Post by
MinisterofDOOM »
https://forums.nicoclub.com/ministerofdoom-u16506.html
Sat May 06, 2017 5:18 pm
I don't have default navigation in my car, and I find that when I'm driving a car with built-in navigation, the interface is always so clunky that I just fall back on Google Maps anyway. I used Waze for a while, but its route finding algorithm is beyond broken. It tends to seek out low-speed residential roads with fanatical zeal, bypassing much quicker and less congested arteries. The good news is: since google owns both, you get congestion and accident updates on either app.
As far as having it on the car screen rather than the phone: yes, I think there's value. However, there are sort of two degrees of this.
1: Things like Apple Car Play and Android Auto are a great way to relay your phone's navigation to the center screen in your car, with basic interface support. But it's very limited for liability reasons, which is a pain. Still, it's better than holding a phone up or needing a windshield mount to eliminate the need to take your eyes off the road. It also means you can have your phone safely stowed and charging (since ongoing GPS can eat away at battery).
2: What Apple Car Play and Android Auto DO NOT do is link with the secondary screen in the instrument cluster. Most in-car nav systems can at the very least relay the next turn on a route through the screen in the cluster. Many can show a normal overhead nav view (though usually condensed). Right now, you can't get this by linking your phone for Google Maps or Apple Maps.
Really, in car navigation is going to be terrible until there is some standardization. We need common modularity support across all platforms, and we need reasonably standardized capabilities between vehicles. Do what you want with the GUI (although if it's anything less than flawless, people like me will never forgive you). But the core needs to be a very lightweight universal backbone that supports externally relayed app information (without a need to download a companion piece car-side) with standardized multi-screen support and interface scalability. Then anyone can develop their own navigation, media, or whatever app and simply use the infotainment system as a remote client to display info and sent interface commands back to the host phone.
This is what made the smart phone work. This is what let Palm dominate PDA sales before smart phones. This is what allowed the home computer market to really flourish. When everyone is building their own wholely disparate system (sometimes many of them, each for a different model of car) with no concern for any kind of standards or future-proofing (updates, security [these things are usually open to public networks with NO security!!!!], you're just going to have a whole bucketload of bad products. But as soon as the industry grows a brain and starts to come together to at least establish some interoperability standards and a general scope-of-features for a certain degree of consistency, the in-car infotainment system is going to flounder in subparity.
The real problem right now is that it's automotive engineers that are designing these systems, not software engineers. Correcting that is the first step toward improvement. Then we can start standardizing. Then we can start including these systems as what they really are: a network-connected OS for a basic RISC hardware architecture. In any other industry, this would be the most basic of steps, but for some reason putting these things in cars caused their design to be tipped upside-down, and what we're left with is really extremely terrible.