Author Archive for Abhinav

Why isn’t the internet hosted in Korea/Japan?

Published
by
Abhinav
on April 18, 2008
in Development and Internet Lurker
. 0 Comments

South Korea and Japan are world leaders in terms of broadband penetration, costs and bandwidth. What surprises me is why most of the world’s hosts are based out of the United States.

Now, the cost of an internet connection is not just bandwidth costs, it includes hardware, technical expertise, etc. but even considering that, it might still be worth it to invest in Korea/Japan.

gTLD Distribution

I understand costs for consumers and corporations will be different, but country-to-country, they run parallel.

  1. United States, 67.23% (52,277,677)
  2. Germany, 5.71% (4,442,041)
  3. Canada, 3.60% (2,802,411)
  4. United Kingdom, 3.37% (2,617,679)
  5. China, 3.22% (2,503,430)
  6. The rest combined, 16.86% (13,111,810)

This is virtually a U.S. monopoly. Compare this with the following chart showing average download speeds across the world.

speed matters

Seems to me like one part of the internet is still stuck in the non-global 1990s. Look out for fireworks in this industry if some people are willing to make bold investments.

Optimizing Advertising for Social Networks

Published
by
Abhinav
on April 17, 2008
in Facebook, OpenSocial and Social Media
. 4 Comments

This comes from an interesting post by Jesse Farmer. There is no clear winner amongst ad networks for social applications on Facebook, MySpace, etc. The only real way to know is to find out for yourself.

Using PHP, you can randomly show ad codes from multiple ad networks with equal probability. Check the cash and find out which works best for you!

function get_random_ad_code() {
    $ad_codes = array(
        'lookery'     => 'Your Lookery ad code',
        'adblade'     => 'Your AdBlade ad code',
        'socialmedia' => 'Your SocialMedia ad code',
        'rockyou'     => 'Your RockYou ad code'
    );

    return $ad_codes[array_rand($ad_codes)];
}

echo get_random_ad_code();

If you want to take it to the next level, you can write more complex algorithms to weigh the ads in favor of the network that works best for you, but that’s outside the scope of this article (for now.)

OpenSocial Video Tutorials

Published
by
Abhinav
on April 15, 2008
in APIs and Libraries, Development and OpenSocial
. 0 Comments

After researching the possibilities of porting my Indian Premier League application from Facebook to Orkut, I have begun aquatinting myself with the basics of the OpenSocial platform which Orkut uses, and these video tutorials provided by Google are an excellent tool.

Continue reading ‘OpenSocial Video Tutorials’

Orkut vs. Facebook in India

Published
by
Abhinav
on April 15, 2008
in Facebook and OpenSocial
. 1 Comment

Facebook still has a long way to go! I’ve been optimistic about Facebook’s prospects in India, but at the moment, things do not look so bright. I was doing some market research for my Indian Premier League application, and the possibility of implementing it on Orkut. Here are some findings.

The most popular Facebook group about the IPL has 985 members right now. The coresponding group on Orkut has 37,096 members

Orkut vs. Facebook India

I guess people are sticking to Orkut because despite all the trouble, it does the job for them. It seems like I’ll have to be the one who has to move over. Reality bites. Ouch!

Where’s the security, Yatra?

Published
by
Abhinav
on April 15, 2008
in Internet Lurker
. 0 Comments

Hashing is considered to be an essential part of securely storing passwords. Remember the old days, when password recovery resulted in your password being mailed back to your email address? There was a reason why it was replaced by password resetting - because companies didn’t want hackers to find out their customers’ passwords in case their databases were breached.

Yatra.com is a Web18 venture and one of India’s most popular travel sites. Now, I like Web18 websites, mainly because they conform to the latest web design trends, but seeing that the password recovery feature mailed back my original password in plain-text makes me wonder that they have a very loose back-end. One breach, and hundreds of thousands of passwords are instantly exposed. What makes it worse is that most people use the same password across multiple websites.

Try out the recovery feature and see for yourself. I’m suspecting Yatra has a very, very weak back-end. That’s the last I’ll be using them for a while.

Human Computation

Published
by
Abhinav
on April 14, 2008
in Computer Science
. 0 Comments

After my previous post about reCAPTCHAs, I looked further into Luis von Ahn’s work and came across his PhD thesis on human computation. Its one of the few most people can understand. It involves tasks that computers cannot easily perform such as image tagging, and packaging it into computer programs like games so that people can perform computation tasks and have some fun too.

Abstract

Tasks like image recognition are trivial for humans, but continue to challenge even the most sophisticated computer programs. This thesis introduces a paradigm for utilizing human processing power to solve problems that computers cannot yet solve. Traditional approaches to solving such problems focus on improving software. I advocate a novel approach: constructively channel human brainpower using computer games. For example, the ESP Game, introduced in this thesis, is an enjoyable online game — many people play over 40 hours a week — and when people play, they help label images on the Web with descriptive keywords. These keywords can be used to significantly improve the accuracy of image search. People play the game not because they want to help, but because they enjoy it.

I introduce three other examples of “games with a purpose”: Peekaboom, which helps determine the location of objects in images, Phetch, which collects paragraph descriptions of arbitrary images to help accessibility of the Web, and Verbosity, which collects “common-sense” knowledge.

In addition, I introduce CAPTCHAs, automated tests that humans can pass but computer programs cannot. CAPTCHAs take advantage of human processing power in order to differentiate humans from computers,an ability that has important applications in practice.

The results of this thesis are currently in use by hundreds of Web sites and companies around the world, and some of the games presented here have been played by over 100,000 people. Practical applications of this work include improvements in problems such as: image search, adult-content filtering, spam, commonsense reasoning, computer vision, accessibility, and security in general.

Here’s the full version of the thesis:

ReCaptcha

Published
by
Abhinav
on April 12, 2008
in Random
. 1 Comment

ReCaptcha is Carnegie Mellon startup that I’ve seen a lot of recently on blogs. I was wondering what the deal is and what makes it different from other Captchas. It turns out that recaptchas use two words instead of one. One of them is a recognized word, while the other is not, what you enter in the latter field helps digitize books.

Smart thinking indeed. I couldn’t find any documentation on how exactly it works, but I’m guessing it uses a probabilistic method based on what the majority of people type in. Sometimes, its easy to guess which word is likely to be the unrecognized word, so it becomes easier to screw around. Moreover, what would happen if the unrecognized word is likely to be confused with another word, there might be trouble with probabilistic methods then?

Considering that, its probably a good thing not many who enter a reCaptcha dont know how it works, or they would probably be too lazy to get the second word right.

Facebook Chat Launched

Published
by
Abhinav
on April 12, 2008
in Facebook
. 0 Comments

Facebook has launched its instant messaging product. However, my personal preference is to keep it disabled. I think integration with friends’ lists is important, as usually people add so many friends on facebook as it is impossible to communicate with all of them simeltanously. Most of my friends have disabled the feature. I think the lacking ingredient is prioritization.

Here’s the official announcement on the Facebook blog.

Human Computer Interaction - The Current Wave

Published
by
Abhinav
on April 4, 2008
in iPhone
. 0 Comments

The first marketed integrated mouse — shipped as a part of a computer and intended for personal computer navigation — came with the Xerox 8010 Star Information System in 1981.

That was the last real revolution in Human-Computer interaction. Twenty seven years down the line, we talk of exciting platforms, Web 2.0, open standards, but ultimately, the vast majority are reading off a screen, typing on a keyboard and clicking a mouse. Fortunately, this decade is pointing the way to the next generation of HCI technology. The current wave of new HCI technologies includes the following:

Advanced Touch - Pointers and Stylus:This started with PDAs a long time ago, and then went into laptop computers with the Tablet PC with better but still imperfect handwriting recognition. With Apple’s use of glass in the iPhone, we can look forward to interaction with bare fingers, and the stylus being relegated to handwriting recognition.

Advanced Touch - Complex Actions: This technology was around in research and prototypes for a long time. The simplest example we knew of two-finger scrolling on Macs. Then Apple bought out FingerWorks and threw this into the mainstream with the iPhone.

Speech: Speech is that technology that has been trying to break in since the mid-1990s, but still hasn’t got much mainstream momentum. Things are changing for the better. At first, we wanted to do everything with speech. Now we’re going back and perfecting basic applications like GOOG-411. With time, speech looks almost certain to play an important part in the big picture.

Motion Sensing: This is another current generation technology that is gathering momentum. Consumer devices such as phones, laptop computers and digital cameras have built-in accelerometers. This is an exciting platform to look forward to, since most current applications are restricted by the manufaturers. It would be interesting to see what iPhone developers have in store for using the accelerometer. Even simple applications like iSnake for the iPhone are exciting.

Visual Motion Sensing: This is where the machine senses the user’s actions by visual aids such as a camera. The hardware implementation is cheap, creating the algorithms to do so is not. That said, some brilliant things have been with this technolgy. The DARPA Challenge is a competition where cars have to be built to negotiate traffic on normal city roads, obey traffic rules and get from one place to another, using IR, video and other tools. In a few years, your home computer is going to be smart enough to do things like this, and when it does, it’ll be nothing short of revolutionary.

Surface Computing: Everyone has their own vision of what surface computing should be like. Microsoft has its own, and is the first tech major to go public with it. Microsoft Surface is a combination of most of the technologies discussed above, and interacts with other devices through wireless networking.

Social Apps - Usage Discipline?

Published
by
Abhinav
on March 31, 2008
in Development, Facebook and Random
. 0 Comments

We developers love it when people use our applications on social networks, but sometimes it can go a little too far, one of my friend’s profile looks something like this:

Continue reading ‘Social Apps - Usage Discipline?’