Hacker News Reader

Comment by equoid | Hacker News Reader

Don't be silly. That measurement technique is to allow you to isolate and measure the effect. You seem to want to infer that the effect does not occur in any other circumstances.

That is as inane as disregarding crash test dummy results because they take place in a controlled environment which is not the same as 'real life'.

View on HN

Actually, you can eat the same "high GI food" like white bread with butter and its GI goes DOWN. This is because eating fat with your high GI foods decreases its GI value! So really, GI is a useless measure for this and many other reasons.

It's not silly, at least not on the face of it to a complete duffer in nutritional science (me). Intuitively it seems like nutrition is a highly interdependent system of variables. The effect of a food on your body and blood chemistry might be completely different when consumed in combination with other foods.

Or it might not. Testable! Give subjects the same combinations of foods (maybe scaled by body weight) a set time before consuming the test food, and measure blood sugar delta.

In any case, it might be wrong but it's not silly.

I stand by my verdict. He was stating that ring fenced measurements are blown out of proportion because they do not reflect real life behaviour. I'd expect a citation otherwise I am happy to dismiss it as silly.

Well, in effect what you're arguing is that because something was tested in isolation, you feel justified in generalizing it to a situation it was not tested in.

Absent further evidence, that seems like a safe conclusion given the caveat that it hasn't actually been tested.

Hold on a second. It's not that there is no evidence, I just personally haven't bothered to crawl the internet looking for it. A quick google search found the following: "High- and moderate-GI choices may be improved by pairing them with low-GI foods or healthy sources of lean protein or unsaturated fats. For example, the GI of white bread will be lowered if it is eaten with a bowl of chili or spread with peanut butter." I.e. eat your white rice with some chicken and olive oil and you'll be "okay".

http://clinical.diabetesjournals.org/content/29/4/155.full

Diet is complicated. Its not at all obvious you can 'control' for an effect in a complex chemical system like your body. In fact its known that you can't.

To be fair, you would also want tests to determine the effect of other recently-consumed foods, because combining foods might not lead to strictly additive results.

Reply on news.ycombinator.com