Fay suggested I play this game to kill my boredom. To prevent further waste of manhours (or womanhours if you’re a girl), I’ll show you the endgame screen so you won’t need to play to find out:

50 levels of insane balloon popping and this?!

Here’s the game if you still want to play:

Continue reading

The Friendster Social Experiment Part 2

Have been very busy for the past week because of exams and real life obligations. Since I got nothing much to write and no camera to take pictures (camera went on holiday to UK), I’ll post a bit of updates on the Friendster experiment.

My profile views have jumped another 5k.

And I’m now officially popular on Friendster! I feel so overdressed and “under-posed”.

I’m still constantly receiving friend requests but I’ve lost count since I deleted all of those mails.

So as a conclusion, you can be popular by writing a piece of software. Continue reading

The Friendster Social Experiment

If you’re from Malaysia, chances are that you have a Friendster account. Or two if you are so popular that you have exceeded the 1000 friends limit in one account. Friendster users are unique because they are from all walks of life and it’s actually a very lively and active online society which holds many mysteries.

If you have Friendster, you must be familiar with the “Who’s Viewed Me?” feature. It shows who have viewed you for a certain time frame (say for the month of September). This is an interesting feature which allows people to track people who clicked on their profiles. This is important because people who clicked on your profile most likely found your display picture to be attractive (since it’s the first thing they see when they look at Friendster users). Or they might know you from a friend.

Anyways, I was dreaming one day when an idea struck me. How many users would actually view your profile after you’ve viewed theirs? And how many of those who viewed you would actually add you as a friend?

As it is impossible to have so much time on hand to view a lot of profiles to gather enough data, I wrote a simple program to help me do it. But before I began I needed a control group. This group consists of female users, aged 16-22, all from Malaysia. Another program was written to help me find profiles matching those criteria. I got about 15k profile matches.

So the task of visiting those profile was left to the program that I have written. In 10 seconds, it visits 12 profiles. Due to some bugs in the program, it took almost 8 hours to visit all those 15k profiles. Still fast. Faster than what a human can do.

Since it took time for the users to respond (some may not login everyday), I waited 2 days before collecting the results. And the results are as follows:


249 friend requests.

As you can see, there is a positive jump in my profile view after “viewing” a large amount of profiles. From the 15k users I viewed, only 3k responded by viewing my profile. But this does not mean that only 20% of the users bothered to view those who viewed them. Some profiles may be inactive, some may have the “Viewing profiles anonymously” option turned on and some may not have logged in during the 2-day period. So it wasn’t a very good way of measuring user response.

But we can say for the 3067 people who viewed me, 8% (249) decided to add me as a friend. Alright, maybe we need a larger group of profiles for better statistics. So I wrote another program that will go to random profiles of people from all over the world and “view” them.

A bit of maths here. Friendster currently has 47 million users. My program roughly visits about 1 profile per second. By doing some rough calculation, it would take me 543 days or 1.4 years to visit all those profiles. Something which I can’t do unless I have 543 connections working at once and visiting those profiles. It would take just a day to finish. Results would be better.

But since I’m not doing a large scale experiment, I just went with smaller samples. I wasn’t sure how many profiles were visited (maybe around half a million). And the results are as follows:

15883 views, including the 3k views from above.

1718 friend requests.

I wouldn’t say this experiment is a great success. There were just too many variables that weren’t controlled properly (inactive accounts, invalid ids, not signing in during the duration of the experiment). The number of views are still increasing and the number of friend requests are still increasing as we speak. But it’s a way to show that just by visiting profiles, you can actually get friend requests.

So as a conclusion, the experiment proved that Friendster users will respond if you make a move to get to know them (by visiting their profiles). This is exactly like real life. You won’t make many friends by staying at home. You’ll only make friends by meeting people. Sadly, I cannot accept all the 1700+ friend requests because then I’ll have to create 2 more accounts.

But the world has changed so much since the birth of the Internet. I think Friendster really appeals to us because it enables us to meet lots of people without ever leaving our desk. Is that good or bad? I guess we’ll see the result of that in another half a century. Continue reading