A frequently asked question goes something like this. 

"I like JSON. What if apps started publishing RSS data in JSON." 

1. My first answer: Sure, why not. I experimented with it in October, and found prior art -- a website that takes a feed and returns JSON.  

2. I wrote both ends of a protocol that depended on JSONified RSS. It worked. I didn't find any dead-ends.  

3. Now, that doesn't mean there aren't any dead-ends. If I were a betting man, based on past experience, I'd expect there are. But they wouldn't necessarily be the end of the process. It depends on how determined one is, and if a consensus can be arrived at as to which of the possible egregious hacks would be used to back out of the dead-end. 

4. Motivation. Now there's the really big problem. Why would anyone be willing to recode their RSS support to use JSONified RSS? I did it as an interesting hack, a diversion, something to pass the time while my brain warmed up for other work.  

5. If you run a news site, could you replace your XML-based feeds with JSON without losing a bunch of readers and potential readers. So you'd be stuck supporting both, for the indefinite future. Congratulations, you just made your life more complicated. What did you get in return? 

6. But the world is switching to JSON. Isn't it? Answer -- yes, that does seem to be happening. For new applications. If it's going to happen in an existing app, like HTML or RSS, it's a very long way off. Do a view-source on any web page. See all those angle-brackets? Now think about all the brains that know how to do that. Installed bases like that come, at most, once in a generation. You can't wish them out of existence. If XML is meant to be replaced by JSON, that process will take many many years.  

So RSS in JSON is technically a real possibility.  

Will it happen? I wouldn't bet either way. It might happen. That doesn't necessarily mean it's a good idea. Or that it's not a good idea.  

Will it replace XML-based RSS? Obviously, I don't think so.