4 ms·

I think there's room for both models. I care about what I eat, but the way calorie counters work now is that they have too much friction for someone who's basic

I think there's room for both models. I care about what I eat, but the way calorie counters work now is that they have too much friction for someone who's basically guessing most of the time anyway. You can add some minor qualifiers to help narrow down the quantities. > "Hey Jemaclus, it's 2pm. What'd you have for lunch today?" > "A burger." > "On a scale of 1 to 5, with 5 being the healthiest, how healthy was the burger?" > "4" (I made it myself) Ping me with a push notification -- hey, time to enter it! -- then ask me a quick question or two to qualify the results. How healthy was it? How many common-sense servings do you think you just ate? My point is that the numbers regular calorie counters give are very exact numbers -- 489 calories for a Big Mac -- but food is rarely served perfectly. Two burgers cooked within 10 minutes of each other at the same restaurant might vary wildly in the number of calories, depending on the size of the patty, the amount of seasonings and butter and what-have-you added to the dish, and so on. At the end of the day, I'm just guessing. On calorie counting apps, the serving sizes are almost always useless. This burger is measured in ounces -- is that before or after cooking? How many ounces is the burger I just got from the restaurant? I know a Quarter Pounder with Cheese from McDonalds is, well, 4 oz (before cooking!), but does that translate 1:1 with my app? Who the hell knows? I think if you're neurotic or otherwise obsessed with tracking your calories super seriously, then maybe you're right. But in my experience, most of my friends who start using calorie counting apps stop using calorie counting apps, because it's so much damned work to track everything they eat all day long. I'm advocating a smarter system. I'm guessing, which is more work than I want to do. Do the guessing for me. Do the work for me. Prompt me when you don't think you can make an accurate guess. I'm happy to say "I think it was 1.5 servings and relatively healthy", but I don't need to sit here and compare a 4oz burger from McDonalds against a 1/3 lb burger from In 'n Out against a "2 patty" burger from Joe's Burger Shack and try and figure out which one more accurately reflects the burger that's sitting in front of me right now.

dragonwriter 12y ago
> I care about what I eat Then you can't avoid learning to estimate portions. And once you do that, the way most standard food logging apps, web or mobile, work should generally work (there is room for improvement in friction, in some cases, but not in the particular area of making the app guess more.) There's no shorter cut from which meaningful results can be gathered. And its not that hard. > On calorie counting apps, the serving sizes are almost always useless. They aren't "useless", they do require you to understand portions. They replace manually logging and tallying what is logged, not actually recognizing what it is you are eating. > This burger is measured in ounces -- is that before or after cooking? Pre-cooked. Its the generally-accepted standard (most logging systems I've seen expressly indicate this, though some don't). > I think if you're neurotic or otherwise obsessed with tracking your calories super seriously, then maybe you're right. But in my experience, most of my friends who start using calorie counting apps stop using calorie counting apps, because it's so much damned work to track everything they eat all day long. Calorie counting apps aren't useful, though, if the user isn't visually making meaningful estimates (and, yes, they'll generally be estimates if you aren't preparing it yourself) of quantitative portion sizes and specific included elements beyond "1 to 5" ratings of healthiness -- because of the very kinds of variability that you yourself point to. Calorie counting apps can facilitate the process of tracking, but they can't replace the recognition and quantification of what you are eating part (which, yes, is initially the hard part of food tracking -- but its also the part that gets easy with experience.) There's some pretty simple rules of thumb for visually estimating portion sizes, etc., that can be helpful if you aren't eating a predefined menu item that is already in the apps database, and its possible that UX could be improved by facilitating access to guidance on those when logging such items. But that's different than the model you propose, which has serious GIGO problems. > I'm advocating a smarter system. I'm guessing, which is more work than I want to do. Do the guessing for me. Do the work for me. Sorry, you're the one who has the sensory input from which to make a meaningful estimate. Its obviously not impossible for a machine to do this, given the right sensors and the right application of them to the food, but its not going to save you any work on a smartphone platform. > I'm happy to say "I think it was 1.5 servings and relatively healthy Yes, but that's not meaningful input from which any useful information can be gained. Unless you have a well-calibrated consistent definition of a serving -- in which case, you can just as well say how many ounces, etc., it was.

dragonwriter 12y ago

> I care about what I eat Then you can't avoid learning to estimate portions. And once you do that, the way most standard food logging apps, web or mobile, work should generally work (there is room for improvement in friction, in some cases, but not in the particular area of making the app guess more.) There's no shorter cut from which meaningful results can be gathered. And its not that hard. > On calorie counting apps, the serving sizes are almost always useless. They aren't "useless", they do require you to understand portions. They replace manually logging and tallying what is logged, not actually recognizing what it is you are eating. > This burger is measured in ounces -- is that before or after cooking? Pre-cooked. Its the generally-accepted standard (most logging systems I've seen expressly indicate this, though some don't). > I think if you're neurotic or otherwise obsessed with tracking your calories super seriously, then maybe you're right. But in my experience, most of my friends who start using calorie counting apps stop using calorie counting apps, because it's so much damned work to track everything they eat all day long. Calorie counting apps aren't useful, though, if the user isn't visually making meaningful estimates (and, yes, they'll generally be estimates if you aren't preparing it yourself) of quantitative portion sizes and specific included elements beyond "1 to 5" ratings of healthiness -- because of the very kinds of variability that you yourself point to. Calorie counting apps can facilitate the process of tracking, but they can't replace the recognition and quantification of what you are eating part (which, yes, is initially the hard part of food tracking -- but its also the part that gets easy with experience.) There's some pretty simple rules of thumb for visually estimating portion sizes, etc., that can be helpful if you aren't eating a predefined menu item that is already in the apps database, and its possible that UX could be improved by facilitating access to guidance on those when logging such items. But that's different than the model you propose, which has serious GIGO problems. > I'm advocating a smarter system. I'm guessing, which is more work than I want to do. Do the guessing for me. Do the work for me. Sorry, you're the one who has the sensory input from which to make a meaningful estimate. Its obviously not impossible for a machine to do this, given the right sensors and the right application of them to the food, but its not going to save you any work on a smartphone platform. > I'm happy to say "I think it was 1.5 servings and relatively healthy Yes, but that's not meaningful input from which any useful information can be gained. Unless you have a well-calibrated consistent definition of a serving -- in which case, you can just as well say how many ounces, etc., it was.