fastoptimizer

Lustig is unfortunately a scientific crackpot. His lecture has many untruthful statements (like those about Japanese) and puts blame mostly on fructose.

As much as fructose does cause some interesting metabolic problems, unlike glucose, it still cannot be characterized as unhealthy (because the very Japanese he praises eat a lot of fructose and the studies are done on fructose extracts which can't even be reproduced with equivalent fruits in fructose amounts).

He makes equivalent claims to those made by Campbell in his China study that casein from milk causes cancer - of course it does if it's 20 freaking percent of your caloric intake (which is practically impossible to consume long-term).

I'm a diabetic and I eat 70% of calories in carbs. I make sure they are all low GI.

My statement was mostly about average people, not sick people. I'm fairly sure diseased individuals are an exception and have to make sure to put their disease in remission (or to meet their genetic predispositions) with proper diet plans.

Sugar isn't useless. It gives you energy to function.

If one overconsumes and allows for that useless sugar to circulate, or useless cholesterol, or useless anything from overconsumption, a plethora of interesting metabolic things happen in the body and some of them cause disease.

This is a very nice skeptical stance on how to observe food in this age of non-personalized medicine and nutrition:

https://atheismfaq.quora.com/Is-this-food-healthy

Of course not.

Is sunlight bad because it can give you cancer?

Are apples bad because they contain cyanide in seeds you can accidentally consume?

Is red meat bad because it causes cancer?

Is alcohol bad because it causes cancer?

If you feel you're living a healthy lifestyle, you can easily check that by doing some medical tests. No diet related disease is going to invisibly attack you at a random moment in life.

Atherosclerosis is fairly visible. Diabetes too. With regular checks you can be sure you're fine.

There hasn't been a study showing that calorie controlled fruit diet will cause you any harm (except potential malnutrition if fruit selection is not as varied).

Just like excessive consumption of protein (but with a calorie cap) in healthy people shows no harm, or similar studies on healthy, physically active people.

Diet is just a part of a healthy lifestyle and a huge variety of diets can be healthy. This blaming of particular ingredient or macronutrients is unscientific, I as a skeptic dislike it very much.

Fruits are rich in fiber which slows down the absorption of their more "potent" sugars, not to mention that digestion of that fiber is far more taxing than digestion of something else.

Of course, if I were to eat excessive amount of calories it's quite probable I'd get the same symptoms as those people consuming the equivalent in refined sugars.

> The hypothesis is that sugar does not make people feel full whereas fat does, so they eat more.

They still eat about 1500kcal of fat and protein - of which I eat about half the amount.

If you were really talking about feeling full, nothing makes you feel more full than a belly and guts filled with lovely starch/fiber from fruits and vegetables, which an average US eats little if not any.

When in fact nothing can be blamed but our overconsumption of almost everything but vegetables and fruits.

I eat about 70% from my calories in carbs (some are refined), but it's just 2500kcal calories per day. I'm pretty sure the fact that average US citizen eats 3500kcal daily has more to do with anything than what particular ingredient one eats.

edit: some irrational unscientific and anegdotal statements require me to remove myself from discussion. thanks.. for the downvotes, I'm outta here boys! :wink:

Do they say how much time is the generation taking?

Is this insanely slow to train but extremely fast to do generation?

> Better operational efficiencies?

Some of them do not realise that the logistics problem (of picking up stuff and delivering them) can't be solved by common routing.

Most of them work mostly on their mobile apps.

Some of them realise that the logistics problem they have is an NP-hard problem and try developing the solution in-house. Most of the time it results in failure (I believe some other firms that fell apart, I believe it was some Google acquired team cleaning services, particularly mentioned they couldn't solve the logistics problem in time and just when they did the funding disappeared).

Some realise that they can outsource their optimisation to services like:

+ open source: optaplanner, OpenVRP, jsprit, mixed integer programming solvers

+ proprietary services (ascii sorted): ArcGIS, ClearD Optima, DirectRoute, Foxtrot, Graphhopper, Optimoroute, Paragon, Roadnet, Route4Me, RouteXL, Routific, Routyn, Satalia, jOpt

All designed for different variants with some features that miss in others, available in all or some.

For some business the quality of routed paths isn't that important, for some it's crucial, especially if one wants to bring down the price.

Oh yeah, the problem is called Vehicle Routing Problem (or multiple travelling salesmen), it has variants that include vehicle capacities, pickup and delivery (one location preceding the other), pickup or delivery time windows, lunch breaks that don't have real location and other features.

That stuff is hard to solve if you have to run a business like Washio.