I think we're both designing the same thing in our heads
I picture someone needing to know what's correct for their steering column, for instance. They can either enter a partnumber or view a exploded schematic and click the part they want to get info on. Say they click the horn brush. Since they've entered the model/year/market of the vehicle they want the info on, the script can show them anything we would want it to: a picture of the item, one of it installed, application(year and model span), upgrades and current sources for the part with user reviews, in regards to fitment and it's match for OEM quality.
Like we've both mentioned, the scale of this is immense. Even with a partnumber database, it would require an incredible amount of input. Luckily, I've got a lot of free time
. Even if it took 5 years to be useful to more than half the people trying to use it, it's not something that would ever be a waste of time. We can safely assume that nobody will ever produce something that makes it obsolete. Everyone that runs a website is willing to provide the simple things, like forums, classifieds, downloads, etc. That's because the systems have either already been written for them or they are easy to write yourself. Something as large as we're discussing is too much work for too small a market. The only reason it would be written is as a labor of love and 99.9% of the site owners don't love VW's that much. If anyone did, we'd have the system already
I'll post my script roadmap and design tree to the forums if you want to see it. You seem to have exactly the same functionality and end result in mind, so I'd be very interested in your thoughts and suggestions on the matter.