When I read that Google Reader was being shut down, I had a mild panic attack. And I never bat an eyelash at web services being shut down, bought out, or otherwise decommissioned. I loved del.icio.us and when it was bought out and dismantled, I built my own replacement. When I got really frustrated at how needlessly complicated WordPress truly was, I built my own replacement. When I thought maybe it'd be funny to ditch the internet and build my own, I prototyped it out.

But Google Reader is something I use almost every hour of the day, at work and at home, on every computing device I own. I absolutely love the Reeder client for iOS and Mac. It's one of the very few apps that I've ever shelled out money for and thought "damn, this is worth it". But while I normally read through feeds on Reeder, I don't mind the web interface for Google Reader at all. It's simple, minimal, and shows me what I want to see, in a format that makes sense. I read hundreds of articles every day via RSS. The interface needs to be simple and it needs to work well.

I love RSS because it's what I've referred to as a "higher-level standard" that only makes the web more fun, accessible, and extendable. The RSS and Atom standards are pretty easy (although somehow lots of people fuck it up). After all, RSS feed aggregation itself is simple. It's just not scalable. Very quickly you have thousands and thousands of posts being stored locally on your server, growing to millions as your user base grows. And fetching and parsing all of those feeds is expensive. But it's worth it — the alternative is visiting all of those blogs/sites/whatever one at a time, manually checking for updates, having nothing to keep track of what you have and haven't read.

So I built FUCK READER. It took a couple of days. Here's what it looks like right now:

It's insanely fucking simple, as the masthead implies. I hate social media sharing features. They're clutter. I can copy and paste a link to Twitter if I really need to. Here are the features I really cared about, and are present in my reader:

That's really it. I have some more features that I care about, and I'm working on them, but RSS readers don't have to be more complicated than what I just laid out. Here are some more advanced stuff I'm going to try to incorporate:

Pretty simple stuff. Again: it's a goddamn RSS reader, there's no crazy social web 2.0 crap to go on here. I want to read my fucking feeds, that's all.

The Technical Stuff

The front-end of FUCK READER is built in PHP with a little jQuery for some tricks. It's stupidly simple. A MySQL database stores the feeds, posts, users, and state info. I can already think of ways to use Raik or MongoDB to store things better. The feed fetching mechanism is all Ruby, primarily utilizing feedbin's fantastic feedzirra gem. I had tried building my own feed fetching-and-parsing in PHP, Node.js, and Go, but gave up because many RSS and Atom feeds are malformed or otherwise extremely difficult to parse.

I have not open-sourced FUCK READER because I just don't really want to yet. I want to hammer out a few things before I make the code public, but I definitely will. I have no plans of letting everyone use my instance of it, which I'll get to later.

The Competition

Of course, when I saw that Google Reader was gonna die, I sought out alternatives. However, none of them appealed to me. They don't seem to appeal to many people. Here are some thoughts:

I've spoken to several people who have tried out these or other ones on the many lists of Google Reader alternatives, and most people have walked away from them disappointed. I'm starting to get some pretty positive feedback from people about mine, so I'm happy so far. But ultimately, I'm just trying to please myself with it.

Sustainability

Right now if I made FUCK READER totally open to anyone, it'd be unsustainable very quickly. I'm currently running it on my Linode box, which has a few sites on it sharing the same server. I've elected to make FUCK READER free, but invite-only, to share among friends and a few interested parties. I have no intention of letting it run amok on Hacker News or Reddit or some crap like that. I'll open source it, eventually, if people want to install it somewhere for their own use. Unless somebody wants to give me some money to spin up another Linode instance for it.

So how do you get an invite? Ask nicely via @cylegage, maybe.