genetic algorithm *engineering

I'm especially amused when it highlights the handbags/backpacks....I just assume this is the prototype for Bender's stealing algorithm...

I like it. Jumping around in front of it HELPS the algorithm.

Freedom of religion is independent of civilian armament.
History shows that religious persecution is normal for humanity, and in most cases it's perpetrated by the government. Sometimes to consolidate power (with government tie-ins to the main religion), and sometimes to pander to the grimace of a majority.

Ironically, in this country, freedom of religion only exists due to armed conflict, albeit merely as a side effect of independence from a religiously homogeneous ruling power.



It's true that Catalonians would likely have been shot at if they were armed.
However, likewise, the Spanish government will never grant the Catalans democracy so long as the Catalans are not armed - simply because it doesn't have to.
(*Barring self suicidal/sacrificial behavior on part of the Catalans that eventually [after much suffering] embarrasses the government into compliance - often under risk that 3rd parties will intervene if things continue)

When the government manufactures consent, it will be first in line to claim that people have democratic freedom. When the government fails to manufacture consent, it will crack down with force.

At the end of the day, in government, might makes right. Laws are only words on paper, the government's arms are what make the laws matter.

Likewise, democracy is no more than an idea. The people's force of arms (or threat thereof) is what assert's the people's dominance over the government.



You can say the police/military are stronger and it would never matter, however, the size of an [armed] population is orders of magnitude larger than the size of an army. Factor in the fact that the people need to cooperate with the government in order to support and supply the government's military. No government can withstand armed resistance of the population at large. This is one of the main lessons from The Prince.

Civilian armament is a bulwark against potentially colossal ills (albeit ills that come once every few generations).

Look at NK. The people get TV, radio, cell, from SK. They can look across the river and see massive cities on the Chinese side. They know they have to play along with the charade that their government demands. At the end of the day, without guns, things won't change.

Look at what happened during the Arab Spring. All these unarmed nations turned to external armed groups to fight for them to change their governments. All it accomplished was them becoming serfs to the invited 3rd parties. This is another lesson from The Prince : always take power by your own means, never rely on auxiliaries, because your auxiliaries will become your new rulers.






Below is general pontification. No longer a reply.
------------------------------------------------------------------



Civilian armament does come with periodic tragedies. Those tragedies suck. But they're also much less significant than the risks of disarmament.
(Eg. School shootings, 7-11 robberies, etc -versus- Tamils vs Sri Lankan government, Rohingya vs Burmese government. etc.)

Regarding rifles specifically (all varieties combined), there is no point in arguing magnitudes (Around 400 lives per year - albeit taken in newsworthy large chunks). 'Falling out of bed' kills more people, same is true for 'Slip and fall'. No one fears their bed or a wet floor.

Pistols could go away and not matter much.
They have minimal militia utility, and they represent almost the entirety of firearms used in violent crime. (Albeit used to take lives in a non newsworthy 1 at a time manner)

(In the U.S.) If tragedy was the only way to die (otherwise infinite lifespan), you would live on average 9000 years. Guns, car crashes, drownings, etc. ~All tragedies included. (http://service.prerender.io/http://polstats.com/?_escaped_fragment_=/life#!/life)






A computer learning example I was taught:

Boy walking with his mom&dad down a path.
Lion #1 jumps out, eats his dad.
(Data : Specifically lion #1 eats his father.)
The boy and mom keep walking
Lion #2 jumps out, eats his mother.
(Data : Specifically lion #2 eats his mother)
The boy keeps walking
He comes across Lion #3.

Question : Should he be worried?

If you are going to generalize [the first two] lions and people, then yes, he should be worried.

In reality, lions may be very unlikely to eat people (versus say, a gazelle). But if you generalized from the prior two events, you will think they are dangerous.

(The relevance to computer learning is that : Computers learn racism, too. If you include racial data along with other data in a learning algorithm, that algorithm can and will be able to make decisions based on race. Not because the software cares - but because it can analyze and correlate.)

(Note : This is also why arguing religion is likely futile. If a child is raised being told that everything is as it is because God did it, then that becomes their basis for reality. Telling them that their belief in god is wrong, is like telling the boy in the example that lions are statistically quite safe to people. It challenges what they've learned.)



I mentioned this example, because it illustrates learning and perception. And it segways into my following analogy.



Here's a weird analogy, but it goes like this :

(I'm sure SJW minded people will shit themselves over it, but whatever)

"Gun ownership in today's urban society" is like "Black people in 80's white bred society".

2/3 of the population today has no contact with firearms (mostly urban folk)
They only see them on movies used to shoot people, and on the news used to shoot people.
If you are part of that 2/3, you see guns as murder tools.
If you are part of the remaining 1/3, you see guns like shoes or telephones - absolutely mundane daily items that harm nobody.

In the 80's, if you were in a white bred community, your only understanding of black people would be from movies where they are gangsters and shoot people, and from the nightly news where you heard about some black person who shot people.
If you were part of an 80's white bred community, you saw black people as dangerous likely killers.
If you were part of an 80's black/mixed community, you saw black people as regular people living the same mundane lives as anyone else.

In either case, you can analytically know better. But your gut feelings come from your experience.



Basically, I know guns look bad to 2/3 of the population. That won't change. People's beliefs are what they are.
I also know that the likelihood of being in a shooting is essentially zero.
I also know that history repeats itself, and -just in case- I'd rather live in an armed society than an unarmed society. Even if I don't carry a gun.

-scheherazade

I went through and can't find the grading example that they had when I was dealing with this with my kids. If I can get the spacing right they showed the student's work as below, with the proper pen marks for 'carrying' if you were doing long hand multiplication:

37
*23
------
111
740
------
851

The marking guidelines stated that this was to be marked as INCORRECT, because the student was falling back and using the algorithm and the correct answer was to formulate multiple different strategies for solving the 'problem'.

A better answer would have been 10 times 37 is 370, so 20 times 37 is 740, then 3 times 37 is 111. So 740 plus 111 makes 851.

Even that though was NOT a good enough answer. No, the BEST answer was the above and then a second method like calculating 25 times 37 and subtracting 37 twice as an alternative solution.