It's not much of a problem when you're running the server for them, but when the server is running on their desktop, as it is in River2, where do you put the publicly accessible content? I could operate a storage system for them, but what happens if the service doesn't take off and never becomes a business? At some point I have to punt, and that leaves users with data that goes missing. Thinking long-term I don't want to set another one of these up.
The right answer is to help them get their own storage working. And Amazon has made that relatively easy, so I spent the time to make it really work.
The users' feeds are stored in Amazon S3. For this I need three pieces of data. The "Access Key ID" (which is the equiv of a username), "Secret Access Key" (password) and a default bucket. That's explained on the Frontier News site.
Frontier News: New preferences for Amazon S3.
I created my own feed using the new tool, you can see it's on Amazon. I wouldn't subscribe to it, while it should keep updating, I'm going to merge it with my lifeliner feed.
http://bit.ly/4PKh9T
