I was watching my friend Robert Scoble’s show where he was getting a demo of the latest version of Bloglines from the General Manager of the service, Eric Engleman. When Mr. Engleman was asked how he describes feeds to those who are yet unaware of them and he replied:
"They are like a time machine…"
Do me a favor and watch Robert’s show (you can watch it on his blog post without downloading anything) so that you can see that statement in clear context.
We all struggle to create simple definitions for complex technologies, but time machine? He makes a good point, in that feeds can really save people a lot of time, because they save the need to actually load every single site the person wants to read on a daily basis. But everything  that I know tells me that when you try to define something, it should be clear, concise, and not leave anything to mystery, or the need of further definition.
Pick up any dictionary and look at the definition of a word. Pick one, there are tons of them. Every definition offered for each word in that book leaves no question as to the meaning of the word. Even fairly complex words that have meanings rooted in history so far back that some of the origins have already been forgotten – yet someone has taken the time to come up with a succinct way of stating exactly what that words means. Sometimes, a word means different things – perhaps even to different people – and so the dictionary offers nearly all of those as well.
So what about feeds? What is a good way to describe feeds to those yet unaware of them?
How would I know? Two years ago (almost to the day) I gave a really horrible definition to what feeds are that I, for no reason whatsoever, called the unified feed theory:
"A “feed†is a way of syndicating almost any type of content (be it articles, essays, plain or rich text, images, audio, or video) in a simple and portable way. Feeds are generally used to distribute frequently updated content, such as news and notifications. An example of this would be CNNs feeds, which provide headline feeds on various topics to keep their readers up-to-date with news. To take advantage of any feed on the Internet, you will need a “Feed reader†or a Web site service that is capable of reading these feeds. These feed readers appear very much like an email application, in essence you’re being ‘emailed’ every time one of your “subscriptions†update their content. Many people who have many Web sites they visit regularly, use feeds to be notified of these updates rather than visiting hundreds of Web sites each day."
I’m amazed I even added punctuation! But, live and learn, right? So as I heard Mr. Engleman say "time machine" I thought about how, two years later, I’d describe feeds to someone. I think I’d say something utterly simple like:
"Feeds are kinda like email. When a web site is updated you’re notified of the change without needing to visit the site."
This description does not describe how feeds work, that you’ll need an aggregator, or even give an example use really – it is straight forward and is built upon knowledge that most people already have. If someone wants to know more, maybe they’ll do a little research or ask for more information. But I think it portrays the idea pretty easily. While my crude definition from late-2005 is accurate, it is a pain to read and long winded.
Two years from now I’m sure that I’ll be able to describe feeds even more succinctly. But I look forward more to the day when feeds no longer have a learning curve at all, sort of like email. You sign up to an ISP and you’re given an email address and a, well – maybe a Bloglines account.
Hey, Mr. Engleman – ever think about getting every single person that logs onto the Internet to use feeds the moment they connect to the Internet by automatically setting up their accounts for them? Something to think about. Kill the learning curve by setting it up for them.