An important distinction is that the author's experience is mostly with the businessy side of data science, and his jab is at people who use buzzword tools that add complexity rather than simple solutions.
In defense of the hype, many tools like storm are worth their hype many times over when used for the right application.
The author makes this distinction, but it can easily be lost in the post.
I originally thought the same. Instead of immigration numbers decreasing during those years, it's possible a lot more people were born in the states relative to the number immigrating then. It turns out, though, that foreign-born population did actually decrease substantially:
I'm not sure. The y-axis on this graph is just the rate of power consumption. It doesn't do much for you to use minimal energy while traveling 0 mph.
Efficiency, in terms on miles/gallon, is always a trade-off between traveling fast enough that the overhead of running the vehicle is less significant but slow enough that wind resistance doesn't dominate. There's a sweet spot in there.
Yeah, I'm definitely confused. I don't understand how a car traveling 0 mph can have a finite value on that graph if the y-axis is [energy]/[distance]. I was hoping the units were [Watts][hours]/[minutes] or something to make everything better.
I found they have another graph with the same units (written "Wh/mile", so that's unambigious). But it has a different shape, with a minimum at 20-25 mph...
That graph shows a maximum efficiency at around 22 MPH (for the 85 kWh battery), which makes perfect sense -- any faster and air resistance becomes a factor. Any slower and energy uses apart from propulsion begin to eat away at efficiency -- the computer, lights, air conditioning, heat, and so forth.
> Yeah, I'm definitely confused. I don't understand how a car traveling 0 mph can have a finite value on that graph if the y-axis is [energy]/[distance].
The car expends energy just sitting at zero MPH -- the computer and lights, air conditioning and heating, things like that. So the graph is correct -- there is an energy expenditure to "go" zero miles per hour.
> I was hoping the units were [Watts][hours]/[minutes] or something to make everything better.
When the speed is zero, it stops mattering which units it's expressed in. :)
When speed is zero, distance is zero and energy/distance is infinite. Hence the OP questioning how could the graph present a finite value at that point.
Personally, I'd wager they followed the old scientific adage: "If you wish your function to be linear, sample two and only two points".
> When speed is zero, distance is zero and energy/distance is infinite.
1. Not energy/distance, but energy/speed.
2. Notwithstanding the cart's labeling ("Wh/mi"), its values aren't predicated on a division of energy by speed. "Wh/mi" doesn't literally mean mean "watt-hours divided by miles per hour", it means "the relationship between watt-hours and vehicle miles per hour".
> Personally, I'd wager they followed the old scientific adage: "If you wish your function to be linear, sample two and only two points".
Yes, except the chart isn't linear. Just for fun, here's a polynomial function that matches the chart's results reasonably well:
I like it. For me, the Twitter integration with @NFLScoreBot is the best part.
I'm sure there are a lot of issues getting video highlights on the site, but if that were somehow possible and you added a link in each tweet, it'd be something I follow and occasionally click on... i.e. a nice app.
Video highlights as a link out would be possible (although I certainly couldn't embed them). Maybe I'll add a highlights link on mouse over, so as to keep it relatively uncluttered.
In defense of the hype, many tools like storm are worth their hype many times over when used for the right application.
The author makes this distinction, but it can easily be lost in the post.