I found this today. It’s apparently some kind of presonal organizer from 37signals, the people who made Basecamp and Ta-da List. It sounds neat, but I have this nagging feeling about ideas like these. I think that the reason why some of these things work is that they are centralized, and that one of the major things we need to do as web developers (those who are) is work out how to decentralize everything.
Look at Livejournal. Livejournal is a community that people on the outside (myself included) don’t really understand. The functionality afforded by their application is not as slick as many of the free blogging software solutions. The trick is that it also provides collaborative friend and interest lists, which is impossible to do with stand-alone blogging software.
What software like WordPress and Movable Type need is an underlying infrastructure for installed instances to communicate with each other to share metadata. This metadata can be used to power the features that make sites like LiveJournal and Basecamp tick.
A technology that could address these issues is XFN, the XHTML Friends Network. It seems like a great way to provide the information, but you still have to centralize it to make it useful. As far as I am aware, there is no way for your site to benefit from XFN info on other sites without something to scrape and reconstitute the information. Even those tools that centralize the information don’t appear to return that information back to the providers in a useful form (OPML? RSS?).
Technorati is another one of those centralizing sites. From there, you can get a list of people who have linked to your site. Why can’t your own blog software provide this information to you?
People who are thinking about these ideas will note two things: First, your blog would have to scrape the web in order to find those links, which would be inefficient. Second, technology such as Pingback is useful for obtaining that information as long as everyone who has a blog supports and uses it. But I would like more.
My intuition says this has something to do with the threaded external blogging idea that has been going around on the WP-hackers mailing list. A distributed datastructure that underlies a post could easily be the carrier for threaded conversation, and also help maintain the threads from disparate, possibly disabled sites. Each site could share a default number of pieces and request the missing ones on demand. Should a piece go missing, the underlying framework would know how to fetch a distributed copy and redistribute it to preserve redundancy.
These are all loose ideas. I needed to set them down so that I had an idea of what I’m after. Not that I’ll be writing code to implement this soon, but when I do, I need to know what points to hit.
