On the Lack of a Content Update Protocol…

I’m still dealing with my annoyance at the lack of a “content update protocol.” Or maybe I’m just annoyed that RSS has lost support over the years.

Someone mentioned Visualping when I posted before, which I’ve been playing around with (they were gracious enough to give me a free account). It’s very well-done – probably at the extreme end of the spectrum for site update detection. Visualping can isolate an HTML element (like the “What’s New” sidebar or something), and it will diff the prior version of that with whatever it finds when it checks.

Here’s the thing – it can diff the text, which is expected; but, it can also diff the pixels. And this is so granular that when I changed the JPG quality of a single image, Visualping actually pointed out the pixels inside the image that were different.

(There’s a quality control play here, I feel like. If you’re a site manager and you want to error-check your site as a background process, you could probably find a way to use this.)

Someone else mentioned RSS.app, but I didn’t get great results out of this. It allows you to make your own RSS feed out of patterns of repeating HTML elements. Problem is, this depends on how the site’s HTML is structured, and it depends on that not changing over time. I assume failures would be silent – you would just stop getting new items if something changed.

(To be fair, I didn’t get a lot of time to work with RSS.app before the free trial expired. Ironically, the only item I ever managed to get in a feed was the notification that my trial was coming to an end.)

So, this brings me back to my annoyance that RSS fell out of favor for some reason.

The biggest hit was losing Google Reader. But, honestly, I blame JSON for a lot of it. When JSON became the new thing, anything “tainted” by XML was effectively deprecated in a lot of minds, which is sad.

I don’t care about language/format in the slightest. Why can’t we make JSON-based RSS popular? (Note: someone has tried it with jFeed. There are probably others).

I know we don’t need another standard, but here’s the thing: RSS readers could just check for the format and manage their own parsers. Is it XML? Parse it as XML. Is it JSON? Parse it as JSON.

I feel like we should abstract ourselves away from something as pedestrian as the serialization format (yawn) and just concentrate on the logical process of communicating a content update. How we actually send that information down the wire seems like a small detail.

If we popularize a JSON option, will people be happy with it again? (Yes, I’m bitter.)

I’m ranting. But the lack of penetration of Deane’s Desired Site Update Protocol is a matter of … value. There’s a lot of value here that we’re all leaving on the table because of some weird hang-ups.