I applied to YC and was flown out for an interview in November 2008. I was told that while it was a promising area, YC is reluctant to fund single founders. Fair enough.
A few weeks later, I responded to a technical plea for help from one of GraffitiGeo's founders in this thread a few weeks after that: http://news.ycombinator.com/item?id=413929
When we got in touch, we had a long and frank discussion about Graffitio and the space in general. I told him about the challenges I was facing, and some of my plans for the future. He brought up the idea that we team up, and I said I probably wasn't interested, but I'd let him know.
Then I got the email. He offered me 1.5% of their company for all of my source code, the name, and my 40,000+ users at the time. He went on to use threatening language such as "Graffitio is going to get stomped out of the picture" and "Your financial alternatives are bleak." I chose to pass on the offer.
The first version of GraffitiGeo that came out in the App Store was an _exact_ clone of my app, Graffitio, down to the terminology such as Graffiti and Walls. They started to differentiate themselves with later versions. Some of those differences were novel, but others were features that I'd mentioned to him in our conversation. To be fair, they could have arisen independently.
Graffitio is over a year old now. It was in the App Store on day 1. I should probably be more annoyed than I am, but honestly, I have a lot of reservations about this space now after being here for over a year. Dealing with the App Store is another headache. I haven't been able to get an update through in months. Anyway, the first round of geotagging apps have fallen flat IMO. Turning them into a "game" like Foursquare and GraffitiGeo isn't going to change the fact that they deliver relatively little utility.
There might be a magic combination out there that makes it work, but I don't think GraffitiGeo or Graffitio in their current state is it. I wish them luck in finding it. If anything, they've lit a fire under my ass to spend more time on Graffitio.
Do you control the web service? If so, throw XML out the window. The open source JSON Framework on Google Code kicks ass. My iPhone app, Graffitio (http://graffit.io), uses Rails on the server side spitting out JSON to the iPhone app. Its awesome.
Use VMWare Player with that image as the base, install your Rails app to it, and go from there. Then you can give your client the pre configured image and they can deploy it however they please.
That image is Engine Yard in a box, and Engine Yard is awesome. You
http://news.ycombinator.com/item?id=763559