Wednesday, August 23, 2017

the eclipse

It is very disconcerting to point your telescope directly at the sun, take the filter off, and immediately look through the eyepiece. I've spent most of my life learning to be very careful not to do just that. But that sight, with true-color bright pink prominences, glowing tendrils of the corona against a royal blue background, and the enormous, terrific void in the center, was hands down the single most incredible thing I have ever seen.

I've seen thousands of photos of the sun and its various types of surface features, and I've seen thousands of photos of eclipses, and read hundreds of elated accounts, and I've seen lunar eclipses and an annular solar eclipse, several partial eclipses and a Venus transit, but this was the difference between reading a textbook about string theory and sneaking a glance through a fleeting crack in spacetime itself. It's the difference between kissing a man and marrying him; the difference between a ferris wheel and skydiving; between a bathtub whirlpool and a tornado. No matter how much you objectively know what is coming, it's impossible to be adequately prepared. I am not at all surprised that ancient humans thought they were staring into the end of times, and they didn't even have a magnified view.

Eclipse progression from Madras, OR

And if anyone should have been prepared, it's me. I've been looking forward to this eclipse for the last 20 years (ironically because it was supposed to be the first eclipse I wouldn't have to fly around the world for. Whoops.) This was even before I owned my first telescope and was still contenting myself with learning constellations, hunting for Hyakutake and Hale-Bopp, measuring latitude with a protractor, and waiting 45 minutes on dial-up internet to download photos of Neptune. (I was not the most popular kid...) I can't adequately describe the eclipse directly, but surely you can infer from that kind of history that I had pretty high hopes and expectations, so if I'm still paralyzed by awe 36 hours later, it really had to have been pretty great.

Partial phase, with (unusual for this time during the solar cycle) sunspots. The grainy surface appearance is due to temperature variation (not digital noise).

Why exactly is it so incredible? I wish I had the faintest idea. I've experienced sunset and the associated sky colors, drop in temperature, and darkness. It's "just" a black circle in the sky. I don't even believe in anything supernatural that could have turned it into a spiritual experience. Sure, seeing things with your own eyes is often much more powerful than in photos, but nothing else purely visual, not even looking into the Grand Canyon, or seeing supernovae in other galaxies with my own eyes, or swimming with manta rays, has come close to giving me an adrenaline rush that took half an hour to dissipate. I've watched my own little gopro video of totality (below), in its gloriously crappy quality,  a dozen times already and it gives me chills just from remembering it. Vividly remembering bungee jumping also gives me a bit of an adrenaline rush, but that's a memory of jumping off a bridge and repeated freefall, not ... looking up. Matt's knees were shaking ages after totality was over, I had trouble firmly taping the solar filters back on the telescope with shaking hands, people around us were crying. I think if we could figure out what primal nerve these things hit and why, we would understand humanity much better.



Matt and I drove up to Oregon to see it with two friends who live in the Bay Area. On the way there I read up on photographing it, which I hadn't had time to think about and which I'd assumed I would only make a very cursory attempt at since there is no spare time during 2 minutes and 5 seconds of totality to waste fiddling with equipment instead of looking at it directly. But I found out that it's really easy to automate the shooting by plugging the camera into a laptop and scheduling a series of bracketed captures in darktable (the open source imitation of lightroom). I set my camera to F4, ISO200, 200mm (the maximum of my good quality lenses), 1/30 basic exposure, but then 10 exposures around that value ranging from 3 seconds to 1/3200 in order to capture different levels of detail. All during the partial phase I was doing something similar with manually tweaked settings as the conditions changed, but I planned on these values ahead of time for totality based on a bit of experimentation the day before and recommendations from the internet. Luckily I wrote them down, because with 2nd contact rapidly approaching I was already losing it over watching the tiny sliver in the telescope break into individual Bailey's beads and then disappear. With 30 seconds until totality or so I hurriedly punched in the settings, pulled off the filter, and pressed start (telling it to cycle through those exposures repeatedly until I stopped it) and looked up just in time to see the spectacular diamond ring, more clearly than I could have ever expected, more clearly than anything I was able to capture photographically (due to not controlling which exposure it was taking at the exact second needed), and I swear more clearly than I've seen in almost any other photo either. Unbelievable.

Outer corona during totality

Shorter exposure showing the inner corona and three pink prominences (at 11:30, 1, and 3).


And just that fast, it was gone, and we plunged into darkness. Venus showed up like a spotlight, and sunset colors spanned all 360 degrees of the horizon. People cheered; Matt played Pink Floyd's "Eclipse", which at 2 minutes and 1 second, almost had to have been specifically written for this time and place... I was completely taken aback by the naked eye sight overhead, but Matt had the sense to look through the telescope and exclaimed about the bright pink prominences. I took a look, and did a double take, and then a triple take, and now that image is permanently seared in my memory. I obviously can't stop myself from continuing to fail to describe it, so one more try: it's not just more beautiful than you expect, it's not just surprisingly moving, it's like staring straight at something you know in your core you're not supposed to be allowed to see, something that may have dire consequences, but that you most certainly can't look away from. And again, I have no idea why. Is it the strongly conditioned hesitancy about looking straight at the sun through a telescope? Is it the hole in the fabric in the universe that looks like a tunnel to the afterlife? Is it simply too alien to process with existing neural connections? Perhaps all of the above would begin to come close to explaining the adrenaline rush it caused. I wish I could at least share a picture, but nothing I can find online matches that view.

Even faster than it began, it ended. A neighbor set off fireworks; skydivers landed at the airport across the street; I flipped the filter back on the telescope as fast as possible with some dubious bits of scotch tape that Matt sensibly reinforced and watched the second set of Bailey's beads form and merge into a larger crescent, and when I looked up ten seconds later it was once again hot and brilliantly sunny, even with only a few percent of the sun uncovered. As soon as the moon fell behind the sun on the tail end we hit the road, which turned into an Oregon crossing at an average of 12 miles per hour. And it was worth every single second.

See you in Oklahoma in 2024.

Any photoshop experts want to help me make a better exposure stack than this one...?

Friday, August 11, 2017

MTurk tips

2 easy hacks to make MTurk/Web/Qualtrics data collection/management easier:

1. isn't actually a hack, just a recommendation to send users to an experiment that is either hosted on your own server or goes through your own server as a landing page (before redirecting to qualtrics or whatever). This serves a couple of purposes. You have complete control and complete records of absolutely every interaction anyone has with the website. You can see the IP and entry point and time of everyone who loads the site, you can record exactly when and where they click, you can manage random assignment to treatment groups yourself in a way that keeps samples balanced or meets any other constraint you have, you can see how many people open the survey and decide not to do it, etc. I also send users back to my server (with an automatic redirect from qualtrics) so I can mark them off as having completed the study.

2. is a hack assuming you do #1: make sure your server keeps all access logs from a long enough time span that at the end of the study you can keep a full record of all interaction. I cannot tell you how many times this data has been useful to have. With the timestamps and IPs I can tell when someone in the lab changed computers because he entered the wrong url and got an error on his first computer, I can tell you who used their phone, who restarted, who finished but just didn't click "submit" at the end, etc etc etc. There are always a few mystery people in the data who didn't do things the right way or had technical issues and I've always been able to track down what happened this way.

3. is a hack assuming you do #1 and #2 in combination with using qualtrics (or any similar third-party platform). Instead of hosting photos or other imported media on the third-party platform, host them on your own server and load them via url on the other platform. Every access of this kind will show up in the access logs, giving you better timing information and IP tracking than those platforms will usually tell you. Even qualtrics, which is phenomenal and and will give you timing information for every page of every survey, only records the time of the first and last click, but if the page is loaded and immediately exited you won't be able to tell, and if something is clicked that opens a sub-question and then closed to avoid the sub-question you won't be able to tell, and if someone fails a verification check and has to re-do a page you can't tell, and so on and so forth. You can make hacks of this kind arbitrarily fancy with custom code.