Journaling on estrip is easy and free. sign up here

Tinypliny's Journal

tinypliny
My Podcast Link

11/19/2011 00:48 #55547

Wachet auf, ruft uns die stimme
Category: music
I am re-discovering all of Bach magnificent Cantatas once more and BWV140 is one of my all-time favourites.



I love the absolutely perfect build up of the vocal ornamentations to the crescendo of the choir from 4:26. Hearing the cantatas is almost equivalent to getting a virtual pair of wings if you close your eyes.



Did you know that Bach wrote a brand new cantata every single week for the Sunday worship service, and that there are nearly 260 of them? Prolific was a gross understatement when it came to Bach.
tinypliny - 11/19/11 23:26
How could it have been misspent if you heard so much good music!!!?? :) You need to post more. I miss your posts!
uncutsaniflush - 11/19/11 22:18
Yes, I did know that about Bach and canatas. In my misspent youth, I was a big fan of Johann Sebastion and a wee fan of his sons.
metalpeter - 11/19/11 17:11
nice

11/18/2011 12:27 #55542

djvu files in Fedora
Category: linux
Evince has djvu file support but just not in Fedora, at least by default.

You need the evince-djvu module to make Evince launch djvu files:

sudo yum install evince-djvu


11/18/2011 09:37 #55540

The Emperor's Waltz
Category: dance
The Waltz is my most favourite kind of classical music after sonorous abstract cello recitals and baroque orchestral works. Wait till you reach around 3:20, when they give the dance a modern interpretation. :)


tinypliny - 11/18/11 17:17
hah, I would love to (e:metalpeter) but I need a partner who would be okay with me being bald some months of the year and who won't have trouble seeing my face from far up there. If I get here, the dress won't be an issue at all. :)
metalpeter - 11/18/11 16:55
Didn't know they allowed guys with guys and girls with Girls ;) .... Ok now go out and practice it .... Oh but it has to be in one of those dresses ...HA

11/17/2011 19:10 #55537

United States of Obesity
Category: science

SUPER EDIT:

Though I like the poster below for how it presents information, I think all of us need to review this really informative book on reality and myths surrounding Obesity: directly leads to the PDF.
Source: Center for Consumer Freedom (CCF)


NB1:
The book has some excellent data presentation - far prettier than the poster below. A pity, its locked into PDF. I am going to see if I rip those diagrams and also put it on this post in direct contrast to the pretty poster in my original post.

NB2:
For me, the real challenge is to examine if I am cherry-picking results or have done a balanced survey of the peer-reviewed literature. I am currently disinclined to accept any pigeonholed set of beliefs on the many sides of this whole obesity discussion. In many ways, I see more confounders in the equation than is comfortably explained away. If there is an epidemic that needs immediate addressing, its one of confusion and indecision.

And now my original post:

Just saw this here:
image

Now, that is classy presentation. Twenty years back, I would have been a 4 instead of being relegated to 2s and 0s most arbitrarily.


EDIT:
The real sources
I posted this because I find the data presentation in this poster very compelling. It is capable of eliciting a strong response from everyone just because it is so directly and simply presented. However, you have raised some concerns over the source of the chart. So I dug more and came up with the real sources of this compilation:
  • The portion size statistics and graphics came from the National Heart Lung and Blood Institute's (NHLBI) website. They have 2 full slide sets for post and present portion sizes:
  • Set 1:
  • Set 2:
  • The BMI chart was adapted from the NHLBI's evidence report titled: "Clinical guidelines on the identification, evaluation and treatment of overweight and obesity in Adults. The evidence report." Available at:
  • The facts for the diagram came from here:
---

I am also going to keep adding equivalent data presentations from non-commercial sources to this post.
image
Source: RAND health, an offshoot of the famous RAND Corporation, a non profit think tank whose name expands to just Research And Development
Data from Sturm, R., J. Ringel, and T. Andreyeva, "Increasing Obesity Rates and Disability Trends," Health Affairs, Vol. 23, No. 2, March/April 2004, pp. 1–7.
image
image
image
image
Source: Centers for Disease Control a.k.a. CDC
See:
--
Food subsidies and the food pyramid
{Re: (e:heidi)'s observation. Last para Comment #66257}
image
From PCRM. Physicians Comittee for Responsible Medicine

Note that the government recommended food guide pyramid is now a plate with %s. I personally think it makes more sense and is more comprehensible even though the new food plate also has its critics. An argument is that the government shouldn't be telling us what to eat. However, people who determine the food pyramid or the plate are just scientists working for the government. This diagram raises the real parallel question for me, why are these scientists who come up with food portion recommendations not on the committees that decide on what to subsidize?!
...
And just for fun, from Apples From Oranges:
image


tinypliny - 04/02/12 23:01
It's even worse than we thought. BMI is not a very good measure. But it actually *underestimates* the level of obesity, especially in women. See: :::link:::

tinypliny - 11/20/11 12:28
See major edits and additions.
tinypliny - 11/20/11 12:17
(e:heidi), I need to address your comments. I have been side-stepping them because of laziness. :) I will be back and discuss. This is a really absorbing topic for me, in terms of how I think about evidence-based primary health care.
tinypliny - 11/20/11 12:09
Added a PDF link to a really informative book. Many valid and some slightly weird points. :)
"Obesity Myths" :::link:::
Source: The center for consumer freedom :::link:::
metalpeter - 11/20/11 11:57
Yes that is what I was trying to get at........

tinypliny - 11/20/11 11:48
I think I see your 66% point (e:metalpeter). (late realization. haha) In statistical terms you are talking about normalization vs. standardization. I think I will do a post on the pros and cons someday, but yeah, I think I finally see what you trying to say. Valid point.
tinypliny - 11/19/11 08:28
Added some more presentations from relatively non-vested sources.
mike - 11/19/11 01:28
while I agree there might be a weight problem in this country I don't really trust a chart put out by the people who make some miracle weight loss product. it seems rather sketchy to me.
metalpeter - 11/18/11 19:44
@(e:heidi) I think that we both agree that BMI is used as a way to say this is how you are supposed to look......

I assume though we differ on the Pizza thing.... They say Tomato Paste counts as a vegetable when one has Pizza... But here is my question does the Cheese count as Diary.... Does the Bread count as grains and does the meat count as meat.... Yes I'm unhealthy but that is why Pizza is such a great food one can get it with all kinds of things that are good for you...... Not that...That is the kind of pizza I got in school.... But even bad Pizza isn't so bad cause it is Pizza.... It would be great if they also tried to encourage other types of Pizza... But what I wonder is when you get pizza as lunch what else does it come with.... Does it mean that they give you corn and on pizza day that is it and no sides and no Salad? Now that being said I think Pizza should be part of school lunch... Picture "Little Johnny " or little "Aquafina ".... Ha sorry about that and they come home ...... "oh how was school" " Green peas ma off to do home work" kid is all down and depressed and school sucked that day... Now with Pizza they come home and are happy and liked Gym class...
tinypliny - 11/18/11 19:11
I admit, (e:metalpeter). I am still kind of puzzled about your question so I am going to dissect your comment here (marked with %). Please chip in if I am going on a wrong track:

% If lets say I picture all the people on ((e:strip)) and people I know from say work and % the internet I can say who I think is fat and who is thin...

Okay - to be REALLY unbiased and random, also count every person you see on the street for say the first hour of your doing this exercise because your circle could have unique characteristics. We want as much randomness and variability in our sample as possible.

% Now of course one would then also have to look at the height and width of
% shoulders.... Then decide again who is fat...

Good idea. Also take their ages and waist-to-hip circumference.

% But to get if someone is over weight is different we would have to take everyone % who is has about the same shoulder width and height and then all jump onto a
% scale.... one would then get the Average weight...

That makes sense. You take averages of similar build and similar sex people. You cannot compare say, me and (e:YesThatCasey). :)


% (now granted I know they have other ways to measure like percent body fat and % they have water tests to see how much water is pushed away and things like that % an weight isn't the only way to go).

That is true.

% Now anyone who is under the Average would be underweight and and one over
% the average would be over weight...

okay.


% To be honest for the math to work out we would need a lot more people then I
% know it would have to be what ever they call a good sampling....

Precisely, so just count people from the street as well. Everyone you see in around 1 or more hours without bias gets to be in your sample.

% Now that being said it is possible that no one would be average if you have many % thin and many fat people....

!!!! It is possible but VERY unlikely. I am in the "normal" weight category, (e:Paul) is, so are you, I believe. And surely you will count us all in your sample. For your assumption to be met, your sample would have to be REALLY biased. And we established before that we are going to do a good sampling!


% with out doing any math the amount of over weight people should be pretty close % to the amount of underweight people....

!!!!! how can you say that?? Without doing any calculations, there is no average. If you took pains collecting your study sample, why not calculate the average? Are you going to throw away your effort?? This is like saying, I will buy the best pie there is at the Coop, come home with the pie and just throw it in the garbage because you think it won't taste good!! How can you say, without tasting it?? I don't get this statement at all.

% I think where we differ is that you are looking over time....

I am not. All these people in your life, they are alive!! We are looking at the present. We are just determining what percentage of people in a random sample of people is overweight or obese. There is no time factor involved. This little study we want to do is entirely a current snapshot - or in other words, cross-sectional.

% People know as compared to some other time... But I could be wrong...

We are just concerned with your original question. You said you didn't believe that 66% of the people could be overweight or obese and this litttle experiment I am suggesting is just to find out if your beliefs (or my beliefs or in fact, that govt data) checks out or not.
heidi - 11/18/11 19:06
Just to prove the last point I made: :::link:::
metalpeter - 11/18/11 18:50
The other thing is that there is a society view that factors into this... Way long in the past being fat was considered Healthy it meant you had good health and ate and if you where skinny you looked like you didn't eat and you where thought of as sickly now it is other way around.....

I think the charts they use are old and outdated and as populations change in size the charts need to be updated.... It is also important to mention that national origin effects what chart you are on.....

Yes we do eat a lot of crap and stuff that is bad and in America Eating is about enjoyment and this town is a food town for sure.... Just go to any festival and see all of it... Deep Fried Cheesecake , French fries with cheese Bacon Bits and gravy.....

I think where we don't agree is that I say you need to measure people at this time not from standards of years ago or maybe I just don't get the math?
heidi - 11/18/11 18:49
I wasn't arguing that Americans haven't gotten fatter or that population studies aren't useful. I'm saying BMI is bullshit. There are many ways to critique it but I like this one: It's based on a nonrepresentative young white European male military recruit sample from 1846. 1846!!??? :::link:::

I strongly object to how BMI is used to label, shame and control people, particularly women, into harming themselves to fit a constructed and illusory "ideal". Dieting doesn't work. Eating well and exercising do not guarantee - and are unlikely to create - sustainable/sustained weight loss. But eating well (not eating junk) and being active (not being sedentary) are very important for healthy outcomes regardless of size. I'm really really not okay with shaming anyone about their bodies, which is what this poster is trying to do. Everyone deserves dignity and respect. This culture of body shame leads to bulimia and anorexia, crazy yo-yo dieting, and bad sex. (See for example, Fredrickson, B., Noll, S., Roberts, T., Twenge, J., & Quinn, D. (1998). That Swimsuit Becomes You: Sex Differences in Self-Objectification, Restrained Eating, and Math Performance. Journal of Personality and Social Psychology, 75, 269-284. )

How reliable is BMI? Completely arbitrary:
"In 1998, the U.S. National Institutes of Health brought U.S. definitions into line with World Health Organization guidelines, lowering the normal/overweight cut-off from BMI 27.8 to BMI 25. This had the effect of redefining approximately 25 million Americans, previously "healthy" to "overweight".[12] :::link:::

Yes, this poster is a very pretty presentation of data. However, it was created by a company that is interested in promoting its product. Check out this article about who is promoting the "obesity epidemic" story :::link:::

"[F]atter cardiac patients were more likely to survive hospitalization and invasive treatments than thinner ones, even when adjusting for age and other contributing factors. In this analysis of 130,139 heart disease patients, 5.4% of “normal” weight patients died, as compared to 2.4% of “obese” and 3.1% of “overweight.” :::link:::

And please remember the ongoing conversations about US food subsidy policy. Broccoli farmers do not get subsidies. Corn farmers do. There are huge structural issues that impact people's access to healthy foods and individualizing and blaming is not helpful. Again, the goal of this poster is to shame people and individualize the problem.
metalpeter - 11/18/11 18:43
Ok I think I see where we have a disconnect.......

If lets say I picture all the people on (e:strip) and people I know from say work and the internet I can say who I think is fat and who is thin... Now of course one would then also have to look at the height and width of shoulders.... Then decide again who is fat... But to get if someone is over weight is different we would have to take everyone who is has about the same shoulder width and height and then all jump onto a scale.... one would then get the Average weight... (now granted I know they have other ways to measure like percent body fat and they have water tests to see how much water is pushed away and things like that an weight isn't the only way to go). Now anyone who is under the Average would be underweight and and one over the average would be over weight... To be honest for the math to work out we would need a lot more people then I know it would have to be what ever they call a good sampling.... Now that being said it is possible that no one would be average if you have many thin and many fat people.... with out doing any math the amount of over weight people should be pretty close to the amount of underweight people.... I think where we differ is that you are looking over time.... People know as compared to some other time... But I could be wrong...
tinypliny - 11/18/11 18:26
NY is close to 60% obesity levels :::link:::
tinypliny - 11/18/11 18:20
Here's an arbitrary exercise. Count around 35 of your close friends and relatives and acquaintances. According to your perception, mark them as normal, overweight, obese. Add the number of overweight + obese and divide by the total number of people. See if you come close to the average overweight/obese population for New York State.


====
  • NOT less than 30+ people because that is where the central limit theorem - a statistical measure that is capable of approximating/extrapolating measures from small samples to populations and vice versa begins to be valid.
tinypliny - 11/18/11 18:15
I am not sure I understand your comment, (e:metalpeter). Alabama's population last year was 4,779,736. 70% of this population, 0.7 * 4779736 ~3345815 people in Alabama are either obese or overweight, based on extrapolated data.

Similarly, the US population was 308,745,538. 0.66 * 308745538 ~ 203772055 people in the US are either obese or overweight, based on extrapolated data.

Even you attributed up to 10% measurement and sampling error to those estimates, the numbers are still pretty high.
metalpeter - 11/18/11 17:55
But see here is the thing you can compare people to other times and that is fine so one can say based how people weighed say as an example 10 years ago.... But based on everyone they measure you will get an average number again based on height and age and all those things... but what the percentage above and below the average is I have no idea but I know that 66% isn't possible.... Now if we are talking about over time that is something different.....
tinypliny - 11/18/11 17:08
I also need to mention that since they cannot really come around and measure everyone, they collect a national representative sample. They ensure representativeness by carefully looking at census data, determining age, race and ethnicity distribution in various area of the US and making sure that their smaller study population reflects this distribution.

From that PDF's last table: For males in the age group of 30-39, for instance,

In 1960-1962, in a sample of 714 males, the average weight was 169.9 pounds (+/- 1.4 pounds)
In 1971-1974 in a sample of 654 males, the average weight was 178.1 pounds (+/- 1.7 pounds)
In 1976-1980 in a sample of 871 males, the average weight was 175.5 pounds (+/- 0.9 pounds)
In 1988-1994 in a sample of 1,468 males, the average weight was 182.3 pounds (+/- 2.0 pounds)
In 1999-2002 in a sample of 704 males, the average weight was 189.1 pounds (+/- 2.0 pounds)
tinypliny - 11/18/11 16:57
(e:metalpeter), here is the data that you might be interested in: :::link:::

I don't think presenting average weight without stratifying by age and sex would be an accurate representation of population averages. That PDF comes from the National Health and Nutrition Survey and gives you the age-standardized average weights from the 1960s through 2000s.
metalpeter - 11/18/11 16:46
Yes we are unhealthy and eat bad food but the numbers on here are wrong and with Out doing any math I can tell you they are wrong based on this:

2/3 of people are overweight or Obese ? That is incorrect if 66% of the people are that heavy then that would mean the average weight needs to be higher... I'll say that again if 66% of the people 2/3 or heavier then what they are supposed to be that means what they are supposed to be needs to come up in #..... Another way of thinking about it is 33 percent of people are at or under weight.....
YesThatCasey - 11/18/11 01:30
That is an excellent collection of data and presentation.

Heidi, I don't agree that BMI is bullshit. It isn't absolute, as it can't account for the content of that weight, but it is extremely useful for gathering general data on large groups over time. It's not as if our bone mass can account for the dramatic shift in BMI over the past 20 years, and we definitely are not more active with the advent of the computer. Of course, when visiting a doctor or a trainer, there are better ways to get a more precise measurement as to the condition of a body.
heidi - 11/18/11 00:11
BMI is such bullshit.
:::link:::

11/17/2011 08:36 #55532

Oh no! Something has gone wrong
Category: linux
You have to hand it to the Gnome3 folks. At least the message looks crisp and smart.

image

Even though what followed was an undeniably ungraceful crash.