Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 18 submissions in the queue.
posted by Fnord666 on Saturday October 20 2018, @12:58AM   Printer-friendly
from the all-your-data-are-belong-to-us dept.

The Verge is reporting that the next data minefield is your car. GM has been capturing lots of user data from the cars they have sold and is apparently planning to sell that (stolen|coerced) data to advertisers targeting, for now, radio advertising. Newer cars generate upwards of 600GB of user data per day. This is causing business leaders to drool because some expect the value of this data to reach more than $1.5 trillion by the year 2030, if the data (capture|theft) remains uncontested. GM is the first auto maker so far to try this. The first batch took data from around 90,000 vehicles. However, there was not much detail given about how permission was gained for this data capture and whether agreement was coerced or through ignorance.

GM captured minuted details such as station selection, volume level, and ZIP codes of vehicle owners, and then used the car's built-in Wi-Fi signal to upload the data to its servers. The goal was to determine the relationship between what drivers listen to and what they buy and then turn around and sell the data to advertisers and radio operators. And it got really specific: GM tracked a driver listening to country music who stopped at a Tim Horton's restaurant. (No data on that donut order, though.)

Also at The Detroit Free Press : GM tracked radio listening habits for 3 months: Here's why.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by RS3 on Saturday October 20 2018, @11:56PM (6 children)

    by RS3 (6367) on Saturday October 20 2018, @11:56PM (#751514)

    > Really you need a vehicle from before about 1995.

    Spot on. OBD2 started in 1995, and 1996 up it's required. I doubt anyone makes a vehicle tracker for pre-OBD2, although vehicle speed data comes through the earlier ports, but not usually power, so there's that too.

    You could go pre-1980-ish and have no computer.

    If they try to force me to use an OBD2 dongle, I surely hope it's internally protected from very large electrical surges.

    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 2) by dry on Sunday October 21 2018, @06:15AM (5 children)

    by dry (223) on Sunday October 21 2018, @06:15AM (#751580) Journal

    Last none computerized vehicle I had was an '86 Nissan 4x4. Mostly used wax to control vacuum and did well in the smog test, even without a catalytic converter.

    • (Score: 2) by RS3 on Sunday October 21 2018, @05:44PM (4 children)

      by RS3 (6367) on Sunday October 21 2018, @05:44PM (#751724)

      That's intriguing on many levels. It frustrated me that American car companies were _forced_ to do catalytic converters in 1974 (IIRC). The first gen "gravel bed" ones were terrible, clogged up easily, very expensive, etc. I thought all imports were required in 1980, but maybe import trucks did not require cats until after 1986?

      I've always been a tuner / tweaker. I want efficiency, which brings max fuel economy, decent power, and low emissions. My non-cat 1960s musclecar did better in tailpipe smog test than my computerized catted 1989 Chevy; admittedly the 1989 had a problem and somehow I was never notified of a recall for an incorrect computer calibration. My frustration: I could not tune the 1989. So I studied and learned that I could buy a different but compatible computer and tune away, so I did that.

      Now I need to tune an OBD2 Chevy and it seems my only option is $1K-2K interface. I'd like to find it in the $100 range.

      Wax to control vacuum? You mean a coolant-temperature sensing vacuum control? I always _hated_ all those hoses under Japanese car hoods. What a nightmare!

      • (Score: 2) by dry on Sunday October 21 2018, @07:04PM (3 children)

        by dry (223) on Sunday October 21 2018, @07:04PM (#751745) Journal

        I meant that the catalytic converter was removed after it clogged up, which was/is legal here, which is Canada. The truck was actually made in America and we generally have the same rules as the US.

        The problem is efficiency and clean don't go totally together. You can advance the timing and HC and CO drops close to zero while NOx goes way up. Somewhere is a compromise where everything is low, and while it is efficient, it is not the most efficient. Same with increasing compression, runs better, but higher NOx.

        Yes I meant little wax pellets that expanded as the engine warmed up and allowed vacuum to the EGR for example.
        While in the late '70's, there was a nightmare of vacuum lines, air pumps and other crap, the Nissan Z motor actually didn't have much pollution controls on it. Two spark plugs per cylinder probably helped along with the head design but otherwise, an EGR, spark timing and a anti-dieseling system that also shut the gas of when decelerating.

        • (Score: 2) by RS3 on Sunday October 21 2018, @08:30PM (2 children)

          by RS3 (6367) on Sunday October 21 2018, @08:30PM (#751765)

          Hi neighbor. Yes, agreed on all fronts. I have a '90s vehicle (made in Canada) that cleared out its own cat. converter. Inspectors don't know that it's an empty can. Yes, NOx is the problem child, and like you say, factors fight each other. 2 plugs are a great idea. You may know small plane piston engines have 2 plugs, and as part of "run up" checks you turn off each ignition and note RPM drop, which shows it runs better on 2.

          I'm not deep expert (too many other interests prevent deeper understanding) but I've been reading and tinkering for a long time. My preferred combination, I mostly stumbled on, is high compression (11:1, 60's musclecar), lower coolant temp (160), high energy ignition with fairly aggressive timing curve, tweaked carb running on the lean side, but not to the point of misfire, moderate cam, headers, aftermarket intake manifold, and it all worked very well. Lots of power, surprising MPG (got 22 on a long trip driving reasonably), and very good smog tests.

          Now I just need to be able to tune this OBD2 thing.

          • (Score: 2) by dry on Sunday October 21 2018, @11:50PM (1 child)

            by dry (223) on Sunday October 21 2018, @11:50PM (#751816) Journal

            Yes, doing something like you did to your '60's muscle car would need a lot of reprogramming. My understanding is that during reprogramming, you have to drive it in various ways while monitoring everything, especially the O2 sensors and adjusting the programming as you go. Basically a 2 person job. It's a shame that the better stuff is so expensive, seems to be just a RS232 connector or such and hopefully a serial interface but I don't know. A wireless dongle or USB connected one and the right software in a laptop seems like would be good enough but once again just guessing
            22 MPG, and I take it you mean American gallons, is pretty good from any American '60's car.

            • (Score: 2) by RS3 on Monday October 22 2018, @03:05AM

              by RS3 (6367) on Monday October 22 2018, @03:05AM (#751850)

              There are several aftermarket companies making EFI retrofit systems. "MegaSquirt" was one of the first I remember. Holley, Edelbrock, and dozens of others are on the market. Some use existing ECU (computers) and reprogram them, make wire harnesses, sensor sets, etc. One of many back-burner projects I think about is an EFI for the 60s car.

              Actually many "tuners" (motorheads) disable O2 feedback system altogether and tune for as close to stoichiometric (ideal air-fuel ratio) as possible. They usually use a wide-band O2 sensor and datalog that. Standard zirconium dioxide O2 sensors are pretty much switching devices- not linear.

              Frankly I think the O2 system is stupid. All it does is measure residual O2 in the exhaust. I might be way off base, but I just think you can't be sure of burn efficiency just because of residual O2. Which is why the car manufacturers are moving to wide-band O2 sensing, and it's much better.

              So for my '94 Chevy I built a simple ALDL (computer diag. connector under dash) to RS232 interface and run TunerProRT on a laptop to datalog all the sensors and some of the internal calculated parameters. Occasionally looking at the TunerProRT "dashboard" on the laptop, and otherwise studying the datalogs. Using the same software you retune the parameters, erase and reburn the EPROM (can be replaced by a FLASH chip), drive and datalog some more, and tweak / tune / refine. I ended up with amazingly improved driveability. The incremental changes are usually subtle, but every now and then I'd put an original ROM in and it was amazing how bad the vehicle would run. Again, tuning for better gas mileage, emissions, efficiency, and you end up with better gas mileage, much better driveability. There's a whole world of EFI tuners including shops that will do it for you, dyno tuning, etc.

              There is a 2-port RAM that I could buy and replace the ROM, called a "romulator". It can be updated by the TunerProRT software on the fly- while driving. That would be too distracting, but you could do it at stop signs, lights, pull over, etc., but you can run auto-tuning software that does a much better job of optimizing. In the '94 computer there's really only 1 parameter that is learned, and there are only 16 "cells" for long-term storage. It's very coarse and crude.

              As far as I can tell, GM OBD2 is the only one needing a very expensive interface to reflash the program / parameters. Very frustrating. An alternate approach that some have done is to desolder the FLASH chip in the PCM and program it in a standard programmer. I may do that, and install a socket. It's still pretty labor intensive to just change some tuning parameters. There's supposedly a "back door" into the CPU in the PCM meant for software development. Soon I will buy a spare PCM and try that approach. I just wish I could buy a programming interface, like most other cars in the world, for $50.