Warning: include_once(/home/nullvoid/blog.mikezhang.com/wp-content/plugins/wordpress-support/wordpress-support.php): failed to open stream: Permission denied in /home/nullvoid/blog.mikezhang.com/wp-settings.php on line 217

Warning: include_once(): Failed opening '/home/nullvoid/blog.mikezhang.com/wp-content/plugins/wordpress-support/wordpress-support.php' for inclusion (include_path='.:/usr/local/lib/php:/usr/local/php5/lib/pear') in /home/nullvoid/blog.mikezhang.com/wp-settings.php on line 217
NullVoid » 2005 » November NullVoid » 2005 » November

Archive for November, 2005

How to use a bluetooth headset on a thinkpad laptop

Wednesday, November 30th, 2005 -- By ET

I just bought a BlueTooth headset from Logitech. It is supposed to be used with a blue tooth enabled mobile phone. But since I spend most of my time with my laptop (a thinkpad with built-in blue tooth support), I thought it would be nice to use it as a headphone for the laptop, too.

It is not as easy as it sounds to set it up. In the beginning, my laptop just complains about not finding any bluetooth equipment around. I tried all kinds of solutions including turning off my wireless adaptor on the laptop and reinstalling the bluetooth driver.

Turned out the Microsoft driver for bluetooth included in Windows XP sp2 does not support bluetooth headsets. So I found that I need to install a Wincomm driver. Spent some time searching and found that IBM has it here.

Updated the driver, and it still complains not finding the headset. After searching around, I found that installing the IBM driver leaves a directory at C:\DRIVERS\WIN\BDCACT, there is a installation executable called BTW.msi, installed the BlueTooth Software Package from Wincomm. This time I got no problem paired the two devices.

By setting the audio device in Skype as “BlueTooth Audio”, I’m able to use this headset to make wireless phone calls through Skype. This is what I call cool. :-)

FireFox 1.5 released

Tuesday, November 29th, 2005 -- By ET

I like FireFox, not just because it is easy to use, but also because it represents something innovative growing under the shadow of something very overwhelming. IE has been a good browser after version 4.0. Before then, Netscape was the best. I gave up using Netscape because the support of Chinese was not as good as IE.

From looking at the great websites of today, I keep finding something amazing. We always talk about mind-blowing innovativeness in ideas to win the market (and thus profit). But think about the big winners: Amazon, eBay, Google, Skpye. None of them is based on a super innovative idea. The lesson we can learn is that you don’t have to be smarter than everyone else to win. Rather, you have to do something really good to win. Amazon focused on selling books, after it has been well established, it started to sell other goods. Many other websites who rushed to build online retail channels failed. eBay focused on C2C auctions, and it lived out Yahoo’s fierce competition. When Google came out with the "pagerank" idea, Yahoo was the dominant player, it would seem funny if you tell people, in 2000, that Google will be sharing the market with Yahoo. (back then, there are so many other potential competitors: HotBot, Excite, Infoseek, lycos, etc. etc. – how many of these can you remember now?) When Skype came, Yahoo messenger and MSN both offer free VoIP services, anyone dipping the toe into this mud should have thought it twice, yet Skype succeeded. All these examples tell us one simple thing: do what you are best at and be excellent in it.

Scientists, be on guard … ET might be a malicious hacker

Tuesday, November 29th, 2005 -- By ET

Long time no posts.  Here is something fun and new: 

Scientists, be on guard … ET might be a malicious hacker

In the next issue of the journal Acta Astronautica,  Richard Carrigan, a particle physicist at the US Fermi National Accelerator Laboratory in Illinois suggests that SETI@Home’s data may contain virus and cause problems for people running the SETI@Home client. 

This is interesting, but I guess Mr. Carrigan does not know too much about the architecture of modern computers.  To cause a serious problem for people (or steal information from these computers, as implied in the article), the aliens really need to be very good at programming for Microsoft Windows, and know a lot about the underlying hardware running the OS.  Back in the early 1990′s, I got my first computer, and I spent a lot of time cracking DOS programs.  The tools I used are PCTOOLS (I guess it was produced by a competitor of NortonTools), and SoftICE.  I use these tools to reverse engineer programs and I literally collected samples of virus programs, and learns a lot about hacking and propogating programs among computers. :-)

I learned assembly language to write small programs and these techniques all paid off later when I wanted to

  • decrypt some encoded files
  • localize English version softwares to Chinese versions
  • find a quick cheat to finish some very hard computer games
  • recovering lost data from a hard drive failure

Oops, off topic again.  Back to the article, if the aliens have a way to sneak code into these signals and contaminate the SETI clients, and then sends information back to the space in some way, they do not really need to take the trouble in the first place.  Why not just intercept the information transmitted on earth directly?

 

Artificial artificial intelligence

Monday, November 7th, 2005 -- By ET

The title of this post is correct, I did not duplicate the word "artificial" by mistake.  Artificial intelligence (AI) refers to the process of using computers to simulate a human being.  Computers can now do a lot of things better than human beings thanks to the AI research in the last 20 years or so.  AAI is the reverse of AI, it leverages human being’s intelligence to improve a computer’s performance.  For example, if asked "is there a letter A in this picture", human beings can do much better than machines. 

This inability of computers recognizing patterns in a figure has been used by websites (including mine) to distinguish human beings from computer web robots.  The idea is to display some distorted graphical text for human beings to recognize.  In order to get a computer recognizing the letters, a lot of AI work has to be done.  This is called "Captcha".

Talking about ideas of using human beings to empower computers, I always wanted to establish a website to systematically display those questions hard for computers to do but easy for human beings to do.  Any human being can browse the questions and answer some of them and get paid, computers then can take the result and do what they do better (storage, processing, retrieval, etc…) 

I’m glad that Amazon has started to do this in what they call "Amazon Mechanical Turk". Basically, they pay from $0.03 to $0.65 for people to do simple tasks as identifying objects in a picture or to write a product description.  They are all practical problems they have in the Amazon search engine, A9 or Amazon the website itself.  This is an exciting first step toward what I imagined.  There are a lot more to be done, though.

For example, this website only allows people to work on Amazon problems, it would be nice to see more websites to offer similar tasks to people (I call this "mini-outsourcing"). Even better, it would be nice to see some central clearing house to offer such a service so that companies can automatically post questions and payment information, and individuals can go to a single place to earn some easy extra money.  If I were Amazon, I would also establish a reputation system to control the quality of the mini-tasks.  In the back end, I would also run all kinds of data analysis to use the human input to improve the machine learning results so that after some training, the computer algorithm can do a better job in identifying the objects.

How about more complicated tasks?  Like assessing the quality of a novel, a research paper… I believe the key lies in the right incentives for people to participate.  In the Amazon case, it has to set the payments correctly so that some people could come and do the job and still feel happy to be paid; in the more complicated tasks, maybe non-monetary incentives works better, let me think about it… 

Trace me through Wi-Fi

Sunday, November 6th, 2005 -- By ET

This is not a figure of nebula.  It shows the wireless usage at MIT campus.  MIT has upgraded the wireless network through out the campus.  Now you can have a look at which bulidings have the most laptops.  The red area on the right should be Sloan school.  It’s tempting to list my name through the link at iSpots, but then I have the privacy concerns.  Why would I allow everyone in the world to potentially track where I am?

On the website: New wireless communications technologies are changing the way we live and work. This fact is particularly evident at MIT, thanks to the presence of two conditions: 1) the very high percentage of laptop computer ownership on campus; and 2) the existence of one of the most pervasive wireless Internet networks on earth, which includes over 2,800 access points and was completed at the end of October 2005.

 


BlogTimer
You are visitor number several since September 1, 2001

Copyright Xiaoquan (Michael) Zhang, 2004-2020. All rights reserved.
All trademarks property of their owners.