We're going much further with this than the last time around. 

Which was in 2004, if I'm not mistaken.  

Here's the deal. Outliners are ways to edit hierarchies.  

Given the Internet, we should be able to join our hierarchies, pretty seamlessly, so when we cross the boundary from my space into yours, it might not be so easy to see we've gone from one place to another. 

With this effect, we create huge multi-author places, where everyone retains ownership of and control of their own work.  

It worked out really well, in 2004, when I was working with Adam Curry on the bootstrap of podcasting. If you ask me, and I bet if you ask AC, the World Outline will, long-term, be a bigger deal than podcasting. Because it relates to how we collaborate on the organization of the information we all create. If we can just entice other people to do this!  

Anyway, here we are seven years later, and seven years older, and more relaxed with a greater appreciation for how huge this job is, we're going slower, but yet reaching much further.  

For example, the number of nodetypes has grown. We now support these types, in alphabetic order: 

blogpost, code, howto, html, photo, river, rss, redirect, include and plain text. 

And more important, there's a callback structure that makes it easy to add new types to the system. Or to override or subclass the defaults. 

And it's open in the only way that counts. You don't have to export your data to gain access to it elsewhere. You can include mine in yours and it continues to live in both places (with only one original so changes flow in a natural way). 

You'll recognize that these are problems that many others are working on. But I think we're uniquely working on it from this point of view: There is no business model, in fact no business at all, that's trying to capture anyone's content. I might offer to host it for you, but there's a pref setting that allows you to store everything in an S3 bucket that you own. Just give me write-access to it, which you can take away at any time for any reason (some have already done this) and we're still friends.  

What I'm going to do, over the next few weeks at least, is post stories to Scripting News that show you how this works. It'll be as if you're a passenger on a plane. Eventually, soon -- I want to teach you how to fly one of these planes. But it's complex, and if you try it without knowing what's going on, you will either: 1. Fly your plane into a mountain, killing everyone on board. Or, 2. You will flap the wings once or twice, wonder what the big deal is about and go on. Neither outcome is acceptable to me. So I'm opting for a third approach. A fairly long tease period, with a few real-world seminars here in NYC and elsewhere (Toronto and Boston, probably first, maybe San Francisco). The only way this can work, in my experience, is with a slow build. If I were doing this inside a tech company, I would just use the people of the company to boot it up. But I'm doing this work on the open Internet. So I have to be a bit more clever about it.  

Anyway, here's the first tease.  

Today I did some development work, which naturally I wanted to narrate for all interested parties. 

The work: I improved the error message for Amazon Route 53 when you try and can't create a "blorkmark." The previous message was fairly incomprehensible. The new message, less so. 

I wrote-up the changes in my outliner, set the type to "blogpost" -- added some details as sub-heads, and saved it.  

This is what the web page looks like. 

And this is what it looks like in my outliner.