Visual metadata in feeds

I think crazy people who say things like “RSS will kill the visual Internet” need to have their own blog. It would save me the effort of having to provide back-story when I blog about them.

Back-story: I used to work for a Gannett newspaper in the early days of digital publishing. A counterpart of mine used to call me out of the blue and say things like “the neighborhood links in your community section are wrapping funny, I’ll send a screen shot.” John was one of two loyal site visitors that kept Lynx (a text-only browser) showing up in our user agent stats.

Apparently, J.P.’s tragic love affair with the Lynx browser has been redirected to the desktop feed reader. If I correctly understand his particular dialect of crazy, he thinks that content distribution and discovery will become so “efficient,” it will travel direct from the publisher to the consumer and never hit the web, thus bypassing visual distractions. I might be jumping to conclusions for him, but I believe what he was getting at was that [newspaper] sites aren’t necessary after all.

Obviously, this is just an old friend jabbing an easily antagonized web geek, but J.P.’s dreamt up news utopia is far too spartan to go unchallenged. The “visual Internet” isn’t just a scheme to sell ads. Forget advertising and branding and navigation for a minute. Visual formatting (font-face, font-size, font-color, link-color, text-decoration, line-height, et cetera) improves readability. Reading A List Apart is different than reading Scripting News. And by “different,” J.P., I mean “better.”

Stefan and I were just talking about the potential benefit of visual metadata in syndicated content feeds. All the attributes that can make one blog more readable than the next (at the site) could provide the same benefit for content being consumed in a feed reader (save perhaps for font-size.) If feed readers allowed consumers like J.P. the option of turning off publisher style specifications or users like Stefan and I to turn them on, it would be a simple matter of adding visual metadata into the feeds.

This might seem like a big change to RSS and Atom specs. Some would likely argue that it runs contrary to the original purpose of these data formats. But hAtom (I wish we could just start calling it hFeed) has a different origin in that it’s simply defining data types through semantic markup of existing content on a page. Why not define metadata (beyond author, pub date, etc.) through that markup as well? This seems like a very intuitive inclusion.

It’s only a matter of time before mainstream feed readers start allowing subscriptions to pages with semantic markup. If a blog author like Dan Cederholm, for instance, had a microformat / semantic html feed that also included some visual metadata to improve the readability of his writing, I would subscribe to his semantic html feed rather than to his spartan xml feed.

I’d love to see a visual distinction between my news feeds and my blog feeds (no, not just serif vs sans serif.) And adding visual metadata for the entire feed (in addition to a bit of post metadata) would be another cheering point for those of us that still see benefit in the “visual Internet.”