Any interest in a self hosted automation system for Tesla?

[tdeveloper99]

Helle everybody.

I'm a software developer/entrepreneur by trade and been working on a Tesla "Server" (of kinds) for little while now. I actually started with a completely different direction in mind, but am starting to lean towards a self hosted solution. (Too many privacy/security/liability type of issues with a cloud type solution). Before I go too far in this direction, and commit resources on my end (My company does other software related things, so I already have a team of developers, but this would be a new project that we would dedicate resources to), I wanted to see what the community feels about it.

This could be one of three types of solutions:

- Software only - Think Plex Media Server type of a server that would be installed on a Linux machine on your network
- A fully configured virtual machine (ESXI/Vmware) with the above software installed and configured out of the box
- A complete hardware/software product that can be plugged into your network. This would be a very low power server (don't need much horsepower), think ~5-7w power consumption.

Either of these solutions would offer additional functionality beyond what is available today, I'll get into the details of that shortly, but the most important thing is that it is hosted by YOU, lives on YOUR network, and YOU own all the data captured/produced by it. I'm not in the business of selling user data or anything like that, I'm in the business of providing quality software that offers functionality that you may be interested in.

I/My company would have zero access to your data/credentials/login etc etc. Our job is to provide you software that works as intended, is secure, up to date and hopefully evolves over time to provide even more functionality/automation.

Thoughts?

 

[4EVar]

I'd definitely be interested. But would prefer a windows server install like my plex is.

 

[fmm]

I'd be interested ....... I run Plex on a high powered iMac and could run this there too in a Linux Virtual Box session if needed. The turnkey hardware + software is also appealing assuming code extensions and new features wouldn't overrun the hardware quickly - otherwise the software only option would be best for me.

 

[heems]

I’d recommend a docker image... can then be run anywhere the user wants.

 

[ElectricOrgan]

Helle everybody.

I'm a software developer/entrepreneur by trade and been working on a Tesla "Server" (of kinds) for little while now. I actually started with a completely different direction in mind, but am starting to lean towards a self hosted solution. (Too many privacy/security/liability type of issues with a cloud type solution). Before I go too far in this direction, and commit resources on my end (My company does other software related things, so I already have a team of developers, but this would be a new project that we would dedicate resources to), I wanted to see what the community feels about it.

This could be one of three types of solutions:

- Software only - Think Plex Media Server type of a server that would be installed on a Linux machine on your network
- A fully configured virtual machine (ESXI/Vmware) with the above software installed and configured out of the box
- A complete hardware/software product that can be plugged into your network. This would be a very low power server (don't need much horsepower), think ~5-7w power consumption.

Either of these solutions would offer additional functionality beyond what is available today, I'll get into the details of that shortly, but the most important thing is that it is hosted by YOU, lives on YOUR network, and YOU own all the data captured/produced by it. I'm not in the business of selling user data or anything like that, I'm in the business of providing quality software that offers functionality that you may be interested in.

I/My company would have zero access to your data/credentials/login etc etc. Our job is to provide you software that works as intended, is secure, up to date and hopefully evolves over time to provide even more functionality/automation.

Thoughts?

I'm a geek and it looks like you are as well. So you are all about the technical details. But do you know your market? I am somewhat technical (Electrical Engineer) but I have no idea what your idea would do for me. What is the utility of a 'Tesla server"???

Before you get into all the tidbits of choices in implementation, what does this thing do?

 

[tdeveloper99]

I'm a geek and it looks like you are as well. So you are all about the technical details. But do you know your market? I am somewhat technical (Electrical Engineer) but I have no idea what your idea would do for me. What is the utility of a 'Tesla server"???

Before you get into all the tidbits of choices in implementation, what does this thing do?

Patience...fellow geek.

You don't just divulge everything at the onset. The intent of my post was to gauge interest in this type of a solution.

That being said, think synchronous/asynchronous events that are exposed and subscribable, an API for external integration, GPIO support etc, from a geek perspective.

From a non geek perspective, think full logging and reporting, macros to automate things, car specific functions, etc.

 

[Tam]

...Our job is to provide you software that works as intended, is secure, up to date and hopefully evolves over time to provide even more functionality/automation.

Sounds great. I want it!

I don't understand much about programming, running scripts... but I can install programs including Plex to record TV in my Windows PC and retrieve it and watch it on Tesla.

...What is the utility of a 'Tesla server"??? ....

I assume it's a data logging service so I can import to a spreadsheet and draw any graphs I want such as:

.the various rates of supercharging throughout the session...
.12V battery voltage through months and years...

Also logging of trips so I can look back and see how much I should need to charge my battery this time...

 

[tdeveloper99]

I assume it's a data logging service

That would be a very small part of it. There's so much more than can be done.

so I can import to a spreadsheet and draw any graphs I want

It'd have a browser based front end with built in reporting, including "pretty" graphs. But yes, you could also export the data out and do whatever you want with it. It's your data.

.the various rates of supercharging throughout the session...
.12V battery voltage through months and years...

Would be two of numerous other metrics.

 

[mjptech]

Would definitely depend on the cost.. free/freemium/perpetual/annual.. whatever you decide, I'm sure that may also impact the decision (it would in my case)

 

[tdeveloper99]

Would definitely depend on the cost.. free/freemium/perpetual/annual.. whatever you decide, I'm sure that may also impact the decision (it would in my case)

Cost is always a concern. I have not finalized any pricing/pricing model yet and probably won't be until the system is more feature complete.

What I can say is that there will be a "beta" period where the system will be free for the duration of the beta. This is fairly typical in the software world and there will be no license restrictions on any functionality during that period. Once the system is "production ready", there will certainly be a time limited trial available for people to "try before buying".

I'm a firm proponent of the belief that if it's "free" there is always a catch somewhere and the user/user data IS the product. I don't want to play that game with an automation system that potentially affects something moving at high speed and/or high energy involved.

I'm confident that if users see value in the system, cost will not be a concern. There will always be a segment that is very price conscious and that, well...it is what it is. Can't do much there. Developing these kinds of systems involves a fair amount of investment, and there has to be return on that investment.

 

[gbailey]

I’d be interested. I host Plex on a Ubuntu server so would be more than happy to try any beta that you develop.

 

[ElectricOrgan]

Patience...fellow geek.

You don't just divulge everything at the onset. The intent of my post was to gauge interest in this type of a solution.

That being said, think synchronous/asynchronous events that are exposed and subscribable, an API for external integration, GPIO support etc, from a geek perspective.

From a non geek perspective, think full logging and reporting, macros to automate things, car specific functions, etc.

It's not a matter of patience. You aren't giving us any idea of what you are talking about. You talk about "solution" but fail to mention the problem!

What does this thing do? Why should I continue to even read this thread?