Friendster goes PHP

An item I saw yesterday but forgot to blog about: Friendster goes PHP. Pretty cool.

Finally on Friday we launched a platform rearchitecture based on loose-coupling, web standards, and a move from JSP (via Tomcat) to PHP. The website doesn’t look much different, but hopefully we can now stop being a byword for unacceptably poky site performance.

I haven’t had much of a chance yet to use Friendster to see if it truly is faster, so I can’t personally comment on that aspect. And predictably, this is going to bring all sorts of people out of the woodwork arguing over the relative merits of Java/JSP (which was old Friendster) versus PHP… just look at the comments on the link above to see it already happening. And while debate and disagreement can be healthy and productive, how about a quick reality check to everyone:

PHP is good. Java is good. Both have their merits and disadvantages. Loudly complaining that [Java|PHP] is the only true way and the other is crap is boring and uninformed.

More Friendster Notes

I’ve noticed from the referrer logs that my earlier Friendster post is the #3 result on Google for the search phrase “Friendster is slow“, so I figured it was high time I revisited Friendster and poke around a bit more, to see what I could find out.

It was still slow, but not as fatally slow as the first time I was playing with it. I had previously created a profile for myself and uploaded a picture, but I had not invited friends to join. I was curious to find out if I could use Friendster without any friends (irony! irony!), and the answer is “yes,” albeit conditionally.

About the only thing you can do when you don’t have any friends—apart from inviting some—is search for other users. However, I’ll save you some time on that right here: you can only search for users from your personal network—that is to say, friends of friends of friends (ad nauseum); if you don’t have any friends, and by extension no network, then you’ll always end up with 0 users found on the search results.

This wasn’t obvious to me from the way the site was set up, but for sake of argument let’s say I’m socially retarded and overlooked the fact that a site that’s designed to network among friends wouldn’t naturally let you search for strangers… anyway, maybe it was obvious in hindsight and I missed it. Moving on.

I invited some friends. Five that I could think of that (hopefully) wouldn’t think I was too weird in sending them emails inviting them to my Friendster network. Okay, nothing to do after that but log off and wait.

A little while later, my brother had registered with Friendster and suddenly I had a friend! But then I ended up asking myself, “What now?” There still wasn’t any obvious benefit to this system that I could see.

Then, later in the day, another friend registered on the site (I got an email notifying me of this). Didn’t have time to check it out at the time, I was heading home from work. Also didn’t figure there would be any more to do with Friendster with two friends instead of one, so it wasn’t a big priority.

But by the time I logged back into it from home, my jaw dropped: I suddenly have 400 people in my personal network! It turns out my brother linked to two more friends, who in turn link out to friends, who link to more friends, etc. Very six degrees of separation.

Now I can see the value in what’s going on here. I have access to a network of people that I can browse, search (by demographics or by interests), contact. (Noting, of course, recent stories about how a lot of Friendster accounts are fake as people assume different identities online or are just playing around.) Very interesting. I haven’t decided what I’ll actually do with Friendster yet, aside from figuring it out.

Here’s something interesting, though: When I logged on and found my network of 400 people, Friendster seemed to run faster than when I had none. This is counter-intuitive; it should run more slowly when it’s sifting through larger data sets (ie. larger networks). The only thing I can figure is that their data queries are either highly unoptimised—perhaps brute-force searching through all the users to find out none were in my network?—or when dealing with zero-user networks (ie. no friends), the database/system/whatever is dealing with NULLs improperly. And any good database tech can tell you that NULLs can be a killer. It’s very odd.


I signed up for a new online tool/technology today called Friendster. Maybe you’ve heard of it; it’s “an online community that connects people through networks of friends” for meeting new people. So far I haven’t really figured out what it’s supposed to do for me, because the site is still very much in beta: most of the pages were slow-loading, the people search didn’t give me any results (they’re rumored to have 300,000+ members, so I’d expect some results), and the site just stopped responding to me after several minutes of use each time I tried.

Perhaps the slowness is due to increased exposure to curious users after the write-up it got in Wired by Xeni Jardin (that’s such a great William Gibson-esque name), though I doubt it. I’ll play with it some more, and report what I find.

Interestingly, what got me to Friendster was a link on Robert Scoble’s weblog for, which is another beta social/community website that’s making the rounds—and Xeni Jardin (there she is again) on Boing Boing talking about and writing that she won’t be “ditching her Friendster account anytime soon”—all of which made me curious. So I’ll probably go and try out now, too, because Friendster is slow.

How’s that for making connections and providing links? I think the ultimate social software application is the blog.