Ok so it won't win any awards but not many startups can have a prototype done in 5 minutes!
What you're suggesting at the core is basically what mobile transcoders already try to do, and any document classification system which needs to work out the most important content. It's a hard problem but there is plenty of thoughts out there to build on.
What you could do, even, is build a "meta-service" on top of something like Google's transcoder, where you pass the site into Google, then transcode /it's/ results to add a nicer style and the usability stuff like the feedback link. That would be a way to get something started pretty quickly, and you could use the feedback to see where Google isn't good enough, and therefore how to improve the service.
Comments
Ok so it won't win any awards but not many startups can have a prototype done in 5 minutes!
What you're suggesting at the core is basically what mobile transcoders already try to do, and any document classification system which needs to work out the most important content. It's a hard problem but there is plenty of thoughts out there to build on.
What you could do, even, is build a "meta-service" on top of something like Google's transcoder, where you pass the site into Google, then transcode /it's/ results to add a nicer style and the usability stuff like the feedback link. That would be a way to get something started pretty quickly, and you could use the feedback to see where Google isn't good enough, and therefore how to improve the service.
How would you make money though?