User Errors on the web a model

So I am working on this model again, something I started about 2 years ago. There is still a heap to do, certainly the user testing and the recommendations drawn from it.

So what is this model I am talking about and why should anyone care?

Well, so far as I can find, no-one has ever mapped out the end to end process of retrieving a web page even though it is covered by some very specific standards. This was my original aim… to make a taxonomy of all the errors that could occur so users could have a better understanding. Things like the Motley Fool’s page not found error page are already starting to help a lot. However I wanted to go a bit further.

There is some research (I have the papers but not to hand) which suggests how one might educate or describe to users how the internet works. Describe the parts of the system which might have gone wrong in such a way that they don’t feel isolated. Internet Explorer and Firefox, and other browsers have started to do this more, but still not enough. My hope is by describing a generic model (with examples of typical configurations) that this model can be used to help build better web site and browser error pages. Things which educate users and point to the best possible tactics they can employ to recover the situation.

I also of course want to advocate the best use of auto-recovery strategies, such as the Motley Fool uses. Why should a user have to email you to tell you about a 404, or even press a button to email you? Why not just email yourself in your 404 page?! That’s just a simple example, but why let your users work unnecessarily.