Richard BF

Having bored you enough with the vague theoretics of controlling the publishing pipeline, which I swore I wouldn’t mention again, perhaps a few practical examples are in order, of the next small step we can make which won’t require Atom/RSS changes.

Reading books, a day at a time, is becoming ever more popular. You subscribe to the feed, and each day a new page from a classic, or not so classic, book is delivered to your aggregator. Here’s two examples: James Joyce’s Finnegans Wake, and Leonardo Da Vinci’s Notebooks.

The main problem of course is, like buying a magazine, you need to subscribe either on the day that the first page appears, or within the few days that page 1 is still being published in the feed. Talk about timely, I thought the Internet was supposed to ease on demand media delivery. If someone is publishing a book feed, I want to start at page 1, and I want to start today!

Instead, how about if the feed was able to reset itself to day 1, and hence page 1, from the moment I subscribe to it? You’re probably thinking cookies and other technical wonders, but it’s oh so much simpler.

Create a dynamic feed URL. Each day, the URL advertised as the book’s RSS feed, includes a date parameter, like this:

http://my.example.book.com/rss/1049066444

Or, to obtain some semblance of consistency, use CGI parameters:

http://my.example.book.com/rss?start=1049066444

Now the book site knows when the feed was subscribed to, and can generate the appropriate pages based on the start date. By using human readable parameters, users could even back and forward date their reading:

http://my.example.book.com/rss?startDay=1&startMonth=4&startYear=2003

Of course a dynamic URL is going to cause havoc with sites such as Technorati and Feedster, but to be honest, that’s a problem they’re going to have to solve eventually anyway, when we move to, you guessed it… controlling the publishing pipeline

How about offering an alternative feed for your weblog, one that drip feeds a post or two each day? If I’m reading you for the first time, it might be nice to read your weblog from the very first entry, with one a day until I’ve caught up.

I’ll let you work out the natural extensions to these ideas for interrogating big media sites, but until they get the whole business model problem first, I doubt we’ll be seeing them any time soon.

If you have a dynamic URL for one of your feeds, post it here in the comments, I’m interested to see what you’ve come up with.

(Originally posted to Synop weblog)