It's always interesting when statistics correlate across multiple frames (time, location, etc.); however that still doesn't prove causality. I'm inclined to think that the causes behind an area being slow to remove leaded gasoline could also affect crime rates.
>I'm inclined to think that the causes behind an area being slow to remove leaded gasoline could also affect crime rates.
The statistics show a 20-year lag between the banning of leaded gasoline and reduction in crime rates. So even if "unwilling to ban leaded gas" and "high crime 20 years later" could be correlated, it seems less likely that the same things that cause "willing to ban leaded gas" are also the cause of "lower crime 20 years later".
That's not entirely true. One factor could be that less progressive urban areas are both less willing to remove lead from gasoline and less willing to fund community outreach programs. Community out reach programs may take years to gain traction and become effective. Banning gasoline is witnessed in a much shorter time frame. This would explain the lag between crime rate drops and banning of leaded gasoline.
Its interesting to model the biosphere hydraulically.
So lead from leaded gas drips into the barrel representing the local ecology, and the lead contaminant drips out of the biosphere at a rate that obviously must depend on local climate. Presumably lead contaminant flows into the oceans or gets buried in the ground at a rate dependent on local weather.
Yet the stats show a nearly perfect and consistent shift 20 years after the addition of lead to the environment stops. That's just biologically weird, like a part of brain development doesn't depend on local lead levels, but on the first derivative of lead levels or something.
It COULD make sense if lead only causes problems when inhaled. Not consumed in food, not drinking water, purely and exclusively inhaled. Even then, local dust storms and the like should totally mess with the supposed global 20 year effect latency.
The font and underlining change are a non issue for me (other than looking terrible in chrome for windows); however the increased font size is jarring and reduces the amount of information that can quickly be scanned.
Interesting. What about including a second column showing the back correlation? An example: 'WPF' appears 6% of the time for 'C#' questions, while 46% of 'WPF' questions include the 'C#' tag. Would be interesting use this to identify ontological hierarchy trends.
This is exactly why metric is so powerful: I also have no inherent concept of a deciliter or decagram, but I can convert to my understood liter and kilogram in moments.
Because evolution relies on survival of the fittest and genetic mutation. Your Chi-Pom-Poodle-Terrier isn't an evolutionary step for the canine. And most of all I don't want creationists to think that 'hand of god' is a valid evolutionary driver ;)
This might be a minority opinion, but I think more than home data collection (which they probably have a lot of already from phone / computer usage), that this is about acquiring a company that has cache and a 'cool' factor within both the design world and more importantly the market sector of young tech inclined home owners - a hugely lucrative market.
I think there's a middle ground where you are giving the gift of x dollars in real food / livestock to a family in xyz. Then the actual logistics of what that means are figured out and told to you.
I agree with you to a point. Joins are your friend. But trying to pull out all of the information about a graph of 'objects' using a single query with multiple one-to-many and many-to-many joins is just as foolish in SQL as in Mongo.
