Click here to show or hide the menubar.
Thread started by Dave Winer on Friday, May 10, 2013.

Should the Community Feed be an RSS feed in addition to being an OPML feed?

This question came up in the Community Feed, which you can read in Fargo, by choosing the Community Feed command from the Docs menu. Or you can read it in the Small Picture Reader if you don't use Fargo. I wrote my answer there, but thought it would be interesting to also post it here. No I didn't use the fancy Blogging 2.0 protocol I described in an earlier post. Soooon!

Well of course it would be nice to have everything, if there were no cost.

It would take time to write the code and keep it running. It would be worth doing if there would be a lot of people using it. But right now the Community Feed a new feature. We're still at the point where we're introducing ourselves. If that's all it does it will have been worth it.

I'm an investor in software, and I have to make decisions as any investor would. I can't buy everything. And right now there are other projects that I think need more attention.

Also, and this is a key point, this is not something you need Kyle or me to do. The OPML feed is public. If you want to write the code to convert it to an RSS feed, you can do it.

Read it once every ten minutes. Use the eTag feature of HTTP to conserve bandwidth. Generate RSS 2.0. How will you synthesize a title for each item? I don't know, that's a hard problem. RSS 2.0 doesn't require titles, but Google Reader did. That made generating RSS feeds a difficult process for data that doesn't inherently have titles. But Google Reader is going away, so we're free to do as we please, you say. Not so fast. The replacements are clones. I bet they're just as picky as GR was. At least until the dust settles, and that isn't going to happen this year even, probably.

But OPML feeds? Ahhh that's easy. Since I'm writing both ends I can make it work. And if I want to change things based on what I learn, I can do that too. That's why the early days on anything are important. And why you should go slowly enough so you can feed back what you learn into the protocol.

Anyway you see these questions sound simple, but when you actually start writing the code, they can become complex.

Bottom-line: My bet is that no one would use an RSS feed of this content. That makes it a bad investment. I've been wrong before, btw.

XML