The model s apps sdk...
Geek/developer disclaimer: some of the content below may not make much sense if you're not one of the above and even then... anyhow.
Also sorry for cross posting and if that breaks any of the rules here, but since tesla's forum (thread was initially posted there) is way too aggressive so I'm posting over here instead.
I've asked Elon in person myself about the SDK (not the remote control app for your smartphone that some of you refer to) in public during the press conference at the Geneva Motor Show last month and his answer though at first may sound a bit disappointing makes a lot of sense from a development methodology perspective. I'm going to paraphrase it here, but I do have a video that I can dig out if needed.
'We want to make sure that the platform is stable enough to support 3dr party apps so we're expecting to release something to developers about one year after the car has been into production'
Personally I'm very happy with this answer because it's a clear timeframe, something very important if you're serious about building a business on this "niche" platform. It also shows that they're serious about it. Anyone who's written code knows that building software is one thing, opening it up to 3dr parties is another. You need maturity in many areas if you are to write an API for "everyone" to use without breaking anything else already in place in the process. So what I'm suspecting is that we'll get software v1 when we get the cars this year, v2 will be pushed to the cars some 6 months down the line (almost certainly OTA) with some 3rd party support (Tesla will probably choose the likes of Pandora/Spotify to work with them) and 6 months later v3 with "open" 3rd party support will be delivered.
One thing I really hope is that they'll manage to make the simulator software only (unlike Ford's sync which requires hardware worth north of $5K) thus pushing innovation in yet another way compared to other car makers. It's quite complicated though because you have to simulate quite a lot of things in terms of core car systems on top of which you have to simulate the unit itself and depending on what type of application you're building you end up running within that environment, but what's to stop you building an app/service that runs on both your smartphone and communicates with your car (think synchronization of queues/playlists)?
In any case I'm very excited about the future.
Geek/developer disclaimer: some of the content below may not make much sense if you're not one of the above and even then... anyhow.
Also sorry for cross posting and if that breaks any of the rules here, but since tesla's forum (thread was initially posted there) is way too aggressive so I'm posting over here instead.
I've asked Elon in person myself about the SDK (not the remote control app for your smartphone that some of you refer to) in public during the press conference at the Geneva Motor Show last month and his answer though at first may sound a bit disappointing makes a lot of sense from a development methodology perspective. I'm going to paraphrase it here, but I do have a video that I can dig out if needed.
'We want to make sure that the platform is stable enough to support 3dr party apps so we're expecting to release something to developers about one year after the car has been into production'
Personally I'm very happy with this answer because it's a clear timeframe, something very important if you're serious about building a business on this "niche" platform. It also shows that they're serious about it. Anyone who's written code knows that building software is one thing, opening it up to 3dr parties is another. You need maturity in many areas if you are to write an API for "everyone" to use without breaking anything else already in place in the process. So what I'm suspecting is that we'll get software v1 when we get the cars this year, v2 will be pushed to the cars some 6 months down the line (almost certainly OTA) with some 3rd party support (Tesla will probably choose the likes of Pandora/Spotify to work with them) and 6 months later v3 with "open" 3rd party support will be delivered.
One thing I really hope is that they'll manage to make the simulator software only (unlike Ford's sync which requires hardware worth north of $5K) thus pushing innovation in yet another way compared to other car makers. It's quite complicated though because you have to simulate quite a lot of things in terms of core car systems on top of which you have to simulate the unit itself and depending on what type of application you're building you end up running within that environment, but what's to stop you building an app/service that runs on both your smartphone and communicates with your car (think synchronization of queues/playlists)?
In any case I'm very excited about the future.
We in the software industry tend to rush things out too often in a half-baked state. It may not be testing at all that might be holding things up but, might be the task of fully-fleshing out (or even starting to build out) all the SDK ecosystem aspects such as sandboxing etc. as AnOutsider said.
