News for hackers. Stuff that matters.

A comment on Who knew cereal was so bad for you?.

Re:Who knew cereal was so bad for you? by ghshephard on Monday June 27, 2011 @02:04PM (#2702293)
In the 1930s to 1950s, anything starchy was bad for you (Ask your grandmother) - In the 1950s - 1980s fat became the new evil. Then it was cholesterol, and saturated Fats. Then Transfatty Acids. For a small period of time there was a niche rebellion against all things carbs, and now it looks like we're coming back against refined sugars (or all things refined white)

After reading a bit of anti-carb propaganda (Taubes) and being drawn into the "Wow, there are really bad things that you can eat" worldview - I decided to educate myself, and wget downloaded about 16,000 journal articles from the American Journal of Clinical Nutrition (They let you do this for $25, with the only request being that you put a 3 second delay between requests. Some awk/sed skills required to parse out the URLs, but not much. Totally download time was 39 hours)

After reading through 40 or 50 of those articles on topics like "metabolism", "glycolysis", "diabetes", (btw - can anyone recommend a good PDF index tool - Lucene is really not easy to plug/play install - I ended up using recursive grep -il on the pdftotext versions to track down docs (very 1993) - but surely someone has solved the problem of indexing 16,000 PDFs in a directory - google fu did not yield much, surprisingly) - I realized that I needed to pick up more education.

Two books I'm making my way through:

  o Essential Cell Biology, Second Edition - $7+shipping.[1]
  o Lehninger Principles of Biochemistry, Fourth Edition - $14+Shipping. [2]
Plus, the one I really, really want - but used 2004 editions are insanely expensive, and there has been a lot of useful developments in nutrition and metabolism since 1999

  o Advanced Nutrition and Human Metabolism [3]
Thank goodness for a generous "Amazon.com Look Inside" allowance - I've probably spent about 10+ Hours perusing this book.

My takeaway, after admittedly putting only three or four months into this topic, is the following:

  o Very little has been donein the way of double-blinded,
    placebo controlled, rigorous nutrition studies.

  o The few that _have_ done these studies, have usually 
    come in with an agenda (Low Fat, Low Cholesterol, Low
    Carb).

  o For how important this whole field is, it's amazing 
    that we've invested so little.
So yes - most cereal (the exception potentially being anything your Paleo ancestors would have recognized as cereal) is probably bad for you. I'll go farther than the OP and say you're probably kidding yourself if you think that "Corn Flakes" or "Crispy Crunch" is significantly better than "Fruit Loops" - it might be marginally better, but none of the refined stuff is doing great things for the western metabolism.

[1] http://www.amazon.com/gp/product/081533480X

[2] http://www.amazon.com/gp/product/0716743396

[3] http://www.amazon.com/Advanced-Nutrition-Metabolism-Sareen-G...

[ Parent | on HN ]
Re:Who knew cereal was so bad for you? by r00fus on Monday June 27, 2011 @04:48PM (#2702955)
Regarding your indexing issue, have you tried SOLR (Lucene powered)? I ran the Blaze VMWare appliance [1], and it worked great for small collection of text-based pdfs.

[1] http://susegallery.com/a/Kr7Ayv/blaze-appliance-for-solr

[ Parent | on HN ]
Re:Who knew cereal was so bad for you? by ghshephard on Monday June 27, 2011 @05:17PM (#2703049)
I installed, SOLR, of course, but then ran into the challenge of figuring out how to automatically feed in the PDFs. I'll take a look at the Blaze implementation - perhaps they've packaged all the components and cookbooked the installation process. Thanks very much for the pointer.
[ Parent | on HN ]
Re:Who knew cereal was so bad for you? by VMG on Monday June 27, 2011 @03:12PM (#2702563)
Do you intend to write up your interpretation once you are done?
[ Parent | on HN ]
Re:Who knew cereal was so bad for you? by ghshephard on Monday June 27, 2011 @05:32PM (#2703099)
So, the challenge, of course, is that there isn't really any correct answers here. Basically, the deeper you go, the more complex you realize things are. A great science writer like Taubes can write an very respectable and carefully footnoted tome (Good Calories, Bad Calories) describing how dangerous carbs are, but then people can come up with counter-examples of asians who have a predominantly white rice diet who live a long, lean, and healthy life. So then he pivots a little, and now it's _refined_ carbohydrates, in particular the sugars and refined flours that are so common in western diets.

It's a little like summarizing "What is the best way to program" - there isn't any one way. A Sr. Developer has lifetime's experience of tools, languages, methodologies, design patterns, and passion that allows them to make a decision on how to approach a problem given a broad pool of knowledge.

Likewise a Dietitian has a whole host of experience, knowledge, tools, and approaches that may be appropriate for different circumstances. Bodybuilders eat plenty of carbohydrates in their off season to bulk up and put on muscle, switch to low-carb to drop fat as they cut, until they get close to their show date, at which point they start adding carbs back to their diet to gain muscle pump. Same person, same profession, same macro-nutrient - different approach.

It's really about understanding the multitudes of mechanisms than it is about any unified interpretation. As much as I love reading Gary Taubes work (and I do!) - I think he has a tendency to make things a little more simple than they really are.

Sorry, no silver bullet. :-)

[ Parent | on HN ]