Saturday, Jan 16th, 2010 by
admin
Category:
Internet,
Programming Tags:
benchmark,
statistics,
web
Ever since Intel and AMD have been selling multi-core cpus, the Erlang hype has been growing continuously. The number of high profile projects using Erlang were flagrantly announced over the blogosphere as the coming of the second C.
We kept hearing about: rabbitmq, couchdb, nearly all of Amazon’s AWS, Heroku’s routing grid, Facebook chat, all switching over to Erlang because it was the fast and concurrent language of the future. No other language could hold a candle to a language run by telecom giant Ericsson which was then validated by Amazon and Facebook! Or not? When I tried to find benchmarks for Erlang, they all showed otherwise.
http://muharem.wordpress.com/2007/07/31/erlang-vs-stackless-python-a-first-benchmark/
http://www.joeandmotorboat.com/2009/01/03/nginx-vs-yaws-vs-mochiweb-web-server-performance-deathmatch-part-2/
Since I initially bought into the hype, I felt compelled to test it myself to see if they made a mistake. To do this, I wrote a simple http server in Erlang, Haskell, and Python that simply outputs an HTTP reply “Pong!”. And here are the results in a graph.
The green line is the maximum req/sec possible. Higher is better.
For more details, continue reading.
Read the rest of this entry »
Sunday, Jan 3rd, 2010 by
admin
Category:
Economics
Many Republicans believe in Reagan’s trickle down theory. They believe that by offering businesses a tax break, they will use the money to hire more people.
You don’t improve the economy by paying people to sit around and not work. You improve the economy by lowering taxes so small businesses will create more jobs.
- Representative John Linder Georgia Republican
How this frequently parroted sentiment is so popular is understandable, but… does anyone else see the problem with this argument?
If you give businesses tax breaks, they are going to horde it because they know that consumers aren’t going to have any money for them to earn.
However, if you give consumers and families a tax break, it encourages businesses to produce more, likely hiring more people in the process because that’s where the money is. Individuals and families also pay a higher tax rate than corporations that like to:
- Deduct everything as a business expense or low capital gains tax. Like personal business laptops and cars.
- Hide it in tax haven states or countries like Ikea and Walmart
- Use it to boost company stock which can be cashed out at a rock-bottom 15% capital gains tax, the same tax rate as someone earning a paltry $8,026-32,550 in 2008.
Potential Rebuttals
- But individuals will also horde money!
Read the rest of this entry »
Wednesday, Dec 16th, 2009 by
admin
Category:
Economics
People hate taxes, and everyone can agree to disagree on them. One thing most people agree on though is that it’s too complicated.
Here is a diagram of the US income tax in 2008.
US Income Tax 2008
What’s wrong with the current piecemeal tax rate?
- It’s complicated.
- People want to earn less if taking a raise barely puts them in the next tax bracket.
- The tax rate increases faster for people earning very little compared to people earning a lot.
What’s wrong with a flat tax?
- The tax rate increases more slowly as your ability to earn more money increases exponentially.
- The more money you have, the easier it is to get bargains, higher interest rates on bank accounts, and investment opportunities that allow you to own controlling shares of the next big company.
- A 20% tax of someone earning $20,000 a year can cause starving and homelessness while 20% of $10 million doesn’t really make a dent.
How do we fix this?
Read the rest of this entry »
Wednesday, Dec 16th, 2009 by
admin
Category:
Codexon
The contact form for this website has finally been fixed!
For some reason the WordPress mail sending function refused to work.
All thanks go to this handy plugin which uses SMTP instead of the PHP function mail.
http://www.codexon.com/contact
Wednesday, Nov 25th, 2009 by
admin
Category:
Psychology
Recently England has mandated that children from the age of 5 would be taught “how to prevent violent relationships“.
One could explain nearly any “odd” but prevalent human behavior through evolutionary psychology. You would think it would be strange to attack someone you were living with and providing benefits. But I found it hard to be surprised when I found a paper that explained domestic abuse as mate retention behavior.
In a 2002 study by Peters, Shackleford, and Buss, around 4000 cases of wife abuse in New York was analyzed. They found that abuse was positively correlated with fecundity. Fertile and younger women were 10 times more likely to suffer domestic violence compared to older women. Large decreases in risk were found after the females were 45 years of age when menopause is likely to occur.
http://www.toddkshackelford.com/downloads/Peters-VV-2002.pdf
If domestic abuse truly serves an evolutionary purpose, schooling might not be enough to curb tendencies.
Subscribe to the feed.
