[wnorris510]

I've done research in this space for many years at Google AI and now at SnapCalorie. The thing I find interesting is how confident people are in their ability to estimate portion size visually, and in truth how wrong they all are.

We published in CVPR (top peer reviewed academic conference for computer vision) and people are on average off by 53% and even trained professionals are still off by 40%. Basically if you want to have a higher level of accuracy you need to use a food scale or something that measure the volume of food, people just can't estimate portion sizes visually.

Oils, cooking fats, hidden ingredients are what people are most concerned about but they actually add far less error to people's tracking than portion. Nutrition5k is the paper we published if you want to check out more details on the breakdown of error most people get when tracking.

[UomoNeroNero]

I have been diabetic for 20 years. I have tried every method, app, plan, and tool, including systems falsely marketed as "smart." No method works or delivers decent results except for using a scale and weighing ALL the ingredients. For a diabetic, eating "out" is always a roll of the dice. The "fun" feedback from post-meal blood sugar is always a reminder of how "eyeballing a plate" is utterly useless.

[foxyv]

It doesn't help that food manufacturers intentionally make it hard to measure nutrition from most of their foods. They play around with serving sizes to hide carbohydrates making you have to do math just to keep up.

Sometimes they will round down on grams of macros after setting the serving size so they can claim it has zero sugar when it does in fact have tons of sugar. Tic-tacs are the worst about this. They claim they have zero everything despite just being sugar tablets.

[vanviegen]

In the EU, food manufacturers are required to label macronutrients (and salt) in mg/100mg or mg/100ml for fluids. Easy to compare, works great.

[knifie_spoonie]

It's the same in Australia as well. I'm a bit shocked that the US doesn't have this.

[oniony]

But are you actually?

[foxyv]

This makes so much more sense than the labels in the USA.

[steveBK123]

US food labelling is insane.

For example - lactose-free yogurt is often just regular yogurt with lactase enzyme added.

If that's what I wanted, I'd buy regular yogurt and take a lactaid supplement.

[BoiledCabbage]

What other method would you deem appropriate for removing lactose from milk? A targeted enzyme that removes it seems pretty wise to me.

Since they're not gonna use tweezers, :) are you suggesting instead engineer or breed a special set of cows that don't produce lactase in their milk?

[UomoNeroNero]

Yes. I can find everywhere on labels the carb amount. I use 2 app too. And after a lot of errors I acquired a six sense (that try to kill me everyday ;-D)

[BobaFloutist]

My favorite example is that cooking spray advertises 0g of fat, giving a serving size of 0.33 seconds of spray

[krferriter]

I think this is one of the crazier ones. It's just canola oil! It's the same as spreading that much canola oil on the surface, the spray is mainly convenient because it spreads it out evenly for you without you needing to contact the surface. But Pam gets to put "0g fat" and "For Fat Free Cooking" on the side of all their cans.

[bluGill]

That might even be realistic if you are spraying a baking sheet - since you cover the whole thing. But if you are cooking pancakes and spray the pan after each one you get a lot of carbs.

[username135]

Alt tip: dont use any fat on the pan when cooking pancakes. Gives the surface that restaurant quaility, smooth evenness.

[bluGill]

Depends on the pan and pancake. Restaurants use a fair amount of fat in their dough so fat on the griddle isn't needed, if you make your own batter (as opposed to store bought) you can control this and reduce the fat such that you need to add some to the pan. The pancakes will of course taste different. In my case I'm making sourdough on cast iron - I've never figured out the trick to make the first couple not stick (whatever I cooked the night before affects something)

[hilux]

Carbs?

[bluGill]

Arrgghhh... Once more I wish I could edit things a day latter to fix all my stupid mistakes. (I blame is on diagnosed dysgraphia).

Oil is not carbs of course. I guess I meant fat.

[pkaye]

The rounding rule is carbs <0.5g can be rounded down to 0 and calories <5 can be rounded down to 0. But I have a feeling even if they properly labeled it without rounding, people would eat the whole pack of tic tac anyway.

https://foodlabelmaker.com/regulatory-hub/fda/rounding-rules...

[foxyv]

So you can just make the serving size so small that everything is less than 0.5g

[tart-lemonade]

The margins the FDA allows for class 2 and third group nutrients are also quite generous. I'm sure they made sense back when they were first introduced, but as food science has improved, the standards have not.

> The Food and Drug Administration (FDA) allows calorie content to exceed label calories by up to 20%

https://pmc.ncbi.nlm.nih.gov/articles/PMC3605747/

> Class I nutrients are those added in fortified or fabricated foods. These nutrients are vitamins, minerals, protein, dietary fiber, or potassium. Class I nutrients must be present at 100% or more of the value declared on the label

> Class II nutrients are vitamins, minerals, protein, total carbohydrate, dietary fiber, other carbohydrate, polyunsaturated and monounsaturated fat, or potassium that occur naturally in a food product. Class II nutrients must be present at 80% or more of the value declared on the label.

> The Third Group nutrients include calories, sugars, total fat, saturated fat, cholesterol, and sodium. [...] For foods with label declarations of Third Group nutrients, the ratio between the amount obtained by laboratory analysis and the amount declared on the product label in the Nutrition Facts panel must be 120% or less, i.e., the label is considered to be out of compliance if the nutrient content of a composite of the product is greater than 20% above the value declared on the label.

https://www.fda.gov/regulatory-information/search-fda-guidan...

Edit: Expanded the quotes to include definitions.

[yellow_lead]

Plus the whole sugar vs. sugar alcohol nonsense, which I still don't completely understand.

[freedomben]

IANA nutritionist or expert at all in this area so take this with a grain of salt, but my understanding from looking into it is that the sugar alcohol doesn't break down in digestion and isn't absorbed, that's why the "carbs" from sugar alcohol "don't count."

I would recommend taking it easy on the sugar alcohols even though they "don't count" because they can cause significant gas ;-)

[foxyv]

It depends on the type. Some are partially digested as carbohydrates. Others are not. You can look up their individual Glycemin Index.

https://cdn.shopify.com/s/files/1/0261/4761/8864/files/Scree...

[rini17]

And some like xylitol are highly toxic to dogs, must be careful around them.

[freedomben]

Indeed. My dog loves the dissolve-on-your-tongue melatonin, but it is super deadly to her because of the Xylitol. I keep it on a high shelf now and am very careful to pick up any pieces that might get dropped. She made it through, but that was a terrifying few days D-:

[MisterTea]

I also hate the games they play with labeling such as "no sugar added." Bought a cherry pie from a local market labeled "no sugar added" thinking it was going to be extra tart only to take it home and taste sugar. Reading the label it listed sugar alcohol which I learned can cause gas and bloating in some people, which I soon found out. Once slice had me doubled over with gas pains a whole night and shit my brains out the next day. I got my money back for that piece of garbage pie. I want to punch whoever thought no sugar added means fuck all...

[Dylan16807]

Sugar alcohols seem simple enough to me. The properties vary but they generally have fewer calories per gram and a low glycemic index, and some of them are much sweeter per calorie.

[interludead]

And I think there's also a psychological angle here, for instance when people see that something claims to be zero sugar or low carb, it can trigger a sense of relief or permission to indulge.

[madrasman]

I worked at Noom and the most successful users weighed the ingredients. Many users were frustrated about not losing weight even with calorie deficit, but the issue was they weren't logging accurately or skipped logging snacks and meals that were clearly not good.

[je42]

What works well is to measure the same food multiple times in a container (every time same volume ).

after a while you can estimate carbs by visually inspecting the contents without scale.

Add to that automatic bolus by a semi closed loop system to correct for errors, you can achieve good results with minimal effort.

[yapyap]

I’m rather unknowledgeable on diabetes, so here’s a question that may seem basic:

does choosing healthier meal, a salad instead of sweet ribs, not suffice for a good blood sugar?

[skyyler]

So, first off, commercial salad dressing almost always has sugar in it. Look at the nutritional facts label next time you're shopping for it. There's a few brands that offer "simple vinegar and oil" style dressings that don't have any sugar in them, but MOST salad dressings Americans come in contact with are full of sugar.

Even low GI foods still cause blood sugar to raise by some amount.

All of the vegetables in the salad have carbohydrates that will raise blood sugar. Carrots, onions, tomatoes, all of that will raise blood sugar. Croutons? Blood sugar.

Obviously selecting a garden salad with no dressing is a healthier choice than "sweet ribs". Most diabetics (that are managing their condition) are not going to be ordering things with refined sugar in them.

Where things get tricky is asking questions like "what's healthier, a honey-miso glazed salmon with brown rice or a salad with croutons and a honey and berry dressing?" or "What's better for you, grilled chicken with a sugary barbeque sauce or fried chicken with no sauce?"

[ryan-richt]

Also watch out for "sugar by another name" ... pineapple puree, white grape juice/concentrate, apple juice/concentrate are very common commercial dressing ingredients to load up on sugar.

Sure always ask for the vinaigrette eating out, but at home make your own salad dressing:

* get a mixing bowl big enough to toss salad in, and a whisk * add 1T dijon mustard, 1T not-balsamic vinegar (balsamic is high sugar! I like sherry or beer vinegar), salt & pepper * drizzle in 1T olive oil while rapidly whisking. * Add 3 oz or more salad, toss, done for 2 servings

[Scoundreller]

I just use old dijon bottles with a bit left in it to get the rest out and shake the heck out of it but I go through a lot of mustard.

The lengths I’ve seen brands go To avoid having sugar as their 1st or 2nd ingredient…

after that you have invert sugar, corn syrup, molasses, brown rice syrup etc. as following ingredients…

[sgerenser]

A few years ago organic/natural products were marketed as containing "Evaporated Cane Juice" (aka Cane Sugar) but my understanding is the FDA put an end to that one.

[reubenswartz]

Making your own salad dressing is really easy and let's you have a salad you really like in a couple of minutes.

My recipe is basically what you have here, although I usually mix some balsamic and other vinegars, and add a bit lemon juice.

I went from feeling sorry for people who were "forced" to eat salads to craving them. (Side benefit of not having the afternoon urge to sleep.)

[barbazoo]

> commercial salad dressing almost always has sugar in it. Look at the nutritional facts label next time you're shopping for it. There's a few brands that offer "simple vinegar and oil" style dressings that don't have any sugar in them, but MOST salad dressings Americans come in contact with are full of sugar.

Making salad dressing is really easy btw in case anyone wants to try. Often all you need is olive oil, vinegar, salt and pepper and you're set for most salads. Even a restaurant should be able to whip that up.

[skyyler]

If you have an immersion blender, making mayonnaise without sugar in it is very easy:

https://www.seriouseats.com/two-minute-mayonnaise

(And it tastes way better than commercial mayo!)

I love this author's recipes; it's the opposite of the normal recipe-preamble-slop. All of the stuff before the actual recipe is relevant information. In more complex recipes, he goes over the testing and process that led to the finished recipe. It's a wonderful view into the world of recipe creation.

[barbazoo]

Awesome, I'll give that a try. What I like about it is that you can use whatever high quality eggs you normally use instead of the cage eggs that mass producers will use. Until now I had to resort to vegan mayo.

[glandium]

"If you have an immersion blender"

You can also make mayonnaise with a whisk.

[nottorp]

> Often all you need is olive oil, vinegar, salt and pepper and you're set for most salads.

Why do you need a "dressing"? In my corner of Europe they put the above by default on every restaurant table and the salad has nothing in it (or maybe a tiny bit of oil and vinegar), you adjust it to taste.

The only places that offer salad "dressings" are american inspired and even those mostly serve it separately so you can ignore it.

[BobaFloutist]

For the same reason you add some spices before cooking, and salt multiple times throughout a recipe.

Plus, it's a little hard to emulsify or even suspend the oil and vinegar right there at the table.

[barbazoo]

> or maybe a tiny bit of oil and vinegar

That's what I mean by "dressing". We're talking about the same thing.

[stretchwithme]

I buy a salad kit at Trader Joe's. It has sugar in it. And I buy arugula and make 4 salads out of that one salad. I add a dash of olive oil and pecans. And end up throwing out 1/3 of dressing that came with the salad.

So I get some of the sugar sources in the kit. Just smaller amounts.

Otherwise, I just use olive oil and balsamic vinegar with arugula, pecans.

Arugula is a good source of nitrates, which are good for nitric acid.

[steveBK123]

Not a diabetic but adult later onset lactose intolerant and the problem is you really have NO idea what restaurants put into stuff, even if you ask.

Even a stupid salad, what's in the dressing, what's in the bread/croutons, what was the meat glazed with. Etc.

Restaurant food tastes good because it is generally unhealthy top to bottom, with quantities of salt, butter, etc no sane person would use at home.

One thought experiment - when was the last time you ate out and needed to add salt to anything? Now thing of home cooking how often you might add a little salt while you are eating.

The easiest thing to do is ruling out restaurants entirely, but then that's rather anti-social.. Not to mention family/friends gatherings, etc.

[UomoNeroNero]

Things are changing nonetheless. My wife is celiac (we’re quite a problematic family: I’m diabetic, she’s celiac), but by law, she is guaranteed that a suitable menu must be available wherever she goes, or at least that waitstaff and business owners know how to handle the situation when she informs them. (I know for a fact that managing celiac disease and the most severe and dangerous intolerances is a mandatory requirement for obtaining a business license.)

[steveBK123]

I think in the US, it's basically an intractable problem the way restaurants operate and are staffed. Low margin, high failure rate businesses with many fly by night small operators. Front of house staff is high turnover, while back of house staff is largely non-English speaking of sometimes questionable immigration/work permit status.

And then there is the supply chain since most restaurants are not cooking every single part of every meal from absolute scratch ingredients.

There was a story about a woman near us operating some sort of celiac friendly/gluten free bakery. One day the donuts were delivered and she noticed some D shaped sprinkles and realized her supplier had come up short and just put some random Dunkin Donuts into the delivery. Good on her catching it, but how in good conscience could she operate a bakery advertising itself as celiac friendly/gluten free if she was outsourcing like this?

If I had any sort of food allergy that could result in hospitalization or death, I'd just stop eating out. I'd rather be a little boring than very dead.

[throwway120385]

Why would you run a bakery and not make donuts yourself? They're dead simple if you have a fryer.

[pixl97]

>does choosing healthier meal,

For a type 1 diabetic, no (gets more complex with type 2).

Your body produces insulin at a basal rate.

https://en.wikipedia.org/wiki/Basal_rate

If you're healthy between the pancreas and the liver you maintain homeostasis and things are fine.

As a T1D you don't get that base rate, so your blood sugar will mostly trend up and stay high, even without eating anything. You simply have to get more insulin to avoid burning out all the systems in your body and dying slowly.

[UomoNeroNero]

Is a really complex game. The basic reasoning is that for every X carbohydrates ingested, you need to inject Y insulin (according to a personal ratio).

However, everything is complicated by numerous factors and the technology you use.

Factors: how you feel, stress, exercise, what you ate in previous meals, your blood sugar level at the start of the meal, and the activities you’ll engage in after the meal (physical or mental).

There’s also the issue of how you administer insulin.

In Italy, up until 3-5 years ago, most of us were using the “multiple daily injections” method, which involved taking a dose of “long-acting” insulin (lasting 24 hours) as a “base” and using “rapid or ultra-rapid” insulin at meals. Clearly, this approach provides limited control and requires a VERY habitual lifestyle (you can’t skip a meal; the long-acting insulin keeps working regardless).

Now (at least here in Italy), we are all transitioning to or already using CGM systems, which are more or less intelligent systems that continuously administer insulin at a “medium” rate. Based on input from the patient regarding the predicted amount of carbohydrates (and fats) they will consume, the system calculates the best strategy for what is called the “meal bolus” (using strategies like multi-phase, direct, etc.) and at the same time, it maintains a continuous but adaptable level of injection to achieve a target blood sugar level (day and NIGHT!!)

In essence, it’s a very nerdy way of dying slowly (hopefully as slowly as possible).

È un mondo difficile E vita intensa Felicità a momenti E futuro incerto

[aziaziazi]

> È un mondo difficile E vita intensa Felicità a momenti E futuro incerto

I was curious, for the other curious:

"It's a tough world And intense life Happiness in moments And uncertain future"

[UomoNeroNero]

Diabetes is a complex and mentally demanding disease. It affects you in the short term and has a significant impact in the long term. Everything is in your hands, fully aware that every mistake has immediate consequences but, worse, accumulates over time. That phrase (from this song) perfectly capture the mental state of my 20 years living with the disease. No tragedy (there are worse things), just deep awareness.

https://m.youtube.com/watch?v=cu3K1njbYqs

[Yizahi]

Salad is great for diabetics. The problem is everything else:). Like for instance I've discovered that 99% of all rice is extremely bad, even good pasta is bad, potatoes are poison, bread also bad, and the list goes on. Fruits are bad too.

Cooking at home can be managed, and still heavily limiting. Eating out is a nightmare. First of all there are no "diabetes" places in the similar style to "vegan". And eating in a restaurant with at least some diabetes friendly selection of dishes is hard. For example there are may be 4 soup dishes. But 3 of them or even all will have either potatoes or pasta as ingredient (and leaving out said ingredient makes for a very mall meal, because those are often added to compensate). Salad section - the same issue, too often they have sweet syrup added for flavor. Anything Asian has rice or noodles in large quantities (I often wonder what diabetics in Asia eat). Second course dishes like meat or fish also sometimes contain sweet "surprises". All in all it is very hard task to find something, in a big city even.

[skyyler]

> First of all there are no "diabetes" places in the similar style to "vegan".

AOL keyword: “keto”

low carbohydrate foods are currently trendy but they’re not called “diabetic”

[ksenzee]

This depends on a lot of factors. There are some type II diabetics like this: they might need insulin after a meal with a high glycemic index, but not after a meal with a low glycemic index. There are some type II diabetics with more advanced disease who need insulin after eating anything. Type I diabetics entirely lose their ability to make insulin, which is why the disease was fatal before insulin was discovered, no matter what the kids (it was almost always kids) ate or didn’t eat. As a general rule, it is inaccurate to equate diabetes with unhealthy eating. The Venn diagram only overlaps.

[joseda-hg]

Healthier isn't a good metric, A carb heavy salad will probably be worse than those protein heavy ribs by themselves (Maybe the rib sauce will tip you over, or maybe you will use a salad dressing that put any "healthiness" to the test)

[barbazoo]

That sounds really hard. Is the purpose of determining the amount of food so you can adjust the amount of insulin? Sorry, I don't know about the day to day of living with diabetes.

[lukeschlather]

Yes, diabetics need to precisely adjust their insulin intake in proportion to carbohydrate intake.

[Jarmsy]

Also when figuring out how much insulin is needed for a given amount of carbs you need to factor in the type of carbs, your individual response to that type of carbs, what fats/protein/fiber you eat with it (fats and fiber tend to slow down the BG rise from carbs, protein can cause a rise when eaten on its own but can also slow down the rise from carbs), what time of day it is (I need around double the amount of insulin for the same food first thing in the morning vs in the afternoon), your mood, what else is in your stomach already, the weather (hot weather can greatly increase insulin sensitivity), your current fitness level, what physical activity you have done over the last day or 2 and what you will do over the coming hours, where on your body you inject, if you are fighting any illness…

[not2b]

For my wife (type 1 diabetic), physical activity is the big one that throws off her calculations as a walk in a hilly area makes her blood sugar drop like a rock. Of course she always has something with sugar with her but then she has to figure out how much to consume.

[Jarmsy]

Hill walks are particularly challenging for me too. I can do rowing or weightlifting with my sugars staying fairly stable or rising if it's really intense, but something about walking steadily up a gentle slope makes it drop massively. There was some interesting research a couple of years back on how exercising the calf muscle is particularly effective at lowering BG, perhaps that has something to do with it https://pmc.ncbi.nlm.nih.gov/articles/PMC9404652/

[pixl97]

Yea, as a T1D myself the amount of insulin I need is massively different from days I'm arguing on the internet compared to days I'm up doing physical exercise. Things get concerning really quick when you're a distance from anything and your glucose starts dropping.

[pixl97]

When you eat, depending on the glycemic index of the food you're consuming your blood sugar starts immediately going up and can quickly peak at dangerous numbers if the food is sugary.

A diabetic will want to take dose of insulin a bit before eating in order to send their blood glucose level on a lowering trend. If you dose it right the two waves semi cancel each other out and your blood sugar goes up some, but hopefully not a huge amount.

If you get the dose wrong, it drops dangerously or rockets up and you have take correction doses.

[m_ke]

At Bitesnap we were surprised at how much interest there was from researchers to use our app for diet tracking. It turns out giving people a piece of paper to write “grilled cheese sandwich for lunch” is not a scalable and reliable way to collect research quality data.

We even worked with USDA on putting together a food logging dataset: https://agdatacommons.nal.usda.gov/articles/dataset/SNAPMe_A...

[wnorris510]

We've also been surprised at SnapCalorie how many researchers have approached us to use the app for more accurate diet tracking for medical study participants. The LiDAR based portion size has been a huge draw for them.

If anyone wants to check out our app or research its on our site: https://www.snapcalorie.com/

PS: Bitesnap was an awesome app!

[nitwit005]

Feels kind of incredible that something as advanced as laser imaging is being used to measure sandwich size.

[felideon]

What happened to your app? I was on such a research team (Scripps) that used your app for the study (PROGRESS).

[m_ke]

Unfortunately it was shut down after I sold the company to MyFitnessPal.

I was a shitty business person who thought it made sense to try and build a free consumer product on a bootstrapped budget. We had some traction on the B2B side that paid the bills but COVID took a dent in it and it would have taken a long time to build back the revenue stream selling to healthcare companies (tip for others, it can take 6-18months to close healthcare deals and another 6-18months to integrate)

We had a few offers to sell the company and took the one that seemed to make the most sense.

If there’s anything I can do to help out my email is michalwols at the Google email provider domain

[felideon]

The study ended so no worries. In any case, congrats on the exit!

[mtntreks]

This doesn't surprise me.

Just trying to keep track of calories for myself stupid things like supersized slices of bread becoming common in stores can really throw off my expected calorie counts.

It seems like this can completely throw off any attempt at figuring out nutrition from an app or research perspective.

[marinmania]

I highly recommend people get a food scale/measuring cups and weighing everything single thing they eat (even small things like nuts and cooking oil) for at least two weeks. After that I think you have a much better appreciation for how many calories your regular meals and snacks have.

[dailykoder]

I counted calories and put everything on a scale, for about 2 or 3 months in 2022 (iirc). And you are 100% right. I had absolutely no idea how much calories some food has. There were a lot of things, but I think cashews were my biggest eye opener (probably obvious to a lot of people). I easily achieved my goal of -10kg and saved A LOT of money, because I always had food prepared. And since I was going for a calorie deficit, I easily could afford a few sweets on the weekend.

Then I obviously got lazy. And while I sometimes still think I can estimate how much I am eating, I am probably wrong, because my bathroom scale says something different. My key takeaway is that it takes quite a bit of effort, but once you got into a routine, it's not hard.

Edit: Also, while I might have tried to ditch "wasted calories", I didn't put too much effort in eating healthy. One step at a time.

[dutchbookmaker]

Nuts are so deceptive.

I love almonds and because I love them so much, if I don't portion them out then I end up eating 2-3X the calories that I thought I did so easily.

I pretty much eat 1k calories per day during the week and then 2-3k on the weekend.

It took constant lowering of calories over a year to get to this point but being hungry just feels normal and natural for me. There must be something with fat adaption going on as I never feel low on energy. Eating during the week has been completely divorced from a pleasurable activity to me too. Chicken breast, nuts, vegetables, not much else during the week. It is just easy at this point for me to wait to eat for pleasure on the weekend.

[Fire-Dragon-DoL]

Might be a good idea, but I'm constantly concerned when I eat chicken breast every meal of the week. I'd rather use legumes, but chicken breast is so low on calories...

[mattlondon]

+1

The killer for me was breakfast cereals. The box shows a full bowl of whatever, full to the brim etc. in reality the pictures are probably 5 or 6 or more servings - a single serving would barely even cover the base of the bowl and even then be 200ish calories before milk.

If you just pour yourself "a bowl" of cereal without thinking or weighing then you're probably having 1200+ calories (or about 50% of your entire daily quota) even before you add milk or anything else, just for breakfast.

[aidenn0]

I don't know if they still do this, but I remember Special K cereal had identical calories listed for their various varieties, despite obvious differences in the ingredients; they just changed the portion size for each variety.

[ksenzee]

Cereal bowl sizes vary wildly, to make it even more confusing. Mine hold two cups (~450ml). Some hold way more than that. Some hold less. Buy a new set of bowls and you might be affecting your entire household’s eating habits.

[WheatMillington]

You don't have to fill a bowl to the top with whatever you're eating lol

[ksenzee]

Well, no, but people do. The whole point of this thread is "people are bad at estimating quantity." If it looks the same size as it used to be, but in reality it's half again as big, that's going to have an effect.

[zahlman]

>people are on average off by 53% and even trained professionals are still off by 40%. Basically if you want to have a higher level of accuracy you need to use a food scale or something that measure the volume of food, people just can't estimate portion sizes visually.

I can typically estimate them accurately without direct measurement, and with feedback that will tend to make errors cancel out over time. My trick is to note package weights, and divide containers into N equal portions. That is: I decide a target portion size first, and then portion it out.

If the task is "measure out an ounce of butter" I realistically won't be 40% off - because I can very accurately divide a rectangular solid in half repeatedly, and the butter comes in a one-pound package. Similarly, I have a pretty good idea how much grilled chicken is on my plate, because I know how much raw chicken I cooked, because I made a whole piece from a pack of N roughly-equal pieces weighing X (values which I noted when I bought it).

[wnorris510]

Yeah, dividing out a known portion size is a good hack that will probably help with accuracy. In our research most people's calories and error came from eating out where they didn't have these hints, but this is a good trick if you mostly cook for yourself!

[ericjmorey]

I started eating half of whatever was served as an individual portion whenever I was at a restaurant and not home cooking. It's the thing that tipped the scales for me when having difficulty losing weight.

[zahlman]

Historically I would rely on the restaurant's printed nutrition info. But I don't really eat out often enough for this to matter.

[tombert]

It can still be useful just to get rough estimates of what you're making at home, especially for portions and products that are roughly comparable.

If I make an egg, cheese, and sausage sandwich in the morning, and forget to weigh out or count how much of something I used, it can still be useful for back-of-napkin estimates if I Google the McDonalds Sausage McMuffin with Egg.

Obviously it's not going to be exactly equivalent, but I usually assume my homemade thing is 20% more than the restaurant to compensate.

It's of course better if you just weigh everything out first, you can get much more accurate measurements and calorie estimates then, but this can work in a pinch.

[ryan-richt]

There is actually an elegant _mathematical_ solution to this problem using sensor fusion and a differential equation model of the science: if you weigh your food almost all of the time at home, and only make portion and ingredient guesses when infrequently eating out, we can actually estimate your personal rate of underestimation and correct for it.

Our startup (BODYSIM.com) has also been doing research on this a long time. As founders, we all have >16 months of daily food logging mostly by kitchen scale weights, aligned to daily BIA-scale weigh-ins, fitness tracker calories, bi-weekly blood tests, monthly DEXAs, 3D scans, etc etc. We also have a science-based structural model of macronutrient balance and muscle hypertrophy. Given all that, we can VERY confidently estimate your TDEE (total daily energy expenditure) and its components, and predict how your fat and muscle mass will change on a daily basis. This is real math/science so you can also run it in reverse. This ("simultaneity constraint") provides enough constraints we can estimate users' individual underestimating/over-indulgence when eating out. In fact, it's better to just NOT log those days AT ALL and we can fill them in. I think this solution isn't more widely used b/c you need all this other "quantified self" type data at the same time.

[nathancahill]

This is really interesting, and I'll probably sign up for your app (I'm training for rock climbing). I've used a kitchen scale for a few weeks at home and got pretty good at estimating portion size during that time. Biggest takeaway was that even if you aren't "over-indulging" when you eat out, the portion sizes (especially in the US, less so in Europe) are just insane. 2-3x portions. Ordering half-orders or starters and letting the food settle before eating/ordering more helped quite a bit.

[ryan-richt]

we'd love to have you! LMK any questions over email or whatnot.

[Full_Clark]

How reliable do you find the calories-burned data from fitness trackers to be? Are there any brands that have higher accuracy than others? Are there any hardware features like pulse monitoring that improve the accuracy?

I find that the raw step count varies up to 66% between my phone and my wrist-worn tracker and I can't close that gap just by making sure my phone is never left behind.

[ryan-richt]

Not great! We have to "correct" them mathematically. Some can be crazy off until you calibrate them, for instance my husband's then-new Apple Watch was 2X off for months until he transitioned from winter indoor treadmill running to an outdoor run where it can calibrate. But even Garmin/Whoop/Oura can be substantially off, 50% easy. A "fun" simple way you can test yourself: do the exact same weight lifting workout but at + or - 10ºF between the two replicates. Guarantee you'll get dramatically different "calories burned" when in reality they are more similar. Your heart beating faster to cool you off in the hotter environment even though you're doing the same amount of (physics definition) "work" is NOT burning 50% more calories. Heart Rate is just a proxy for "work" and is confounded with other causes of heart rate change, such as ambient temperature/cooling.

[taeric]

This goes a long way to further convince me that it is portion sizes in the US. Having traveled, it is quite absurd to see the difference in standard order sizes.

Even for zero calorie things like water and unsweetened teas/coffees. You just get smaller cups. I'm not even sure you can get a US large in Japan for most drinks?

[stevesimmons]

Portion sizes in the US are ridiculous... often 2-3x larger than here in Europe.

When I regularly visited New York for work, and we'd get takeaway sandwiches, I'd have to open them and remove half the filling. I just couldn't physically eat that volume of meat, cheese or especially mayonnaise. For all drinks, I'd order small.

[bityard]

Where in Europe? I haven't toured the _whole_ continent but I've been to restaurants in Germany, the UK, and Ireland and did not find their portions to be any different than what you'd get at the average corner restaurant in the US.

Now, there are plenty of food vendors and restaurants in the US where big portions are considered part of the experience. Especially hamburgers, subs, and other sandwiches. I once ate at a place that served a plate-sized burrito completely covered in french fries. 12 inches wide and 6 inches tall. SOME people can eat that amount of food but most people cannot, and nobody is expected to.

Finally, large portions in NY street food are often customary because for lots of people with demanding jobs and 12-16 hour shifts, lunch is often their only meal. Or, half of it is lunch, the other half is dinner later on.

[lotsoweiners]

New York deli sandwiches are certainly not representative of what you get everywhere in the US. They are famously large.

[dfxm12]

And expensive to match the size! A pastrami sandwich at Katz's is ~$30. A croque madame at a similar place in Paris is ~€15.

People generally split a Pastrami sandwich over a couple meals or with someone else.

[taeric]

They may be famously large, but I don't think they are abnormally large for most of the US nowadays? I certainly didn't think they were particularly big when I visited.

[jacobr1]

Katz serves roughly 3/4 lbs of meat. That is particularly big. You can get triple hamburgers which would be similar is size - but most people are ordering singles or doubles. And you can find other kinds of large sandwiches around the country ... but it is not the most common of sizes.

[taeric]

I think what made them not seem excessively large to me, is that it didn't really come with much else? Yes, it was more meat than I would get on a sandwich, typically. But... that is about it?

Maybe I got too used to some of the obscure burrito places around Atlanta that would put way too much on them?

[vladvasiliu]

Probably depends on where you go.

I don't know about "the US", but as a "European" I thought serving sizes were comparable to what I get in restaurants at home. Drinks were an exception, since basically all restaurants had unlimited soda for next to nothing. This was actually great, since I was riding a motorbike in the desert in July.

For reference, I live in France and visited LA and random towns in the western states.

[vidarh]

It absolutely varies a lot within Europe too, but my feeling at least is that the difference between European and US portion sizes gets bigger as you move towards low-end places. High-end restaurants are pretty similar in portion sizes almost everywhere I've been, presumably because they're not competing on portion sizes, while lower-end places are much more susceptible to local expectations of what is good value.

[ellisv]

I'm a big fan of European serving sizes compared to U.S. for food – but when it comes to beverages, particularly water, I can't believe how much they charge you for how little they give. I understand everything comes in bottles with VAT but even asking for tap water I found they'd only bring a very small glass.

[jeltz]

In some European countries water is free. I am from Sweden where all places have free tap water and fancy places often have free sparkling water.

[kjkjadksj]

Along the mediterranian seemed like the only place to get free water were the ancient fountains that spittle out a stream. But then you’d have to wait for the inevitable old man to finish washing his head and arm pits in that fountain. Beer was usually substantially cheaper than the water offerings.

[bigbacaloa]

In Spain by law all eating establishments have to provide tap water for free if asked.

[taeric]

I'll be visiting France soon, so will be able to compare on that front. But I think it is an understatement to say that things are universally smaller.

And on the drinks, even places in Japan that had free refills still gave, at largest, an 8oz cup. Usually, I think they were even smaller. Even getting popcorn at Universal, the bags were large, but nothing compared to what I'd expect over here.

Some of this, I'm sure, is having gotten used to ordering the larges. For a time, it was not unheard of to get a 32oz soda at any given convenience store. May still be normal? I don't know.

(And, of course, this isn't getting in to the sizes of vehicles.)

[zeroonetwothree]

I don’t see why it would be bad to get more water to drink

[zahlman]

There's actually something of a stereotype that Japanese places will give you unreasonably small portions of water with meals. (Dogen plays off this in some of his videos.)

But then, I think it's only been Americans I heard this from, so.

[taeric]

Ha! I hadn't heard of this before, so it caught me completely off guard.

The coffee was the one that really surprised me. Order a coffee and get a 6-8oz cup. With nothing on the menu to indicate you can get a 12-16oz. Was surprising. (Not bad, mind. Just surprising.)

[dfxm12]

In both my trips to Japan (one recent, one 20yrs ago), I never noticed this, and I think I drink a lot of water in general, and especially as a tourist because I'm doing much more walking.

[kjkjadksj]

The free water cup in a lot of places in the US is like a 6oz slosh now.

[taeric]

I didn't mean to imply it was. My point was that everything is smaller.

[WheatMillington]

>I'm not even sure you can get a US large in Japan for most drinks?

I've seen what large US drinks look like, and you definitely can't get that here in New Zealand. Like a litre of soft drink at a fast food place, it's absurd.

[LPisGood]

>and even trained professionals are still off by 40%

I find this very hard to believe, unless the term “trained professional” is quite broad. When I was much more into fitness and weighed every meal to the gram, I could tell if a bowl of cereal was a serving to within a gram or two.

[wnorris510]

If you have a known bowl and fill it to a known position every day with the same type of food, then you can probably do better than the average for that specific meal. In our research we've found a majority of calories for most people come from when they're eating out and consuming new dishes where they don't know the ingredients or portion sizes.

In the study we gave people a variety of dishes to make their estimate on, some they were familiar with, some they were not.

The professionals were nutritionists who had trained in portion size estimation and were shown 2D images on a computer screen.

For what it's worth, we've had a lot of people who have claimed to be very accurate at portion size estimation from a long history of using a kitchen scale. We've paid many of them to do a quiz to see if they're above average accuracy and they have almost always ended up around 40% accuracy or worse.

[shawabawa3]

estimating from a photograph is always going to have huge error because you just cannot know e.g. the size of the plate without some external reference

[wnorris510]

Modern phones have LiDAR that give you the near exact volume and dimensions of everything on a plate. That's what we do:

https://www.snapcalorie.com/

We co-authored a paper on this with Google AI and showed it got about 2x the accuracy of a nutritionist because of it.

[maeil]

I'd love to do such a quiz -I might even be willing to pay for the privilege! I'm quite convinced I'm really accurate at calorie estimation without using a scale but would love to be proven wrong. Zero food industry experience here, just from reading hundreds of food labels per year since very young, maybe 8 years old.

Thinking about it again, I'll probably do a lot worse from a picture because I can't have a bite of the food! Just having a spoon makes it so much easier in terms of ratios.

[watwut]

There is no profession that would require you to estimate portion sizes up to grams visually. So, trained professional will be someone who was trained in something different - a doctor for example.

I guess, maybe cooks should have the best precision for this.

[zeroonetwothree]

If you read the paper it’s pretty easy to see what they mean by this. They tested “4 professional nutritionists”. I don’t know if nutritionists get any special training at estimating portion size but my guess would be they do not.

[wnorris510]

Some do, some do not. We put them through a standard portion size training course regardless to be sure.

[ses1984]

Isn’t that a bit of a special case because you know your cereal and you know your bowls? What about some cooked foods like meats which can vary in density and shape when raw, and also vary even further due to inconsistency in cooking, with more or less moisture cooked off?

It’s possible to calibrate your estimates, but if you haven’t done that, it’s probably safe to assume you’re not particularly accurate.

[wnorris510]

There is definitely a lot of variation in density, moisture content, fat percentage between regions, cuts, cooking amounts and methods. IMO using an average number here is probably best because to some extent it's hopeless to account for all of these things.

Most people don't stay consistent in tracking long enough for any of this to matter, so really it's about what is the most accurate approach to achieve your goal and sustain longer term.

[LPisGood]

Oh I would only weigh things raw - if we’re talking about guessing the portion sizes at a restaurant for example, you might say I’m cooked.

I wonder how good an ML model might be at that task. Maybe given a photograph of the plate and the menu description.

[acomms]

I think they're suggesting that the portions you are judging have not been practiced hundreds of times.

[mrgaro]

I downloaded SnapCalorie to try it out on Android. I went all the way through the sign-up phase, only to discovery that I would need to activate subscription in order to have the 7-day trial. Ended up uninstalling the app :(

[wnorris510]

We're an early stage startup and the models are expensive, we're trying to get the price as low as possible, but yes we need to charge to cover costs right now. Sorry about that!

You might get a yearly discount offer that is less than $2/month if you get lucky (A/B test split). But that's less than the cost of running the model for people, so hopefully others will consider paying full price.

[mrgaro]

Thanks for you quick answer. I want to clarify that I would have liked to try the app for a few days before activating subscription.

Now with the current flow I would need to activate the subscription and then immediately go to Play Store settings to deactivate the subscription so that I would not forget it.

[Dilettante_]

That's pretty standard for free trials in my experience. Amazon prime, audible, musescore, I'd be harder pressed to think of a service I've recently tried where it was not like that.

[wnorris510]

We don't discourage anyone from doing this! Free trial with access to all features stays live for the full 7 days even if you cancel immediately. Hope you enjoy it.

[tln]

I'm on an annual plan from another app (Calory, $30) otherwise I would have bit.

It gives some features with a totally free plan. That makes the IaP feel less like a bait and switch.

The proposition of SnapCalorie is compelling. Calory ui is decent and I use a scale so accuracy should be good but I think their database is shitty. Meatloaf will vary from 1.5 kcal/g to 3, steak will show as 1 kcal/g, stuff like that.

[wnorris510]

You can start a 7 day free trial and cancel immediately. There is also a freemium tier at the end if you don't end up converting. We don't normally advertise this because we've been struggling to handle the load and costs of new users, but hopefully as we scale up we'll be able to support a free tier for everyone!

[ploynog]

I think the complaint is that you only get told that you require a paid plan AFTER signing up. At least on a brief look on the Play Store page and your website, it does not immediately mention it prominently.

That seems like a very dark pattern and is, honestly, pretty scummy.

[wnorris510]

This is not a dark pattern, it's just a constraint that the app stores place on the pricing disclosure that is very non-intuitive. You have to mark your app as "free" to download if you charge a recurring subscription fee. You can only mark it as paid if there is a one time fee to download the app.

Our FAQ and pricing pages all list that it is a paid only app. All of our ads explain that it's subscription based. Anyone who asks we're very transparent about it. If there's somewhere else where you think we can list it to make it more clear I'm happy to add it, just not sure where that would be.

[felideon]

What FAQ and pricing pages? Your website makes no mention of pricing at all.

Edit: The "dark" pattern is in the registration flow. It doesn't mention that the app requires a subscription anywhere until after you've created an account. Surely you could add a disclaimer before creating your account? This has nothing to do with the App Store.

Edit 2: I'm not saying you intended to implement a dark pattern. Just perhaps a UX oversight.

Edit 3: The download page would be another great place to put this info, since that's the primary CTA on the home page (there's 4 prominent download buttons).

[wnorris510]

Looks like we used to have it in the description on the app store along with the FAQ but a team member made the decision to remove it because of complaints about it being inconsistent with the way Apple was localizing pricing to different currencies and regions.

We can't hit people with the paywall before they've registered because we need to assign the trial to their user record. We've tried adding more language during onboarding but no one reads any of it, they just click through.

You're mistaking challenges in building a global app for malicious intent. I left a job paying a lot more to do this because I wanted to help people.

We'll add something back to the FAQ on this, thank you all for pointing it out.

[amendegree]

I don't see any mention of the price in the FAQ[0], which I had to guess the url of because it doesn't have a link anywhere on the homepage. Trying to guess the url of the pricing page doesn't yield any results.

[0] https://www.snapcalorie.com/faq.html

[idiot900]

https://www.snapcalorie.com/

I cannot find a FAQ or pricing page on your website.

[DrillShopper]

It doesn't seem like - it is

Which makes it par for the course in the scam that is mobile development

[DrillShopper]

This is the problem with any fitness app.

They either need to show you ads, charge you for premium for services that used to be free making your free tier functionally useless (looking at you, MFP who gated barcode scanning behind their honestly ludicrously priced subscription), or sell your data, and they often do all three.

The entire industry is like this, and honestly an app that charges one time and fucks off would be ideal but given the amount you'd probably need to charge as a one off (or for major upgrades) most consumers would rather have the slow bleed of $10/mo than $25 one time.

[wnorris510]

You really don't want to pay a one-time fee, it incentives the developers to stop maintaining the app.

[Dylan16807]

I like the general idea of ongoing revenue, but I want to pay something on par with buying a full version every 3-5 years. Subscription software usually costs much more than that.

[bigstrat2003]

I would love for developers to stop messing with most apps.

[wnorris510]

Haha, you say this until Apple does a breaking change to the barcode library or Apple Health export and things stop working. Then you probably want them to change some stuff :)

[toast0]

It would be nice if OS vendors would stop breaking things, too.

[saulpw]

This would often be a feature.

[pixl97]

Except most app stores require future maintenance and compliance to keep publishing the app. Someone has to keep the lights on.

[ellisv]

Most don't require a subscription before a trial.

I'm paying for a fitness app subscription that annually is less than 1 month of gym membership. But I had a 7 day trial which got me hooked before I had to sign up for the subscription.

[mattlondon]

Thanks - just read this comment while it was downloading and installing, so uninstalled straight away.

Back to fitness pal and scanning barcodes (which is not really much of a hardship tbh)

[michchinn]

From the paper (https://openaccess.thecvf.com/content/CVPR2021/papers/Thames...):

> We asked them to estimate the mass of each ingredient present on the plate and subsequently converted these values into nutrition estimates using the same USDA [9] values we used to create our dataset

I get that there's a linear relationship between the mass of a food and its calories, but I'd expect that nutritionists would be better at estimating the calories in a plate of food than the mass of a food item. Most people aren't doing the math in their heads, they're using a frame of reference that recognizes calories. Did you have this in mind? Is there any research on this?

[wisty]

Not too shocking really:

https://pmc.ncbi.nlm.nih.gov/articles/PMC1322248/

> Aiming to pour a “shot” of alcohol (1.5 ounces, 44.3 ml), both students and bartenders poured more into short, wide glasses than into tall slender glasses (46.1 ml v 44.7 ml and 54.6 ml v 46.4 ml, respectively). Practice reduced the tendency to overpour, but not for short, wide glasses. Despite an average of six years of experience, bartenders poured 20.5% more into short, wide glasses than tall, slender ones; paying careful attention reduced but did not eliminate the effect.

A plate is a very wide 'glass'.

[tombert]

I'm currently dieting again, and the only way that I've been able to properly portion calories is to weigh nearly everything I eat and then add the numbers together in Google Sheets.

Eyeballing a portion of a lot of food can be nearly impossible to determine how much food you actually got, but weight is fairly straightforward and objective (at least to an ounce or so of granularity for most kitchen scales, which is good enough for dieting).

[xandrius]

Fats are definitely the easiest way to mess up counting, especially when a lot of people can't compute not cooking without oil/butter.

If the goal is losing weight, I found that maximising volume (e.g. Minimising calory density) works wonders for me. I am used to being full before feeling satiated (mostly upbringing I guess), so this is my trick.

[porphyra]

As someone who takes a photo of every single meal I eat, I was very excited to try out Snapcalorie but it was completely wrong for all the pictures I tried giving it. I uploaded a picture of a recent meal of tomato egg, baked octopus tentacles, and shrimp, and it identified it as pasta, mushrooms, and chicken. Also, it doesn't work for typical home meals that are eaten family-style.

[rapjr9]

I was part of a project that did some work in this area also, we developed a machine learning based wearable to detect chewing:

https://www.ah-lab.cs.dartmouth.edu/publications/detecting-e...

[Fire-Dragon-DoL]

Even when you weight, the weight needs to be on cooked food, otherwise it's useless. I'm not going to cook a separate meal for every member of a family

[darkhorse222]

This should be an indication that tracking as a personal health methodology is inherently flawed. Your body is your most accurate measurement system, both in terms of precision and accuracy but also in its multidimensional, intersectional measurement apparatus that completely demolishes the poor substitutes found in personal nutrition, which are continuously shown to be either flawed in theory or in practice.

Tracking takes more work and is less accurate. Bad trade.

The only use I see in tracking is to perhaps help one inform one's intuition. But that's as far as I'll go.

[sureglymop]

When I was tracking calories to lose weight I just always overestimated by default – doesn't hurt if one week you happen to lose a bit more weight than you set out.

[tejohnso]

> you need to use a food scale or something that measure the volume of food

Isn't that obvious? Basic high school science projects would have students using measuring devices. Are you saying that it's common for nutritional studies to tell people to eyeball their portions and that is then used as actual data?

I see from the article "Nutritional epidemiology studies typically ask people to keep a food diary or complete questionnaires about their intake over the past 24 hours, a week, or even several months." I find that hard to believe. How could any study like that be taken seriously? That's like having someone stand at a street corner for an hour and observe the population to then come up with an average BMI for the neighbourhood.

[pavel_lishin]

I would wager that just paying attention to, and thinking about what someone eats has a decent impact on their health - so it feels like it's working, and like your estimates are accurate.

After all - once you started doing it, you started losing weight/building muscle/achieving whatever result.

[ericjmorey]

Which is a reason why keeping a food diary is an often recommended technique for changing your diet and eating healthier.

[ellisv]

Yes, the mere act of monitoring (including self-monitoring) leads to behavioral change.

This alone can be sufficient for some people.

[jacobr1]

One factor is just the sheer volume of snacks and treats - outside of the portion size of any particular meal. If your were not self-aware of constant eating that can have a big impact - at least it did for a few friends of mine.

[disgruntledphd2]

Yup, I discovered this 14 years ago and wrote a proposal to do barcode scanning to help, but left academia soon after.