Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

How on Earth is ten days enough to test anything in regards to diet? Pretty sure humans don't notice much change in that time, let alone enough to measure? Day to day numbers fluctuate by a lot, so wouldn't it have to be tested over a longer period for it to be valid?


I was also alarmed by that paragraph.

    > Various doctors, dietitians, and food scientists 
    > have been reviewing our formula and providing 
    > suggestions as they see fit. We have been listening 
    > to them and testing these changes in our beta 
    > program. Each modification requires making a new
    > batch of Soylent by hand, shipping them to our
    > beta testers, and gathering their feedback, a
    > process that takes at least 10 days per revision. 
The idea that the dietary advice of a doctor could meaningfully be tested in a few weeks by a beta program of laypeople with no scientific controls is utterly ridiculous.

Even if you had a large beta program and ran each revisions for 100 days, the feedback would be garbage unless you had proper controls and measurements in place.


>The idea that the dietary advice of a doctor could meaningfully be tested in a few weeks by a beta program of laypeople with no scientific controls is utterly ridiculous.

But you are assuming the purpose of the test is to test the merit of the dietary motivator for the change, the testing described here might just be user acceptance tasting. That is, does this this change make it taste bad? Give you gas? etc


They are probably treating it like more of a smoke test. In 10 days you can certainly get a good idea about changes in flavor, texture, preparation, etc. You also ought to get some feedback on whether a formula change resulted in any immediate health concerns.


They aren't getting any feedback on immediate health concerns in just 10 days unless they decide to add peanuts or arsenic to the mix.


Keep in mind that this may be the only form of nutrition the 'beta testers' are getting, and possibly have been getting, for a significant amount of time.

If you're eating normal food then some deficiency in one component of one meal may be masked by a surplus in another, and in particular if you eat a varied diet then that deficiency could start and stop within one meal. If instead your sole source of nutrition is suddenly deficient, it's not a stretch to think that you could notice a difference within 10 days.

I'm not trying to defend the product, I just don't think it's inconceivable that 10 days is long enough to notice a big problem with a formulation change.


In general they don't seem so bothered about the scientific method.


Even the (not very scientific) test from the EU Food-Safety-Agency make 90-days trials on animals (and that does not tell a lot imho).

If you want to know, if you are allergic, a short test should be sufficient. If you wanna know, if allergies might be initiated from the food, 10 days don't do much good. If you want to know, if some known things (nutrition, et al) are missing from the "food" 10 days won't do you much good and might only be detected in really rigorous blood-tests.

If you want to check for problems with unknown unknowns (and there are a lot of things we yet do not know, regarding food and our bodies) 10 days are nothing more then a nice joke. (Sorry, but living with a biologist make me quite cynical, when it comes to things, that might just endanger us.)

A really scientific method is something else. They might stumble upon some things, but I'd like to know, if every beta-tester gets his blood-work done before and after every test. How does the testing process in the first place look like (blood tests as said? just taste? subjective feeling after n-days of new soylent version? and so on).

> so everyone will get their complete amino acid profile

No everyone will get a one size fits all amino acid profile. And then comes the problem, that humans aren't "one size fits all". They will probably get a solution, that fits one, two or probably max. three sigma of the targeted population.

To be fair, I love food, so soylent would not be for me. I really think, that a cheap food-replacement that does give most people mostly everything they need in terms of nutrition, might really do a lot of good, if distributed in times of disaster or in regions with problematic food distribution (third world). But that is not the targeted audience as I see it.

It seems to be people who do not value food, but have enough money on their hand, to replace it with this product.


(This is completely off-topic, but you only needed about 1/3 of the commas you used. For instance, "if you wanna know, ..." does not need a comma. Your final sentence does not need any commas. Etc.)


Here's a classic deficiency case:

"The onset of symptoms of scurvy depends on how long it takes for the person to use up their limited stores of vitamin C. For example, if the diet includes no vitamin C at all, the average onset of symptoms is about four weeks"

http://www.betterhealth.vic.gov.au/bhcv2/bhcarticles.nsf/pag...


I would assume they mean testing it for factors like taste, etc. which would be immediate rather than doing it for nutritional testing.


Why would you ask doctors or dietitians for advice about the taste of the preparation?


I assume he meant that the input from doctors etc is for nutritional reasons but when they ship out to testers they are taking the nutrition changes at face value and are only concerned about feedback from testers on whether it affected the taste/consistency/gives them gas etc.


They're at the point of making changes to "taste, texture, and smell" - the dietary side of the formula is nailed down already, and I am willing to believe was tested more rigorously.


> I am willing to believe was tested more rigorously.

Really? Their site launched with a FAQ on safety that essentially read "well, we haven't died yet..."


Why are you willing to believe that?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: