OSDash Genesis Thread - Was: OpenVIS - The Open Vehicle Infotainment Service

About a year ago I started a project on sourceforge called OpenIVI. Its objectives were to provide a number of webservices and client software for integrating the car with the cloud. This web service was to be pluggable and open so that you could run it on your home server, your carpc itself, or bigger institutions such as mp3car could run it to provide features to their users.

These web services/client would be developed using crossplatform C#/ASP.NET code and provide services like tracking, crowd sourced traffic, and many other possible services. It would be pluggable so platform specific components could also use the cloud for things.

The project got a code commit, but I'm involved in too many other things to have time to develop it myself. This would be a cross-frontend/cross-platform effort so why not open up development to the entire community?

Bugbyte and I threw together some ideas and I'm hoping to get even more ideas and excitement around this project.

I won't do this all by myself, so if you are interested in helping out with this effort, sound off, otherwise the idea will likely die.

OpenIVI agenda:
- Define plugin architecture
- Define useful services to start out with
- Write core server and web frontend
- Write client software service to run on hardware.
- Establish a test server.
- Get Frontends to use it.

We need C#, ASP.NET, and web developers to jump on board. I think the client plugins that integrate with the frontends will either come from the frontend authors themselves or loyal users.

---------------------------------------
Edit by Bugbyte 12/21/2009: This is the genesis thread for the OSDash project. It has been moved to the OSDash forum now that a more formal effort to realize the vision has been started.

Kev, I'm having a conversation with mp3car on Monday about putting some resources behind this.

I'm onboard for sure. I'd like to try and get more people interested in cloud computing and applications for the car.

It sounds like it could be interesting....i'll hold judgement until I see what type of services your planning on offering. Most useful data specifically disallows anything other then end-user use. So I guess thats a maybe from me.

Services off the top of my head:

* OBDII statistics upload (requires some sort of user account)
* Crowd source traffic (anonymous and user)
* Weather, ie provided by government sites can be used
* Car settings/user preferences
* Playlist management
* Media sync'ing
* Calendar/Contacts

I'm sure if I spent some more time I can think of a bunch more that would be useful. I'm sure others have better ideas.

media and playlist syncing could be a big one... and add a send address to car service (like bmw and a few others can do right from google maps). User settings could be cool too especially if tied to your phones bluetooth address. You get in a car, and it can grab your favorite radio stations or even start playing the last song you were listening to.

my question with some of the others is why choose the mp3car calendar service over google calendar? or its weather over the weather.com rss feed? or traffic generated by half a dozen mp3car users over yahoo traffic?

Not trying to be pessimistic just asking questions most people will ask when you present the services.

Quote:

---

Originally Posted by justchat_1

my question with some of the others is why choose the mp3car calendar service over google calendar? or its weather over the weather.com rss feed? or traffic generated by half a dozen mp3car users over yahoo traffic?

---

It's hard to say precisely why you might choose one over the other. Maybe support for one is discontinued or someone comes up with better features for another. What's important for me is that you can pick and choose one or the other and it will just work.

Quote:

---

Originally Posted by justchat_1

my question with some of the others is why choose the mp3car calendar service over google calendar? or its weather over the weather.com rss feed? or traffic generated by half a dozen mp3car users over yahoo traffic?

---

Well, who says the client can't have a plugin that talks directly to google calendar? Also Bugbyte is right, say google starts charging for caledar service, or they drop it altogether. On the server, we drop the google calendar plugin and use another service. All the clients running on 100+ machines don't have to be changed at all. For this reason, it'd be useful to have a generic interface of services, and separate out the backends. Also consider locality into the equation. Traffic service X may not provide traffic for Switzerland. In which case you'd use plugin B which does hook up with the right service for the locality.

Well none of those services are going to be paid only anytime soon but the point about easy transitioning is a good one. All i'm saying is concentrate on the services with the most user benefit first and then re-invent the wheel.

Traffic service is going to be incredibly difficult to get reliable data for. Now i'm assuming your referring to statistical (aka historical) traffic and not live traffic. In order to pull that off you need two things which don't exist in the carpc world yet, standardized location data (maps) and a large enough userbase to get valid data. Even assuming you break road segments by intersection and use fuzzy matching, you violate terms of use for the maps.

3 words: OpenStreetMaps. You are right though, it'll be not entirely useful until large numbers of people use it. I don't expect this project's success overnight. If it takes a couple years to become really useful, that's okay with me. I realize that none of this will be very useful to the 90% of carpc users. But if the trend continues, it'll be very relevant very soon.

I agree, focus on what's most useful at first.

I'd be great to have another c# dev on board besides myself.

This is exactly what i have been waiting for - I have resources to apply to this, both financial and human.