So here's an update from my last ice cream post. I did go ahead and try the spicy chocolate idea. One of the first things I learned about ice cream from Antonia was that you have to mix it up sweeter than you think because you taste the sweetness less once it's frozen. And the first thing I learned from this experiment was that this rule does not apparently apply to spiciness.
I started off with just tiny pinches of chili powder in the chocolate mix (from Ben & Jerry's Ice Cream Book) but could hardly taste it. So I kept adding more... and more... and more.... And maybe somewhere in here we also got some sort of adaption factor to make me taste it less, which would only complicate things. Anyway, I finally decided it would probably be okay, so I went ahead and put it all in the ice cream machine. And once it was frozen it was hot! This is not ice cream to eat lightly. I'd probably halve the chili powder if I were to make this again. If I had measured it in the first place.
Good, though!
Infrequently updated these days, but there are lots of thoughts about lots of things here.
Thursday, September 20, 2007
Saturday, September 15, 2007
eMusic and Last.fm Recommendations
I don't know if this is new in eMusic or if I only just noticed it, but it's kind of cool. It generates recommendations of artists you might like based on what you've already bought, which is nothing new. But the visual display of it is pretty slick. You can see which recommendations come from which other users via which artists, and you can watch the connections change just while you mouse over the different names in all three columns. Plus, of course, you can play samples right there. Nice.

The problem here is that I don't really have a reason to care much about things eMusic is recommending based on just a couple months of purchases. What I want is this UI on my Last.fm recommendations, which is based on 3 years of what I've actually listened to. The Last.fm radio is cool and everything, but it would be fun to have this alternate way to browse recommendations more deliberately. Though with so many more recommendations, this design would probably get unwieldy, so you'd have to work out some way to just examine specific parts of the network at a time. E.g. limit it to friends, or neighbors, or tags, or using the obscure/popular slider.

The problem here is that I don't really have a reason to care much about things eMusic is recommending based on just a couple months of purchases. What I want is this UI on my Last.fm recommendations, which is based on 3 years of what I've actually listened to. The Last.fm radio is cool and everything, but it would be fun to have this alternate way to browse recommendations more deliberately. Though with so many more recommendations, this design would probably get unwieldy, so you'd have to work out some way to just examine specific parts of the network at a time. E.g. limit it to friends, or neighbors, or tags, or using the obscure/popular slider.
Saturday, September 08, 2007
Your Brain On Music
Particularly interesting to me was the information about perfect pitch. A number of studies have shown evidence that many people, even non-musicians, are able to store absolute pitch information in memory. E.g. they could be trained to recognize or sing back an assigned note that they learned. In other cases, when asked to sing a favorite popular song, they would do so in the right key without any prompting (or even awareness that they were doing so). My guess is that if we taught babies to recognize notes the way we teach them to recognize colors, a great many more people would have perfect pitch. It seems to be built into the brain, but most people just don't know how to use it or develop it. Granted, the people who seem to have it "automatically" make it seem inaccessible to the rest of us, but I think that's just a natural misinterpretation of ordinary person-to-person variability.
Another interesting concept regards what it takes to become an "expert" level musician. It seems that amount of practice time really is the biggest factor in becoming really good at something, even more so than perceived talent. Ten thousand hours is the usual amount of time you have to put in to really reach that top tier of world-class experts. And that same number seems to apply not only to music but to everything else that has been studied in this context: playing chess, figure skating, fiction writing, and more. Unfortunately, 10,000 hours is about 3 hours a day for 10 years. Sigh. But I guess that's why we're not all the best musicians in the world.
I was also amused at how well Levitin "predicted" the existence of Last.fm... a few years after its creation. Personalized radio stations mixing stuff you like with new recommendations to try? Yep, got that. :-)
Tuesday, September 04, 2007
I Scream for Flavor Suggestions
So far I've made two flavors, both quite yummily successful. First was cinnamon-cardamom, from a recipe in one of the Moosewood cookbooks. (Thanks, Cass!) Second was nectarine, which I did with just a basic strawberry recipe and the obvious substitution.
I'm more than open to other suggestions, though, so leave a comment with your ideas!
Saturday, August 18, 2007
Fixing My Sleep
What prompted this re-reading for me was coming back from my vacation last month with a severely screwed up sleep schedule/ability. Between jet lag, dance camp, Harry Potter, and various kinds of stress, I was a wreck. I had three days in a row at one point where I only slept a scattered handful of hours, and even after that I'd have nights where I was exhausted but unable to fall asleep for hours. Then I'd alternate nights of no sleep with nights of sleeping like a rock. So I started experimenting a bit with my sleep schedule (what did I have to lose at that point?) and also started reading this book to see if what I was doing made any sense. Luckily it did.
What I started by doing was just moving my bedtime later by one hour every night, while still getting up at the same time in the morning. Ever since I took Dr. Dement's "Sleep and Dreams" class in college, I've tried to be very aware of my sleep debt and of allocating enough time to keep it low. The problem is that sleep debt can actually be too low, to the point where it negatively affects your sleep efficiency, and allows other causes of insomnia (of which I have plenty) to come in and take over. My alternating nights of good sleep weren't erasing all my sleep debt, but just enough to cause this problem, given the state I was in. My idea with the first stage of this experiment was to try to regain/maintain a regular sleep schedule, but not let myself get quite enough, so that I could still fall asleep before my mind starts running away with me each night. After adjusting for a little while, this was fairly successful. (Moderately improving stress levels have helped as well.)
The downside to this, of course, is feeling sleepier than I'd like during the day. In the current phase of the plan, I'm alternating nights again, but in a much less dramatic way than when I had no control over it. I just switch between staying up the extra hour or not, so it's not a huge difference, but so far it seems to be balancing well between getting more rest and still falling asleep without too much trouble. There's also the plus that I can time my later nights so that I get more dancing (like the fact that I actually stayed at FNW all the way to the end last night). This does go against the standard advice of maintaining a rigidly consistent schedule of sleeping and waking, but Dr. Dement does stress in the last chapter of his book that it's really just important to understand the principles, and that different things will work for different people. My original idea was to eventually work back to having a fixed schedule, but I'm starting to reconsider this. If this alternating pattern continues to work out well, I may just tweak it a bit to make sure the total sleep time is enough, but then leave it there. We'll see.
Another thing I've been doing -- independent of but beneficial to the sleep times experiment -- is working out and running more. Dr. Dement recommends regular exercise as being good for sleep, but adds that it's best to do it more than 3 hours before bedtime, so as not to keep you awake. This is another case, though, where it's important to really understand how I personally react to it. My favorite time to go running recently has been about 10 PM. Luckily, this not only helps tire me out, but it can be incredibly relaxing. Running is very meditative for me. Nowadays, my mind -- my sleep nemesis -- is never as calm as when I'm running. And when I'm done and cooled down, I can continue relaxing myself for a while afterwards and maintain much of that feeling until I'm ready to go to sleep. So for 3 or 4 nights a week, that's very helpful.
Subscribe to:
Posts (Atom)