Hacker News Reader

Comment by autarch | Hacker News Reader

Soylent is definitely suspect. The idea that human nutrition can be boiled down to a minimal list of ingredients that are the only things you should eat is nonsense.

That said, there's no reason to think this article is any better. Honestly, I didn't finish it, because as soon as it said that you might crave animal protein heavy products because of a protein deficiency I knew that the author was as full of it as what he or she is criticizing.

The citations look good until you start digging. One of them cites a naturalnews.com article (not a peer-reviewed journal) that in turn cites the Weston A. Price Foundation site, a towering bastion of pseudoscientific nonsense. I wonder how many of actual scientific articles cited even support the points the author cites them for.

In summary, the nutrition science in this article is about as poor as the nutrition science behind Soylent.

View on HN

> The idea that human nutrition can be boiled down to a minimal list of ingredients that are the only things you should eat is nonsense.

It's perfectly sensible that there is a mathematically minimum sufficient diet, or mathematically optimal diet, with the caveat that:

1. There's a high probability that this varies for each definition of "you", and for different times of day / month / year / within a lifetime

2. Soylent is, with almost perfect certainty, not it. But there is a nonzero chance that a purely soylent diet is healthier than the current median human diet.

Will we get there someday? Probably. I"d say there's a better than 50% chance it happens within the next century. Whether we want to get there is another story. FWIW, our bodies aren't really a whole lot more complex than feline or canine bodies, and we've managed to come up with a sufficiently nutritious kitty kibble (or at least, we think we have).

Remember, deciding what to eat is a decidedly modern problem, as is looking at food as a combination of nutrients. For the vast majority of human existence, nutrition consisted of avoiding starvation. When deciding between options, the proper answer was "all of the above".

> Whether we want to get there is another story.

NASA could probably have come up with a Soylent-like product by now, but they haven't. I guess that being stuck in a small can in space means that nice food becomes important.

So, on the ground, there are two hurdles. i) Portion control, ii) Choice.

Some people will not eat two scoops of ice cream. They will eat a pint of ice cream. They won't eat 2 cookies, they'll eat the packet of cookies. Some people will smear the cookies with peanut butter, and crumble them over the pint of ice cream.

Will Soylent help these people? What happens if someone eats double the recommended daily amount of Soylent? Or triple? Or quadruple?

Some people will walk into a supermarket, walk past all the fruit and vegetables, and into the doughnut aisle. Then they'll walk past the meat and fish and into the crisp / chip aisle. Then they'll go to the weird salt / sugar / fat / slop instant meal aisle and buy bizarre "food".

I have no idea how Soylent will help them.

As far as I can figure out NASA's original solution to space food WAS a Soylent-like product. Astronauts responded by smuggling a corned beef sandwich into space. [1]

The point of Soylent is to separate the pleasure from eating. Rather than being a emotional, sensory activity, treat eating with has all the excitement of popping a pill - insert nutrients into body. What do I eat? Soylent. How much? The amount it tells me to. The questions of "Do I eat the broccoli or the ice cream?" or "Do I have seconds?" are taken off the table.

I wss in the hospital for the better part of a week a year ago. They had me on the "nothing by mouth" diet for a couple days - simply an IV drip for nutrition, then a liquid diet for two more. When my friends came to visit, the first thing I said was something along the lines of "As soon as I get out of here we're getting the best cheeseburger in this city." Could that desire be trained away? I'm not sure.

[1]: http://history.nasa.gov/Apollo204/zorn/grissom.htm

This is my exact outlook on it. You have a very fair chance of Soylent being significantly healthier than your current diet (as an American). At the very least, it provides the nutrients that a hundred years of studying dietary deficiencies have proved necessary and will possibly give some indication of necessary nutrients that we have overlooked.

It's pretty cool, actually. If you're entrepreneurial it's good to know that even hackers will fall for anything provided it's being marketed by someone who seems like a hacker, too.

"The idea that human nutrition can be boiled down to a minimal list of ingredients that are the only things you should eat is nonsense."

why?

http://www.priceplow.com/blog/wp-content/uploads/2013/07/pat...

I didn't realize that you could respond with image macros on HN.

Ok, so it's a complicated image. The first thing I thought is it looks like the schema of a simple arithmetic operation in a processor. Meaning that it looks complicated, but it doesn't tell us anything. I see dots and lines and pretty colors, but I don't see an argument.

> it looks complicated, but it doesn't tell us anything

That's because the vast majority of the detail has been omitted in order to make it look pretty. Here is a slightly more detailed version: http://primalmeded.files.wordpress.com/2011/05/metabolism.jp...

And another: http://3.bp.blogspot.com/-MJisxWVfZPA/UEF_-1krN-I/AAAAAAAAr9...

For more info, read Principles of Biochemistry: http://www.amazon.com/Lehninger-Principles-Biochemistry-Davi... (and if you aren't familiar with organic chemistry: http://www.amazon.com/Organic-Chemistry-1st-Edition-ebook/dp...)

This doesn't seem like a good argument. You could provide a diagram of a modern car that would be just as complex as a section of this image, but I don't think anyone would argue that it's inherently impossible to understand the necessary inputs to that.

You are asking a totally different question. It's not about understanding the inputs, it's the claim that the inputs can be super simple and can ensure your body still work as normal.

Because nothing is simple in life, our bodies are complex, and there's Wikipedia if you have tons of super broad questions like that.

Complexity is not, has never been, and will likely never be an insurmountable barrier to learning more about it. Let's take an individual example: The brain. It's very complex, has an extremely large number of variations from human to human and nuances of operation. And yet that hasn't stopped neurologists from trying to unlock, simulate and stimulate its every function.

There's no reason nutrition can never boiled down to the sum of the ingredients in some form (hopefully a form that tastes better and isn't liquid). And let's face it, Soylent (in a way) tries to basically be liquid Tofu.

That said, In that regard, I completely agree with autarch's take on the article content itself.

I also don't like it that the article started off with a preconceived bias: "Rhinehart, a 24 year old software engineer, believes he’s solved the befuddling food conundrum, and has named his product Soylent. It is intended to replace your entire diet — one food to solve it all." Emphasis where "software engineer" was underlined.

So what if he's a software engineer? It doesn't mean his profession precludes him from learning about nutrition.

We've made things fly - even all the way to space. We've transplanted organs. We've developed vaccines that eradicated diseases. We are making progress towards deciphering the genome. What once required a library the size of a city block can now fit in your pocket. You can have a conversation with someone on the opposite side of the planet. Or even someone who isn't on the planet.

Nothing is simple in life, and our bodies are complex. But if you EVER ask me the question: "Will people figure it out?" Unless the speed of light factors into the discussion, I'm putting my money on "yes" every single time.

I'm not saying anything is impossible. But all the solutions we find are usually complex and not easy. A phone is complex. A computer is complex. Transistors are complex. Organs transplant is complex.

So why do we expect the answer to food to be a simple bunch of ingredients mixed together?

A simple bunch of ingredients mixed together? Isn't that basically a recipe by definition. Or a diet?

Getting a bunch of ingredients and turning it into something nutritious just isn't particularly difficult. Finding such a combination that is both nutritious and palatable is the challenge. The interesting bit with soylent is not that it is a fantastically revolutionary recipe. One could probably concoct something out of rice, beans, and a brassica cultivar that would serve the purpose of being a complete human diet just fine.

Rather, it's people saying "I'm going to give up the pleasure I get from food in exchange for removing whatever distress the uncertainty in deciding what I'm going to eat causes me." It's a mental hack, not a culinary one. Are you're willing to make that trade? The culinary and scientific problems just aren't that difficult on the scale of culinary or scientific problems, but I don't think most people would be willing to do it.

> A simple bunch of ingredients mixed together? Isn't that basically a recipe by definition.

No, the usual recipe of a daily meal is way more complex than the composition of Soylent in terms of the variety of elements/proteins you get.

I'm not saying whether or not Soylent is the real deal.

But I don't understand why you think it's not possible to come up with an optimal combination of input (food) into our bodies that maximizes it's output (longevity?).

First, longevity is not the main output of the human body. At least there's no single output. What if Soylent is good for your brain but detrimental to certain organs? Or the opposite?

I have a biochemistry background and from what I know the digestive process is extremely complex and still certain parts are not fully understood. So I highly doubt a bunch of whiz kids can find the optimal formula for food without any scientific background, sufficient research and proper studies to support the claims they make.

Plus, it's not like there are no variations from one person to another in terms of what you need in terms of nutrition. We know genetic differences exist, in some parts of the world you have deficiencies of certain enzymes to break down some types of food, and so on.

By the way, if you want a simple diet, you can probably live on rice and bread alone (mostly). You don't need to stuff your body with some shit people have no experience with. Rice/Bread may not be a very good diet but it satisfies a number of nutritional requirements and we know for a fact people can live (maybe not very well) on such diets (because they did for a fairly long amount of time).

Don't forget that enjoying meals with other people and having direct connections to the earth and the people who produce your food is socially and emotionally healthy. The psychological damage to most people due to a strict Soylent diet would be significant, and psychological stress causes physical health problems.

I agree with you but that has nothing to do with the point being discussed here in this particular sub-thread.

So if someone is looking for optimal food inputs with respect to longevity, I think it's relevant to point out that how you eat may be almost as important as what you eat. That's all, really.

Kinda like why human flight is impossible?

First of all, trying to find just the right foods reeks of premature optimization.

Second we don't know what our requirements actually are. And lab rats aren't going to fix that... one of the arguments why there has been a rise in autoimmune disorders in this century is our obsession with cleanliness... we may not be getting enough dirt, microbes, etc to keep our immune system in check.

Third, nutritional science doesn't have unit testing. We don't have integration testing.

Nutritional science has bug reports from the field, we have minimal requirements and max quantities. We can compare two systems, but we can't debug them. It's like trying to reverse engineer COBOL from machine code, when you don't know either.

Further, we don't know exactly how such things interact with other factors, sleep, genetic and environmental factors, exercise. Nor do we necessarily know all the elements of your health effected by nutrition. Fertility, resilience to disease, mental state.

Soylent is an experiment. And if people taking it as their food source keep good records of what goes on with them on it, we may learn from it, by comparing people on other diets in the same environments.

You want better nutrition? Log your food, accurately, for years, along with your feeling of health, and environment and any other factors (weight, exercise, amount of sleep, etc.)

Publish the data regularly, and maybe someday we'll be able to analyze it and build a nutrition model.

However, I think it's premature to decide we know what we need. We are learning. We know some things. But we don't know enough.

Are any other citations suspect? Concerning the overall argument of the paper, that citation was in support of a very minor point concerning whether dietary cholesterol is necessary.

Obviously the main point of the paper was to explain our relative ignorance of nutrition as a whole. The sub points appear intended to show that Rhinehart is not even properly informed of the details that have evidence, let alone the massive unknown.

In summary, its unfair for you to summarize what you did not entirely read. Read it in its entirety and find more insufficient citations before you debunk everything.

Reply on news.ycombinator.com