The authors make it clear in the very first paragraph that they are studying gender non-contentedness (the desire to be the opposite sex, for whatever reason) and not gender dysphoria (incongruence between one’s experienced/expressed gender and assigned gender):

There’s obviously a world of difference between “I wish I had been born a boy” and “I am a boy, my body got it wrong.” Unfortunately this distinction seems to be getting lost in translation. Kevin Drum (who is usually more careful) seemed to miss the distinction completely, summarizing the paper in a blog post titled Raw data: Gender dysphoria in teens, with similarly mislabeled graphs. I expect Drum’s misunderstanding is accidental, but it’s unfortunate because such a misunderstanding can be used to further the argument that teens should be denied gender-affirming hormones and/or hormone blockers.

And right on cue, within hours Wesley Smith at the National Review writes Study: Most Gender-Confused Children become Gender-Conforming Adults. He uses the term “gender confused“, a non-clinical term that is usually synonymous with gender dysphoria but also used by assholes people skeptical of transgenderism to refer to general issues around trans identity. Regardless of Smith’s intended meaning he quickly gets to his main point:

Smith’s argument against puberty blockers would be non-sensical even if they were being prescribed for gender non-contentedness. After all, the whole point of blockers is to delay puberty (and the permanent physical changes that come with it) to give a child, parents and care team time to decide the best course of action — which is exactly what you want if there’s a chance you might “grow out of it.” But puberty blockers aren’t recommended after the age of 14 or so (the body needs puberty to promote bone density), so at that point there needs to be a decision: stop blockers and let the body’s natural hormones start to surge or start taking hormones of the appropriate sex (testosterone for trans-male, estrogen for trans-female). Either decision leads to permanent body changes so it’s not like doing nothing is the “safe” choice, and in the end the child, parents and care team will need to make a decision.

The fact that a lot of children grow out of gender non-contentedness certainly weighs against deciding to take permanent action for children who aren’t suffering from gender dysphoria — which is presumably why doctors don’t do it. As described in this 2022 paper on gender affirming care:

Hopefully this study won’t become the latest cause célèbre of folks who want to discredit gender-affirming care, but if it does then you heard it here first: it doesn’t say what they say it says.

Until modern times little was known about winds in the stratosphere because it’s too dry up there for cloud formation, but a rare opportunity came in 1883 when Krakatoa erupted with so much force that ash was thrown as high as 20 miles up. Over the next few weeks the high-altitude ash caused brilliant sunsets around the globe, and by tracking when these sunsets occurred scientists were able to determine that high-altitude winds around the equator were easterlies (i.e. coming from the east), and were a nearly hurricane speed of 65-70 mph. These winds were appropriately called the Krakatoa Easterlies.

Some 20 years later scientists started launching weather balloons and noticed that at least some of them showed winds to be westerlies, but the consensus view remained that stratospheric winds around the equator were mostly easterlies with the occasional westerly layer. It wasn’t until the 1950s that we started sending up daily sounding balloons and realized something much stranger was going on. The winds weren’t easterlies or westerlies — they were actually a stack of alternating winds that slowly descended like a barbershop pole at a rate of about 1 km per month. What’s even weirder is that the altitude where direction shifts follows a pattern that’s regular but not seasonal or annual. At first scientists thought it was bi-annual, but even that turned out to be wrong: it’s quasi-biannual, with a period of about 28 months. The stack also favors easterlies in the upper stratosphere and westerlies in the lower stratosphere, with westerlies dominating the lower altitudes for about 20 of the 28 months. In spite of being incredibly regular it wasn’t until just the past decade or so that scientists have finally able to explain the phenomenon to the point where it can be modeled.

The QBO was really important to Loon for two reasons. The first was maneuverability: Loon balloons maneuvered by catching winds at different altitudes in a range of roughly 110 to 55 hPa (15.5 to 20 km), which spans both the top of the tropopause and the bottom of the QBO. Since the top of the tropopause tends towards easterlies year-round, whenever the bottom of the QBO was westerlies (about 70% of the time) it was likely that we could go back and forth across that boundary and effectively stay in one place! Yay station keeping! Of course for about 8 months out of 28 we’d likely need a lot more balloons to keep service over a region, so that made it important to predict when that switch was coming so we could plan accordingly.

The other reason was more of a surprise. In the early days of Loon we were launching most of our flights either in the US or New Zealand, but by early 2016 we had shifted our strategy and were testing in earnest around Peru. We had downloaded an archive of historical wind data from NOAA spanning from 1980 to 2015, but found that our flights were far out-performing our historical analysis. I’ll save the details for another day, but suffice to say that after about a month of digging through our simulation code to figure out what was going on we eventually concluded that stratospheric winds in the summer of 2016 was just plain wonky in a way that happened to be to our advantage — and that’s when Rob let us know that atmospheric scientists were sounding the alarm that for some reason the QBO pattern was being broken for the first time since measurements began in 1953! The pattern righted itself a few months later, but this one occurrence made us all the more wary of relying too much on historical data in this era of rapid climate change (and all the more interested in tracking the pattern to get early warning if it strayed again).

When we started Loon we had no idea whether delivering wireless internet via long-lived high-altitude balloons was even possible, much less as a sustainable business model. No one had ever done it before, and there was no road map or guide book. So we consulted with whatever experts we could find, collected what data we could, and made our best guess. Often our first (and second and third) ideas didn’t work, but with each iteration we learned something new that brought us a little further.

I’m happy to say that those lessons will now live on past the end of Loon. Today we (collectively) announced The Loon Collection: a compendium of specifications, algorithms, design documents, raw data and lessons learned over the project’s ten years. The collection includes a 437-page PDF book documenting key lessons that we learned the hard way, which we hope can act as a mini-guide to all who come after us. We’re also making available the telemetry and sensor data from all 2,131 of our flights, comprising over 127 million samples (over 218 flight-years of data, mostly from 15 to 20 km above sea level). We hope that this data will be of use to meteorologists, climate scientists and other researchers studying the stratosphere, especially those that develop the forecast models that were so vital for our planning algorithms.

[Keeping Loon’s legacy aloft]

[Migrated from Google+]

[Migrated from Google+]

Two of my coworkers, David Stork and Jorge Moraleda, have recently worked out the mathematics for what they call Lake Wobegon Dice. According to their paper (currently submitted for publication), Lake Wobegon Dice are

a set of n non-standard dice that have the following paradoxical property: on every (random) roll of a set, each die is more likely to roll greater than the set average than less than the set average; in a specific statistical sense, then, each die is “better than the set average.”

The name, of course, comes from Garrison Keillor’s famous tag-line about his fictional boyhood town where “all of the women are strong, all of the men are good looking, and all of the children are above average.”

As an example, say I offered to play you in a game using three six-sided dice that have been specially manufactured to have the following number of pips on their faces:

The rules of the game are as follows: you pick one die which I must roll, and you roll the other two. If my die rolls higher than the average of your two dice (or equivalently, if my die rolls higher than the average of all three dice) then I win. Otherwise, you win.

With the faces chosen as above, there are only four unique ways the dice can fall:

As should be clear from the table, I have a two-in-three chance of winning the game, regardless of which die you make me roll.

Their paper presents a proof that there exist such a set of dice for every n ≥ 3, so long as you are free to choose the number of sides on each die and the number of pips on each side, and also provide a method for finding the optimal set. They also show that for any set of n dice it is possible to choose a number of faces and pips for each die such that only one die will ever roll below the mean on any given roll, and each die is equally likely to be the low die. This means in the game described above, for any set of n dice I have a probability of (n-1)/n of winning.

If you want to use similarly-sided dice then for more than three dice the number of sides required gets large very quickly. The optimal set for n=4 requires 12 sides per die, but for n=5 and n=6 you need 60 sides. That’s because their construction method splits each die two groups, each group having a probability of 1-in-1, 1-in-2, 1-in-3, etc. up to 1-in-n. It’s much easier if you allow heterogeneous dice, e.g. for n=6 you could use a combination of six-sided dice and a ten-sided die — a good excuse to break out your old collection of D&D dice.

Back in 2005 I posted about Zubbles — colored soap bubbles that don’t stain your clothing. Well, three and a half years later they’re finally shipping product! (Currently available as a two-pack of Presto Pink and Blazing Blue.)

Blue-eyed humans have a single, common ancestor

New research shows that people with blue eyes have a single, common ancestor. A team at the University of Copenhagen have tracked down a genetic mutation which took place 6-10,000 years ago and is the cause of the eye colour of all blue-eyed humans alive on the planet today.