While doing some additional web research on Tesla, I came across an article that got Ye Olde Gray Cells really cranking away re a future state for
I'm not going to post the URL because I can easily summarize the article's ONE main point.
After spelling out all the safety shortcomings there are in all current driverless cars, and are likely to be unresolvable for an extended period, the author advocated that ALL the manufacturers of such vehicles JOINTLY develop an open source approach to developing the software, and the data the hardware generates.
Well Linux hasn't become a mainstream OS, but that has as much to do with MS, as well as the majority home user market unwilling to put the work into it to make the transition; and being free obviously isn't sufficient motivation for the majority.
So I wondering re this topic how this board feels about driverless cars using open source software (rather than in-house propriety software)???? And need I remind anyone just how good some of us are at reverse engineering code regardless of whether it is legal or not!!! Is it that hard to imagine someone somewhere erasing the proprietary stuff and installing their own???
I was thinking that the Microsoft Developer Network Model would be a way to organize the personnel who would like to get involved (I must admit that I have zero idea how, for instance, a Linux distro like Ubuntu internal organization is formed from a Human Resources Dept model re "staffing", quality control, etc., gets done).
So there's nobody to carry the can? I'd suggest not. But, I can see it happening.
I can see these things being used as weapons, possibly for murder.
"It wasn't me! It was the car!"
Blind people are getting quite excited at the prospect of independence from others. Quite how they would gain control in a dangerous situation has not been explained to my satisfaction.
Exactly my thinking, Katzy. Whilst Open Source is attractive, I'd suggest that allowing anyone, and everyone, access to said software is fraught with
danger. There is a lot of money tied up in the automotive industry and, that is the reason why cheap spare parts are not as readily available. We were
involved in developing a software controlled wiring harness for cars back in the 90s and the algorithms were such that the major manufacturers had not
worked out how to perform the tasks that we had without requiring, then, way more computing power. For that fact, cars still have not developed the
control systems that we had. Briefly; 4 wires to control the entire car, CPU behind the dash, smaller units in each major slave area (lights, doors
etc). We could have even used it to drive the engine too.
I agree, that an open source model for certain segments may work (but only open to the industry) but I suspect that it may not work for all given the different way cars are designed and built. e.g. you may have a led lamp to indicate an airbag failure another mfg may use an audible signal, another may use a scrolling message, whereas another may actually shut the car down citing safety, and so on.