This. I recently read a book that explored the history of the Supreme Court and some of its major decisions. I walked away with one thing: the Constitution can mean virtually anything to anyone. It really does come down to the social winds of that particular time.
Well, the alternative is to be an originalist, trying to determine what a bunch of dead white guys in the 16th century -meant- when they wrote it, which is big in a number of conservative circles, and amongst a number of conservative justices.
Personally, I find that an unhelpful approach because the world, and yes, society, change. But it -can- be a helpful guiding star. Those who don't learn from history, etc; well, we have this document that was very intentionally trying to protect against a variety of evils that had been directly observed. Reintepreting it to today gives us value, in that it's a warning against behaviors that are detrimental to society. What exactly is the extent of freedom of speech? The document won't tell you that, or what a good extent of it should be, but it -does- tell you that -it's important-. The countries with the most restrictive speech, the ones where the wrong words routinely get you thrown in jail, where there is little to no free press...there is no document the government is predicated on saying that freedom of speech is important. Etc.
> Well, the alternative is to be an originalist, trying to determine what a bunch of dead white guys in the 16th century -meant- when they wrote it ... I find that an unhelpful approach because the world, and yes, society, change.
That's precisely why there are ways to amend the constitution. And we have been doing so for a long time - the most recent amendment is only 25 years old. That's the originalist way to dealing with the world and society changing, and the benefit of it is that it makes the change less arbitrary, and more democratic (since it's up to state legislatures, which are elected) than judicial fiat.
Freakonomics did an episode not too long ago with the new CEO of Ford. The guy was practically salivating at the mouth about all the data new vehicles will be collecting and how Ford could potentially monetize it all. Scary times ahead.
I submitted that article on HN but it didn't get any attraction.
It seems Ford CEO thinks they can collect and monetize drivers data:
-- So the case I would make is that we have as much data in the future coming from vehicles, or from users in those vehicles, or from cities talking to those vehicles, as the other competitors that you and I would be talking about that have monetizable attraction.
--The issue in the vehicle, see, is: we already know and have data on our customers. By the way, we protect this securely; they trust us. We know what people make. How do we know that? It’s because they borrow money from us. And when you ask somebody what they make, we know where they work; we know if they’re married. We know how long they’ve lived in their house, because these are all on the credit applications. We’ve never ever been challenged on how we use that. And that’s the leverage we’ve got here with the data.
It really comes across that he doesn't understand what he's talking about around tech. "Transportation Operating System"? It's cargo-culted technobabble.
It is a smart business move as there is money to be made.
Insurance companies did it first by offering dongles that you can plug into your OBDII that would basically feed your driving data back to them.
I agree that it is scary and just feels not right. General population is so ignorant these days that most honestly don't care. They will just accept these things.
This data will be valuable to insurance companies, government, car manufacturers and who knows how else it could be used. You can literally tell who made a modification to their car and read all the data from the vehicle.....meh, future sucks.
One day we will wake up when we have zero freedom, all of our moves will be tracked...and we won't be able to have any privacy at all.
Ironic how everyone emphasizes speed and ease-of-use nowadays with new toolkits and frameworks and all I witness is more clutter and slower, clunkier applications all over the web.
I understand it currently only works with Waze. Despite Google owning Waze, I'm not sure if it has been integrated into Google Maps. Here's a story about the challenges of installing beacons in Chicago [1] vs tunnels near Logan Airport in Boston.
My Google Maps still goes a little crazy on Lower Wacker.
Also we need beacons in the loop. All those tall buildings really mess up navigation downtown.
I don't know what Uber did to make their implementation of GPS so terrible. Google maps can definitely have issues at times in Chicago but I rarely see an Uber driver use the built-in mapping. They use Google maps.
Side note on Uber express, if it says northeast corner, it will always be the southwest corner.
It doesn't work with Google Maps. Source: I was using Google maps to get to/from Northwestern Hospital on Lower Wagner today, and the GPS dropped the entire time I was down there.
Yeah, did that. Problem wasn't the initial exit, it was the rapid series of turns after exiting lower Michigan from lower Wacker that was the Problem. Took a few loops around the block before the GPS reacquired. I used to spend a bit of time in the area, but it was always on foot. First time really driving in that area (coming in from West suburbs, normally, I'd take Metra and walk/cab it), and first time to NW hospital (wife needed to see a specialist)
Honestly, either is fine. It depends on your learning style and priorities. If you want to jump quickly to getting state of the art results with rich datasets like images and natural language text, then start with the deep learning course. If you want to get a deeper understanding of feature engineering, model interpretation, and (especially) techniques for tabular data, start with the machine learning course.
Also, the machine learning course was originally the initial introduction for a masters program, so it is a little less intensive and fast moving. But it also assumes a little more math background in parts, since everyone in the program already was pretty familiar with linear algebra, probability, and statistics.
In the end, the two courses go together pretty well, so I don't think it matters too much what order you do them in.
Yes probably. For instance, there's been interesting recent examples from Pinterest and Instacart (amongst others) who found that DL-based approaches for tabular data reduced the engineering and maintenance headaches due to less need for manual feature engineering.
Although decision tree approaches maybe will continue to be faster to train in many cases. At the moment, it really helps to be familiar with both approaches.
