River4 blog: Bare bones River4 howto.
River4 blog: River4 and the local filesystem.
The net effect of making River4 work with the local filesystem is that it's simpler to set up. Originally I made it work with S3 because we were deploying on Heroku, and since it runs in the Amazon cloud, it made S3 an economical choice (and Heroku apps don't have a persistent filesystem). But now we want to branch out, and setting up the S3 connection was tripping up a lot of people who set up their own rivers outside of Heroku. Now it should be easier. And we'll make easier still.
We know that it can get very easy, because the first Radio had the aggregator running on the end-user's machine, in the background. The users were running a web server on their desktop, though not many knew that. Proving that it can be made so easy that it just melts into invisibility.
One more thing, River4 is designed to plug into the new RiverBrowser software. All this stuff I'm working on that may seem so diverse and possibly random is part of a grand scheme to get more content flowing over the net with open formats and protocols, without being locked into any silos. The trick is to make it easy. We're getting there.
This post appeared originally on Facebook.
If they were open, then I would be interested. Because that means I could use my tools with it. At least with Facebook, there's a lot of openness, I'm writing this using my software, and if I weren't I could easily get it out (though honestly I haven't tested that yet). Here's a screen shot.
This is the same problem I have with people pouring love into Medium. I don't care if they pay you to write there, then it's just like Vox or Huffington Post or Quartz, Buzzfeed or any of a thousand other sites that employ writers. But if you do it thinking this is some cool way to publish, it's the most uncool way! Really seriously wrong that people do this thinking they might be doing themselves some good. It's the bad bargain tech companies always try to make with users. And users still fall for it.