Hacker News Reader

Comment by revnode | Hacker News Reader

So Mercedes' solution is to offer a product that isn't usable? Why bother releasing it?

Tesla's solution offers a product that occasionally tries to kill you and people around you. The only reason it doesn't is because drivers are forced to pay attention and take over at a moment's notice at all times.

Mercedes' solution is a car company taking actual responsibility for their software. If they feel the lawsuits/insurance claims/legal snafus are worth the risk, that means their software is probably pretty damn good in that limited scope. Otherwise they could literally bankrupt the company with lawsuits! That's a lot more confidence inspiring to me than Elon's repeated pie-in-the-sky claims.

It's perfectly usable in its intended scope: it allows you to focus on other things while driving in heavy traffic. When they're confident that they can do so safely, they'll extend it to other situations.

Mercedes explains the purpose in their press release:

> Conditionally automated driving on suitable motorway sections where traffic density is high

> During the conditionally automated journey, DRIVE PILOT allows the driver to take their mind off the traffic and focus on certain secondary activities, be it communicating with colleagues via In-Car Office, surfing the internet or relaxing while watching a film. In DRIVE PILOT mode, applications can be enabled on the vehicle's integrated central display that are otherwise blocked while driving.

https://group-media.mercedes-benz.com/marsMediaSite/en/insta...

How heavy does the traffic need to be, I wonder?

Based on that I'm assuming it's following other vehicles rather than following the road, which does make me wonder what happens when you get a clump of DRIVE PILOT vehicles which are all trying to follow each other.

Edit: It seems like it needs a vehicle ahead and under 60kph and uses a mix of other vehicles and road markings. It seems quite usable (but I still have to wonder how many vehicles you can get to follow each other in a chain): https://www.youtube.com/watch?v=0yiPaKfKLZs

It's the other way around. Tesla is selling an extremely expensive beta test that you can't use at all anywhere in any conditions with any kind of safety expectations.

Mercedes is selling a product that has a small set of well-defined cases where it can actually be used.

When I commuted in the city, there was a traffic jam almost every day, and I'd be stuck 15-20 minutes driving at walking speed. On especially bad days it could be up to 45min.

If I could have read my emails in that time it would have been really nice.

The proper solution to this is trains

everyone knows that public transport is the solution to it but I can't buy a railroad track and a train to take to work so people are going to do what they can do

You can't personally buy a railroad but you can vote for people who will apply your tax dollars towards it instead of oil subsidies and auto bailouts :)

let's not turn this into a chain of saying obvious things that everyone involved already knows about, it's not going to add to the conversation and -hopefully- nor will it give you internet points in this website.

With it's current limitations, the only application for Mercedes' solution I can think of is during heavy traffic on highways. But calling it "not usable" does seem a bit harsh.

Of course if you prefer, move fast and brake... maybe

Traffic on highways is also by far the most frustrating part of driving basically for the same reasons it’s an easy-ish target for automation, so seems like a pretty good place to start.

IMO that’s just good product strategy.

It’s useable in certain situations that have a high probability of safety, and allows them to capture data and grow the program safely over time.

Reply on news.ycombinator.com