As you are no doubt aware, Google Reader is going away. The public outcry hasn’t brought it back, nor open sourced it, by now so that’s not going to happen. There are some options, but I decided that I wasn’t going to switch until May 15th. That will give the smart people an opportunity to vet the options and tell me what to do. I haven’t tried the options, obviously, so I can’t comment on them. Instead, I want to reflect on what I want from an RSS reader.

But first, a word about RSS. I’ve read and listened to a lot about Reader’s demise lately. A lot of people are characterizing RSS a a two-faced beast (most notably John Gruber, but I can’t remember if it was The Talk Show or some podcast on which he was a guest). One face is the user-interface like Google Reader. It’s how I read web pages. I almost never read a website that doesn’t have a feed. In fact, I read less than a handful of sites that don’t publish the full content in their feed.

The other face is this back-end plumbing, this architecture, this infrastructure that is RSS. How you read websites is a personal choice. But whether you pop over to DDoE to see what’s new or you check a feed reader, this site still publishes an xml file. You could write some software that parses that file. You could buy some software that does it. You could do something interesting like the Spreadsheet Page’s Excel Blog Headlines Page. The XML file is out there. Use it, don’t use it, but don’t get rid of it. It has value. Someone needs to build an interface that gets the masses interested in reading pages from a single source. Just don’t call it RSS and it will be fine.

What do I want from a reader? Here’s what I love about Google Reader:

• Great syncing – Google updates quickly. Really quickly.
• No duplicates – Google does a good job figuring out which posts I’ve read and which I haven’t. There have been a few glitches along the way but all-in-all it’s been solid.
• Good rendering – For the most part, the feeds render the web pages really well. I hate the way it shows DDoE code, but other than that, it’s good.
• Shortcut keys – Google respects the mouse-a-phobes and provides keyboard shortcuts for easy navigation. I can learn new shortcuts, but if they’re not there it’s a deal breaker.
• Ubiquity – I like Reader on the web, in Chrome for iOS, and everywhere else I’ve used it.

Here’s what I don’t like:

• Screen Real Estate – Nearly 1/3 of the screen is used up by non-value added bullshit. Some of that is Firefox, but everyone has to do their part.
• No sorting – My folders are sorted alphabetically, which is the just about the worst way you could organize them. Let me assign a number to each folder that determines the order. Some feeds I read no matter what and some I read when I get the chance. I want the former at the top.

Every feed that I read falls into one of four categories.

News
News is the least important, the highest volume, the most skimmed (as opposed to read), and the least missed. I have some news feeds, but if they ever amassed too many unread entries, I have no problem blowing away large swaths. Some of my news feeds are:

• KETV.com
• AP Top Headlines
• A couple Reuters feeds
• Stack Overflow feeds
• Netflix new releases

You see that they don’t have to be news in the traditional sense of the word. In fact, I should stop calling them that. Their most defining characteristic is that I don’t care if I miss something.

Content
Content are feeds I read because they’re damn good. If I thought I missed something through one of these feeds, I would go visit that site to make sure. I’m going to read every entry posted to these feeds. That may be because I know the feed has short entries that take no time to read (web comics) or the content is just so good that it’s worth my effort to avoid missing any entries (kottke). Content feeds include:

• Family memeber’s blogs (let’s call this content ‘highly relevant’ rather than ‘good’
• Web comics like xkcd, Oatmeal, Savage Chickens, Dilbert, NIH, Scenes from a Multiverse
• Signal vs. Noise
• Street Smarts

Not only do I not want to miss a single post, I want to read these right away. When xkcd publishes, I don’t save that gem for later, I read it right now.

Obligation
Some feeds I read out of obligation. I feel it’s my job to know what’s happening in the world of Excel, so I read a whole crap load of Excel blogs. Some are great and some aren’t. If it’s not a blog that solely exists to sell a product, I’m probably reading it whether I find it currently useful or not. I also read feeds about manufacturing and petroleum. I’m not going to miss a post on these either, but I don’t need to read them today. Mike’s Data Explorer post was awesome, but if I read it five days after he posted, it would still be awesome. I’m not going to list all the Excel blogs I read, you can read my OPML file if you’re so inclined.

Saves
Saves are either long reads or things that I wish I were interested in. The Harvard Business Review has some nice articles, but I save them for a rainy day. Same with Kahn’s Corner, Lowering the Bar, and a host of other sites. I may get to them someday or I may not. But I want them there.

My reader should have two panes. The left pane holds Content feeds. The ones I want to read right now. The right pane holds News feeds. It will be pretty full, but I’ll blow through them in no time. When I have no content feeds, the left pane turns to Obligaton feeds. The Saves feeds never show up unless I specifically ask. I can put any single post into Saves while still keeping the feed in another category.

I can set an expiration on any pane. I may say, for instance, that entries from News feeds go away if I don’t read them for five days. For me, personally, every other pane will be set at ‘never expire’, but others may want their Obligation entries to fall off after some time. The panes show all of the unread entries from the feeds assigned to them. If you want to see entries you’ve already read, you can search for them. Entries can be shown in chronological order or the reverse. Feeds can be rated 1 to 5 stars. A combination of age and rating determines what’s on top in a pane. The feeds inherit all these features from their pane, but can be overridden on a feed-by-feed basis. For example, I read almost everything chronologically so that’s how my news pane will be ordered. A notable exception is Stack Overflow, where I prefer to see the newest (least likely to have been answered) first and I will set that feed to be reverse chronological. All other feeds in the News pane will inherit the chronological property from the pane.

So there you have it. Give me two big panes, keyboard shortcuts, effortless syncing, and a semi-consistent experience across web and iOS and I’ll be fine. Oh, one more thing: I’ll pay $100 per year for the service.