Hacker News Reader

Hacker News

Comment by moralestapia | Hacker News Reader

Full thread

moralestapia

A bit tangential, but all of these are arbitrarily chosen constraints:

* Within some specific and relatively unimportant geographic area

* 1 year intervals ...

* ... that start at Jan 01 ...

* ... in the Gregorian calendar.

* The number 110 ...

* ... as a threshold of temperature in Imperial units.

I don't disagree with the essence of TFA (i.e. it's getting warmer) but it's also quite absurd how you can come up with any "record" you may need if you are controlling for the variables that suit you.

I'm pretty sure I hold several world records, like the fastest cyclist:

* on my hometown

* who is male

* with a graduate degree

* that likes hawaiian pizza

View on HN

Does it matter though? Nothing in this metric indicates a slight of hand by the author.

Picking a year, starting with the first day of said year in the most prevalent format of calendar and then counting days in which it has been uncomfortably hot is exactly the way I would choose to get the point across.

The fact that there exists such a place in the US is probably more important than that this place was choosen because it scored highest in above metric.

>days in which it has been uncomfortably hot is exactly the way I would choose to get the point across

I'm glad it works for you but what if 100F is uncomfortably hot for me?

>The fact that there exists such a place in the US is probably more important than [...]

Come on, I'm sure it's not the hottest place in the US, a quick google search gave me Death Valley, CA.

They really aren't that arbitrary, except for 110, but even that is reasonable given our decimal system:

* Arbitrary but socially meaningful geographic boundary. Basically, a notable US city.

* 1 year intervals are actually canonical given that years are based on Earth's rotation

* Gregorian calendar / starting at Jan 1 is again socially meaningful, but in particular doesn't split summers in the northern hemisphere

* 110 and imperial are sensible given usage of Fahrenheit and decimal systems in the US.

Your comment makes it sound like the article is making a claim equally as meaningful as "at latlng 53.193,38.8493, Earth's temperature has been above 482.3 kelvin for 4 of the past 3952 days" which is really not.

Yeah, there's definitely some "p hacking" to make spin this into a newsworthy headline, but not nearly to the degree your comment makes it seem. And not to mention the issue of "110f in Phoenix" has been newsworthy previously (eg 3 straight weeks of 110+), it makes even more sense.

>socially meaningful geographic boundary

For you ... and maybe a minor percentage of the US, which, btw, makes up only 4% of the total world population.

>again socially meaningful

As I commented elsewhere, nature doesn't care when you start counting days.

>110 and imperial are sensible given usage of Fahrenheit and decimal systems in the US.

And so? Countries that use Celsius are warmer? Colder?

>the issue of "110f in Phoenix" has been newsworthy previously

Do you live in Phoenix? This is the first I've ever come across such piece of news. Ever.

> For you ... and maybe a minor percentage of the US, which, btw, makes up only 4% of the total world population.

Yes truly strange to find a discussion about the American city of Phoenix here on this website given that the US “only” has 4% of the world’s population.

Well..

Is a year really "arbitrarily chosen"? After all it's defined by our solar system. And for the start date of Jan 01, I would go out and claim, that years are most of the time compared starting Jan 01. It would be much more random to start this record on Feb 20.

Yes one may debate if 110 as the cutof to count days in is "the best choice", but well, it's in the US and we humans like nice numbers ;)

But all these values are not excluding like your example.

>that years are most of the time compared starting Jan 01

In accounting, maybe, nature DGAF about when people file taxes, lol.

I feel like a lot of the climate change claims are just confirmation bias

Reply on news.ycombinator.com