Thursday, July 30, 2009

Even a bad recession is good

I was reading the article Entrepreneurship during a slump by Tim Draper.
A nice read. The author cites companies like GE, IBM, Microsoft, Shell Oil, AT&T, Merck, Johnson & Johnson, Sun Microsystems, Skype, Kodak, Polaroid, HP, and Adobe which were all started during economic downturn.

A very short summary of the article:
According to the author, recession is the best time for starting up companies and takes it positively and states that recession is good :) .
The reasons being:
1.Managers think creatively.
2. During recession a long-lasting frugality is built as a culture in t
3. Enterprenuers dont face “venture fratricide.”

Here is a sentence(verbatim) I liked in the article.

That is my advice for entrepreneurs in these times: If you take an entrepreneurial risk, make sure you go after something big. Extend your imagination. Think flying and self-navigating cars, holo-decks, brain enhancers, salt water purifiers, fusion energy, and space travel.


Another short counter by him.
Bad news is good news to the press
The article is from Communications of the ACM, August 2009.

Is Google search always the best????

I was searching for a basic tutorial on threads in Java. These days I use Bing for my search. But, this time Bing didn't give satisfactory result. As usual, I tried my luck on Google and the results were no better than Bing's. :(

Finally, I tried if Yahoo search could be of any help. and yippie.... I could find related content. :)
This has been my experience even before, whenever Google fails to show relevant results, I would try them on Yahoo and the search results on Yahoo! would surprise me. Results were more relevant which I never saw on Google's first few pages.


Now this raises a question in my mind... is Google search always the best?
When is Yahoo better than Google?

Hope the new deal between Yahoo and Microsoft brings some change in the search market.

Anyways, the tutorial by Brian Goetz on java threads is good. You can get a pdf version here.

This tutorial is presented by developerWorks.

Friday, July 24, 2009

How Robots.txt can be useful sometimes.

What's new from Google? No doubt, most of the internet users would be eager to click any article with such heading or text.
Indeed, I am one among those. ;)

At this point of time, I recall a moment when I won a bet of good food. I was able to gain access to one of my friend's yahoo account 6 years back. This was out of sheer luck that I knew his personal information and the security question was very easy to answer. Yeah, that was it. This friend of mine recollected the incident few years back and that was the time when I got the idea of having a security check for such hacks.
My idea was to record the previous login time (And make it non-editable).
Finally, Google has come up with a similar feature in gmail(adding login time, ip address and time spent). Storing the ip address is too good an idea. This would actually give us an idea of the location where the hacker is located. A useful feature enough to make sure that no one else is reading your (personal) mails.And this is one reason why I like gmail.

Well, coming to the actual point. I was actually searching for a parameter in the robots.txt to set the maximum number of scans a bot can perform on my site. Request-rate is the required parameter/setting I can use.
And while I was reading few articles, Google's robots.txt http://www.google.com/robots.txt caught my eyes. Google allowing most of its content not be crawled surprised me. But, this was no waste an effort. I came to know of Google Ventures , useful for entrepreneurs..
So robots.txt gives us some useful information too(if not for robots).
Perhaps, I may be gaining some useful information from other site one day.

Many lines in Google's robot.txt but with a Disallow tag prefixed.
No harm, I am no robot to be disallowed :P

Wednesday, July 22, 2009

Looking for Internship in France???

Xerox Research Center Europe is part of Xerox Innovation Group with over 550 researchers and engineers and is working in the areas of Parsing & Semantics, Machine Learning, Large Scale Data Mining, Textual & Visual Pattern Analysis, cross-language technologies, statistical analysis, xml, visualization or software development.

Organization: Xerox Research Centre, Europe.
Location: Grenoble.
Working Language: English
Employees and Researchers: ~ 550
Website: www.xrce.xerox.com

The center also has internship positions for Master's and PhD students in mathematics, linguistics, psychology.
http://www.xrce.xerox.com/internships/home.html

A good news for Indian students :)
Xerox also has Open Innovation partnerships with institutions in India.
More details about this are available at:
http://www.xrce.xerox.com/internships/home_India.html

There are few demos available on their site
http://www.xrce.xerox.com/competencies/content-analysis/homepage.en.html

The language guesser/ identifier demo is appreciable effort and it would be good if such work is done for Indian Languages too. :)

I feel XRCE is a good place for students pursuing their Master's degree to do an internship.

PS: Please update me of any broken links in this post.
PS2: The information provided is subject to change. Please check the main site for more details. :)

Monday, July 20, 2009

Installing Flash Plugin for Firefox 3.5 on FC11 x64

Three simple steps for installing adobe flash player for Firefox 3.5 on FC11 x64

1) Download the libflashplayer tar ball (libflashplayer-10.0.22.87.linux-x86_64.so.tar.gz) from Adobe website.

2) Uncompress the tar ball

[praneeth@inferno ]$ tar -zxvf libflashplayer-10.0.22.87.linux-x86_64.so.tar.gz

3) Now copy the file libflashplayer.so from the current directory to /usr/lib64/mozilla/plugins/

[praneeth@inferno ]$ cp libflashplayer-.so /usr/lib64/mozilla/plugins/

4) Restart Firefox :)

Saturday, July 18, 2009

H5 Technologies

I was browsing and found this company that works on providing solutions for IR and document analysis for legal departments and other law firms.
They have offices in SF and New York. They also have an office in Mumbai.
The work is interesting and here is a brief description of the company.

Location: SF, NY
Work: IR and Document Retrieval for law firms and legal departments.
Areas: NLP, IE, IR, Linguistics.
More about the company at http://www.h5.com/about/index.html

Here is a short description of the company:

H5 is the leading provider of information retrieval and document analysis
services for Fortune 500 corporate legal departments and leading law firms.
H5 was recognized with a number three ranking among the fastest growing
technology companies by Deloitte's Technology Fast 50 Program. H5, a
privately held company with strong venture capitalist backing, was also
included in the "Cool Vendors in Content Management, 2007" report by
premier analyst firm, Gartner, Inc.

H5 is an information retrieval firm that helps law firms and corporations
search, assess, and manage electronically stored information. Through a
full range of advisory, document review, and litigation support services,
H5 finds the information clients need for litigation and investigations,
compliance, and litigation readiness. Our approach - which combines
advanced technologies with expertise in law, linguistics, computer science,
and statistics - consistently reduces clients' information management costs
while minimizing their risk.

Employment opportunities with H5 are differentiated by our unique value
proposition. We offer an environment where you will partner with a diverse
team of sophisticated knowledge workers who share a drive to succeed, a
passion for solving our clients' most challenging problems and a strong
intellectual curiosity. Our expectations for performance are high and each
team member has a clear line of sight connecting their individual
contributions toward the achievement of department and corporate
objectives. For more information please visit the website listed above.
http://www.h5.com

I would personally classify this company as NLP, IE/IR company.
Natural Language Processing /Information Retrieval because document analysis would require a bit of NLP.

The work is interesting because it would reduce a lot of manual effort. Let me know if you come across more information about the nature of work.

On-Demand Webinar: Scaling Hadoop for MapReduce Applications

Google's map reduce mechanism has had a considerable effect on computing.

https://dct.sun.com/dct/forms/reg_us_2005_941_0.jsp?

Hadoop, an framework in java for carrying out distributed processing is gaining importance. The framework is specially designed for data intensive applications.
Hadoop is inspired by Google's Map Reduce

Organizations like Facebook, A9, Powerset, NYTimes use Hadoop for distributed computing.

Hadoop is created by Doug Cutting, . He also contributed to Nutch, an open-source search technology.

Friday, July 10, 2009

A smart tutorial by Smart on Lucene

Found this quick introduction to Lucene by Smart.

http://www.informit.com/articles/article.aspx?p=461633

He explains the three basic steps involved in using the lucene library.

1. Creating an Index

2. Indexing an Object

3. Full-Text Searching

Wednesday, July 8, 2009

Changing default Java in Ubuntu

Just perform these steps to change the default Java in Ubuntu. I have been using the absolute path all these days...... but now I need not do that anymore :)
1. Check the current version of Java using the command

java -version

This may result in some output of the form.
java version "1.6.0_06"
Java(TM) SE Runtime Environment (build 1.6.0_06-b02)
Java HotSpot(TM) Server VM (build 10.0-b22, mixed mode)

2. Check the available Java installed in your system.

update-java-alternatives -l

java-6-sun 63 /usr/lib/jvm/java-6-sun
java-gcj 1042 /usr/lib/jvm/java-gcj


3. Now select the Java version you want as default.

update-alternatives --config java
There are 4 alternatives which provide `java'.

Selection Alternative
-----------------------------------------------
1 /usr/bin/gij-4.2
+ 2 /usr/lib/jvm/java-gcj/jre/bin/java
* 3 /usr/lib/jvm/java-6-sun/jre/bin/java
4 /usr/bin/gij-4.1

4. Enter you Java selection number.

Thats it! You are done.........
Now check the new default java using

java -version