DK the Human

The Battle for PrincetonFML

For the uninitiated, PrincetonFML is an FML site directed towards the Princeton community. It was launched as one of many college FML sites created by a Harvard student under the company CollegeFML. (Ironically, Harvard is one of the few schools that doesn’t use a CollegeFML FML site.) As anyone can see from visiting or, both websites (and all the other CollegeFML sites) are down. I don’t think anyone really knows what happened.

When the website went down a week or two ago, the impact of its absence was noticeable in the Princeton undergraduate community. The topic of its breakage was brought up in numerous conversations, and that’s when I realized that I had to make a clone.

I quickly nabbed the domain and got down to work. Even though we were nearing exam week, making this website became my top priority. Why? I knew that the longer I took to make the site, the more people would get used to not having an FML site – and in the worst case, someone else on campus would get the same idea and make their own replacement.

And that’s exactly what happened. A day into working on the site, I saw this post on Facebook:

Facebook post on new PrincetonFML

Fuck. Someone had beat me to it. The page took forever to load and the upvote/downvote functionality didn’t work, but it was something. I flipped to release-as-soon-as-possible mode because I knew that the longer I took to make my version, the more the other site would gain popularity without competition.

Six hours after I saw the post, I typed ‘git push heroku’ and went live. My upvote/downvote system let anyone vote multiple times if they refreshed the page and I had slapped on Disqus last-minute to allow visitors to comment, but it was something. I posted the site as my Facebook status and waited for visitors. Over the course of the night, more than 700 visitors came to check it out. Not bad. I started feeling pretty confident and thought that being second to launch wasn’t so bad after all.

Then things took a turn for the worse. The Daily Princetonian, Princeton’s student-run newspaper, had caught wind of the other site’s launch and had released a piece on it being the “new PrincetonFML”. Damn it. I was paying the price of not being first. Although a bit discouraged that the other site was getting a ton of free publicity, I was determined to make my site the superior option. After all, the other site was just an exact clone of the previous website – a simple theme slapped on top of a Wordpress install. I had coded mine from scratch using Ruby on Rails. That had to count for something, right? People would care that my site loaded in the low hundreds of milliseconds whereas the other one loaded upwards of five seconds, right?!

Some of my readers are probably laughing by now, because they know the truth: users don’t give a shit. As long a website is Good Enough, they’re highly unlikely to switch to anything else. However, despite the fact that the Prince (Daily Princetonian) featured the other site, my site continued receiving healthy amounts of traffic. It definitely helped that the first comment on the Prince article was a link to my site.

So where am I four days out? Today, someone submitted this as their FML post:

No one cares about this website; it clearly lost the traffic battle to the other pfml website. Being stubborn != successful business model #pivot

This was obviously not an FML submission, but someone trying to send me a message. Over the past couple days, I’ve gotten used to receiving hate posts (examples: “This site is not impressive. FYL” and “This site is pretty lame. FYL, mods.”) I found posts of this nature so amusing that I accepted them and had them displayed on the site. Haters gonna hate.

But this latest post was less an attack on the website itself and more a claim about something I’ve put a lot of thought into: which site will become the de facto PrincetonFML. Ultimately, only one will remain.

Is the claim in the submission true? It’s entirely possible that the other site has been receiving more traffic after being featured on the Prince. All I know that I’m still getting hundreds of hits, and the number of users is growing. Am I losing money hosting the site? No. (Thanks, Heroku!) Is this a business? Hell no. I built something for fun, and I’d say that it’s been fairly successful. If I’m stubborn about anything, it’s about making my site the best it can be. I have a pretty big vision for where a site heavily frequented by the Princeton community can go. So if you’re planning on telling me that my website sucks and that it’s not going to be a successful business, I’d like to kindly tell you to fuck off.

I don’t know whether my site will continue to be successful and grow or if the other site will end up winning this battle to replace PrincetonFML. If anything, I’ve gained experience and had a ton of fun seeing this go from conception to launch in a few days. Here are my major takeaways:

  1. If you want to make something quickly, a great option is Rails + Bootstrap + Heroku. You can create a functioning, presentable website in mere hours.
  2. Launch early and improve quickly. Even though I originally had Disqus comments that required users’ emails (users hated it) and my voting system was broken, I am glad that I launched when I did. If anything, I wish I had launched earlier.
  3. Fuck the haters. Don’t let anyone discourage you when you’re putting in your hard work to make something awesome.

Update (5/29/2012): CollegeFML and its sites have been restored.