[icefox]

> The problem is one of content generation, as opposed to lack of tooling.

I also have been pondering docs for many years and I agree with this statement. The core value I was going to build my solution around was by incorporating / generating examples from real code. Rather than being docs that were published it would almost be close to a code search engine. While the developers are lazy and rarely document you could bootstrap the documentation by finding example usage in the wild and presenting it and from there allow the users annotate it (and the developers if they ever get around to it).

There are many problems with todays docs, another one that is missing has to be analytics. If I own an API I want to know what users are constantly looking at and discussing so that I can make that part better, less confusing, etc. And conversely if there are no example usage and no one has ever looked at an API that would be useful to know too as it could be code that could be ripped out and no longer maintained. What api are users searching for, but never find? There is more, but these are just a few off the top of my head.

By generating content rather than relying on the developers to sign up and publish the site would explode in size and utility. Think Yahoo directory v.s. Google ala 1998.

The idea that someone would generate static html files and put them somewhere is almost barbaric compared to the richness that could be provided.

Unfortunately I don't really have a financial story to go with this idea. One was to be free for open source projects and charge for commercial. But honestly more likely would be that I would build it for a few years and then someone like Facebook would just call me up and buy it out to shut it down and solve their API documentation problems. Not saying that is bad, but I would have to make it until that would happen and I would prefer to have found another solution than a aquirehire end game.