Tuesday, June 18, 2013

Assuming a Benevolent Universe

I recently read a (fictional) book (title omitted to prevent inadvertant spoiler) - where the protagonist, beset by crumbling personal relationships, becomes semi-obsessed with locating a missing girl. At some point, the searche for the girl is discontinued, but he entertains the thought of offering his help to the father of the missing girl. He is not sure whether his help will be accepted but eventually he offers and...
.. he is kidnapped by the father who assumes that the person who contacted him is his girls' kidnapper.
What surprised me when I read this story is that this turn of event took me wholly by surprise. Objectively, one would expect a person showing excessive interest to be a suspect, possibly to get in some trouble. And yet, since the author skillfully did not hint at that direction, I (and assumingly, other readers), divorced that thought off the realm of relevance. 
Very often in life, relations, or work, we expect the universe to be fair or benevolent. Being penalized for trying to help is patently unfair - but it was a possible result of the behavior of the protagonist, a consequence he did not consider or prepare for.  It is not that we should change our moral compass to accept those outcomes, but we should change our expectations to steel ourselves for the inevitable occasions when no good deed goes unpunished. 

Thursday, May 02, 2013

Mindmaps ARE trees!

Mind maps are a useful tool to organize thoughts, options, and ideas - Wikipedia.mindmap
The idea is to draw a circle with the central idea, and multiple spokes as the sub ideas, and continuie (see picture above).  My not surprising observation is that mind-maps are simply trees. So if you don't need the spatial fairness (there is no first element, last element, etc), a simple equivalent is the layers bulletpoint:
- sub idea 1
- -  sub sub idea 1.1
- sub idea 2

The beauty of this observation is that is lets you 'draw' mind maps in email, or on a text editor, without fancy graphical tools.

Thursday, March 28, 2013

Lottery Winners, Hedonic Treadmill, and Financial Planning

It is believed that people return to their basic 'happiness' level despite sudden changes in their environment.
That is, winning the lottery will not keep you happy for long.
OK. So, that made me think. If we derive pleasure/happiness, in a large part, not from the absolute value of our situation (e.g. disposable income) but from the change thereof, we should plan our lives to be continously improving. It also makes life into one obscure differential equation.
If this theory is correct, then we should prefer an income of 30k at the age of 30 and an income of 50k at the age of 50, rather than a constant income of 40k throughout. (assume numbers are inflation-adjusted). The first scenario will include many opportunities for promotions and achievements, and many opportunities to improve your standard of living. The second will not. In that regard, lawyers (who make more as they gain experience) may be better off than technologists (who generally do not past some point).
From a financial planning point of view, this may indicate that a good retirement financial plan should not assume only cost-of-living adjustment but a real increase in retirement income (at least, when possible) even at the expense of a slightly lower initial expense or slightly delayed retirement. Interesting.

Wednesday, November 22, 2006

The great razor-life test

Our dedicated lab rat (me) is trying out two theories to enhance the life of razor blades.

I use a new razor blade a week, or my skin begins to complain. At $2.5/blah this is a $125 a year, or $4000 over the next 40 years. And it is clear Gillette want to sell more, not less, razors...

The Theories
Searching the web, the two 'life extension' theories that stood the Test of Duh were
- 1# Leaving razor blades in petroleum gel between uses may reduce moisture and blade deterioration
- 2# Applying alcohol-based gel hand sanitizer before use will reduce hair accumulation on the blade (and reduce damage) and also instant-treat any micro-cuts, resulting in a usable shave for longer
- 3# Nothing helps

The Test
No way to use a double-blind test; my second-bet approach was:
1. For one week, use #1, #2, or #3 on right cheek
2. Report smoothness/microcuts at the end of the week
3. Use a different combo on left cheek
4. Ensure each theory is attempted on both left and right cheek

The test is already in progress.... final report due next week. Which theory do you believe in?

Wednesday, November 08, 2006

The best DIY energy drink known to Men - EnerT

We all work out, we all need re-hydration, we all buy Gatorade... wait, stop. EnerT, this simple do it yourself tea-based energy drink, may be a better (and cheaper) option.

An energy drink needs sugar (of some form), electrolytes (mostly sodium and potassium) to accelerate absorption, and may contain other ingredients. I opted for tea as caffeine has a beneficial workout benefit, and tea provides a low-level caffeine, striking my personally preferred balanced between 'waking' and 'wired'. Any level of sugars under 8% is acceptable, but I find Gatorade to be overly caloric for those of us trying to balance increased workout performance with weight loss/maintenance.

[Note: Initial idea inspired by similar energy drink I found on the Diet Everywhere site ]

EnerT Ingredients (32 oz or about 1 liter):
2 Tea Bags
6 level teaspoon sugar (24 grams)
A pinch of salt (0.5 grams)
2 oz lemon juice (about 55 grams)
30 oz boiling water

100 calories
500 mg Sodium
60mg of Potassium (from the lemon juice)
Caffeine varies, est 100mg-200mg caffeine (depending on the tea used and infusion duration)

Estimated Cost:
Lemon juice: $0.10
2 Tea Bags: $0.06
Sugar, Salt, water, etc: negligible

Compare to 32 oz Gatorade:
200 calories
440mg Sodium
120mg Potassium
No caffeine
Cost: $1.29

I use a lower potassium level as research indicates that Potassium loss is secondary to Sodium loss in hydration. You can add more lemon juice in, up to 4oz - if flavor allows. On the other hand, removing the lemon juice will still yield a good energy drink, but will reduce price by 66%


Added per the huge amount of interest and questions:
Q: So, is this different from a regular tea?
A: Yes, but only subtly so. The quantity of salt and sugar, and lemon juice (for Potassium) are specifically set for easy absorption. But mostly, energy drinks ARE flavored sugar water. Gatorade is water, sugar, sodium, potassium, and flavor.

Monday, October 30, 2006

Walmart is not my enemy

Sunday, October 22, 2006

Inequality and could/should it be reduced?

Very interesting article here. I agree.

Saturday, September 03, 2005

A modest (flood damage prevention) proposal

When the Katrina rescue efforts are at last over, the inhabitants of New Orleans would no doubt want to rebuild.
Should they?
In fact, people building houses on mud hills, in flood plains, and other danger zones places a burden on the ret of society. It is their right to do as they wish with their property, but it is our obligation to see that the true costs of those decisions lay on those who opt for it. Maybe houses on the flooding riverfront should not be so common. Here's my proposal, which of course would not be popular with propery owners in risky areas:
1. A risk criteria will be federally established, defining areas as low, medium, and high risks.
2. All building in medium and high risk area will require adherance to standards (we have this, but the standards are low)
3. All sale and rent of property (houses and land) in medium or high risk must have the seller disclosure and the buyer's written acknoledgement (e.g. a line in the contract) that the property is in a high risk area. Not doing so is a criminal offense.
4. All built properties in medium and high risks should be assessed a federal 'mandatory rescue costs' tax, maybe 0.25% or more. This should be structured so over time, those taxes would pay the cost of rescue, federal help, and other damages to the US economy from building in those areas.

That's it. Maybe half of New Orleans should not have been built on. and no, federal help for drainage or levy building was not appropriate. If the true cost of a $100,000 home in New Orleans was $100K + $30K in infrastructure cost, this cost should be borne by those living there, not by the entire country.

Wednesday, July 27, 2005

Being risk averse

I saw this post on chaostheory and it really made me think - how risk averse are people in general (and me in particular?)

I think trading $1000 for 60% chance of $2000 might make sense, but trading $1000 for 0.06% for $2,000,000 does not, for me. There are two ways to look at this:

1. Expected resulting satisfaction: I will be happy if I win the 2 million, of course. but not a thausand times more. and most likely I'll be disappointed. If I try this 100 times, I will quickly 'learn' that this is too much frustration and not enough (or any) gain

2. Declining value of money: The value of the $10,000 between $100,000 and $110,000 me is quite high, as it would affect my investment.
The value of the difference between $1,000,000 and $1,010,000 is low. Therefore, the value of a dollar is a function of how much I already have, nd declines with wealth (duh?). So $1,000,000 does not really offer twice as much benefit as $500,000 do, just as ten Mars bars do not offer ten times as much benefit as one bar.

Sunday, July 24, 2005

On profiling and game theory

Two sources I've encountered this weekend were this paper, which uses 'quality control' logic to show (among other points) why profiling is effective, and Bruce Schneier's post which raised the legitimate point that profiling based on some attributes (say, ethnicity) will motivate terrorists to masquerade those attributes (e.g. dye their hair blonde). However, this possibility does not mean that profiling is ineffective. Since the opponents have finite resources, and since masquerading has an associated cost (for example, the supply of likely terrorists might be reduced if they would need to take accent-reduction classes, or learn to convincingly masquerade their origin.
It is easy to see that while the approach of ‘only inspect people with suspect attributes’ is ineffective, equilibrium (at max probability of screening) is reached when the screening strategy chooses people of ‘profiled’ attributes more often than people without it.

This is really common sense, however.

Possibly in a future post I will run some simulations with fictitious numbers to demonstrate how the numbers work out.

Saturday, July 23, 2005

When will blogging peak?

see this article - I agree, blogging can't be that important forever, not when half the bloggers out there would figure 5 people are reading them :)

Friday, July 22, 2005

Fallacies - How arguments lie

this webpage discusses a classification of fallacies. While I don't care how you classify them, the list of fallacies (or misleading ways of arguing) is very helpful.
My favorites:
Ad hominem - we use this one all the time. While an interested party may make biased *decisions*, his logic should stand (or fall) on its own merits.
black&white thinking - all or nothing thinging.

I highly recomend you read the website above - and I know you trust my judgement


Thursday, July 21, 2005

Job choice is like the stock market

Look what I found here:
I thought education and career are like stock exchange, you should always buy low and sell high, but most people tend to buy high and sell low.

Saturday, July 16, 2005

Wedding Bills

"Maybe there's a good reason that so many people shed a tear or two at weddings while the rest bear a grin of pure schadenfreude. The average cost of a traditional wedding in the U.S. is now up to $25,000" says the Christian Science Monitor.

Link brought to my attention on Plastic.com

The food we eat

This link discusses how deceptive labeling is hiding the true content of the food we eat. While I usually am more pro-business than pro-newage-liberalism, this is at least somewhat alarming. They are out to get you, or at least, get their profit margines.

Thursday, July 14, 2005

MCSE certification hurts salary? really?

This blog post discusses how MCSE holders aherage $10K less than people who hold no certification. "The title is a little misleading, but this article is a pretty interesting read."

Actually, statistical analysis shows that those numbers mean little for the person trying to decide whether to get certified. It stands to reason that people who feel the need to get MCSE are less credentialed (or otherwise less successful) than those who do not - no-one who was just promoted to a development manager, for example, would have the time or inclination for a resume-building effort. It is those who did not get the promotion/job/assignments they wanted who will. And those are just the people who would be paid less.
Yet, stands to reason isn't statistics, and reason is often wrong. My take? inconclusive. an interesting double blind would be to have half of a group (randomly chosen) get certified, and the other half not, revisit them in 2 years and see their salary delta. Double blind? not quite, but almost. Is this experiment going to happen? I thinketh not.

Prisoner Dilemma: Tit for Tat or Suicide Bombing?

As described in this article, the latest prisoner dilemma competition (game theory is very cool) was won not by the usually 'tit for tat' algorithm ("be nice, fair, but tough") but by a "identify your friends, collude with them, walk over everyone else" system.

This shows that, in this game theory problem at least, there is a balance and there are weaknesses. Overly-altruistic (say-civilized) are vulnerable to overly aggressive approaches; as the number of aggressors increases, cooperation gives a competitive advantage, turning the tide.

Also read this plastic article

Saturday, July 09, 2005

Firecalc and retirement

using firecalc, a tool that calculates the probability of being able to withraw a specific pension from fixed savings based on past stock market behavior, I got the following:
assuming a desired withdrawal of 30K, the probability per initial investment is:
700k - 89%
650k - 78%
600k - 72%
550k - 56%
500k - 51%

What is morality?

This quiz is an interesting way to get a good insight on what you consider moral. Thanks to Paula and Ruth who had the original link (or link to link, or...).

So what do I think? I think almost nothing which harms no-body is immoral; and I don't believe the source of morality is some higher power. And yet - society has the right to prohibit behavior which isn't strictly immoral.
--- if you plan on taking the quiz, do it before reading the rest of the post ---

The results analysis discusses 'the yuk factor', the tendency to call something disgusting 'immoral'. That is human nature - delude yourself as to the reason of your actions.
However, it is hard to take this test's results seriously. They do specify in the question that "you should assume the action does no harm". But that reminds me a story:
Q. How many wheels do 10 SUVs have?
A. 40
Q. If I told you an SUV has 6 wheels, how many wheels would 10 SUVs have?
A. 40
It does not matter what they claim the results of an action is, the reader must (and will) reject unreasonable assumptions. "Assume you kill all the bad guys and only them, noone ever gets hurt" is a fine question. But we do know that this isn't reality; someone will get hurt, and the bad guys have friends.
In actuality, when we read a statement such as 'assume this has no consequences' we read 'assume you were told this has no consequences'. Which leads many people to 'well, I assume I was told wrong'.


More and more it seems some people insist on seeing the terrorists' side of things. "They only kill innocents because that is their only way of expression".
Fvck that.
If your only way to fight a war is to murder innocent civilians, you shouldn't be fighting. you lost. There is no justification to terror, of course they have reasons, people always have reasons, Hitler had reasons, I'm sure the Devil (if I believed in him) had reasons, so what?
The word 'understanding' has two, conflicting meanings.
Understanding the enemy point of view in order to best defeat him, that is one meaning. That's good. understanding the enemy in order to explain away immorality, that's an equine of another color.
.. and now, Hollywood wants to make a 911 movie. just great.