Jeremy Keith posted a link on Twitter today to a post he wrote at XTech in May 2008, about a talk by Steven Pemberton called “Why You Should Have a Web Site: it’s the law! (and other Web 3.0 issues)“.

I was interested by the main focus of the talk, most notably this passage on Metcalfe’s law:

These are examples of Metcalfe’s law in action. What should really happen is that you keep all your data on your website and then aggregators can distribute it across the Web. Most people won’t want to write all the angle brackets but software should enable you to do this.

The concept is that you store your photos on your own web site along with your audio files, your blog posts, your listening habits - then provide an easy machine-readable method for distributing this content to other web services via aggregation. So you still use Flickr, you still use Twitter, etc - but your content resides on a server you control so that should the service go away, you don’t lose all your data.

We like to sit pretty and think we can rely on these services, avoiding the hard work and time it would take for us to set up a decent personal aggregation system - after all, Flickr and the like won’t just disappear, right? The problem is, it doesn’t take much for times to change. Just ask any frequent users of Pownce.

This concept has got me thinking about how I want to approach the topic. I’m leaning towards sticking with services that provide an API, then writing some simple scripts to either send the content I upload on my own site to the relevant service, or pull content I’ve uploaded to a service to my own site. This would allow me to still gain the benefits of said services (like last.fm’s iPod scrobbling app that integrates with iTunes or Twitter’s SMS functionality) while providing redundancy and going some way to ensure I don’t lose out if a service shuts its doors.

Potentially I could take it another step and do automated nightly back-ups of my own server to my local machine, which in turn could then upload to a service like Amazon S3 or Mozy, so that I was covered in the event of my own server getting hosed. Sounds like a lot of work for content that is mostly only of interest to yours truly, but then I really like being able to look back on things years later to see how my life has changed.

I’ll no doubt update with progress on this as I go and I’ll make any scripts I produce for the purpose available under an open source license, in case the idea appeals to anyone reading.