Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

nytimes.com

Post reply on HN
Print view View original

301–310 of 383 posts

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#301

Post by amyjess »

puranjay wrote: post #96
keeptrying wrote: post #77

Waymo will have the first self driving cars on the road. They are so ahead of everyone. I think everyone including regular car companies buy the tech from Waymo in the short term.

I'm curious where Tesla fits into this scheme. In terms of road miles, I reckon Tesla would have Waymo beat easily. They have far too many cars on the road and thus, way more data. Does this factor into the quality of self-driving tech at all?

I'm sure that Google gathers so much data from human drivers using Google Maps and Waze that they can build a decent model.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#302

Post by repsilat »

maxerickson wrote: post #253
markdog12 wrote: post #242

Earlier quoted context omitted.

Eternal shame on merriam-webster.

A dictionary that fails to document usage is a lot worse than one that documents usage you don't like.

A good dictionary will also document widespread distaste for particular usages, though. It's a losing battle with any intensifier -- we've already lost "really" and "very" to this problem, and "literally" will pass too. That doesn't mean it's not worth trying to extend its life a little longer, though.

("Objectively" is also being used as an intensifier in casual conversation, though I think always with a touch of irony... Good fun.)

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#303

Post by komali2 »

amyjess wrote: post #301
puranjay wrote: post #96

Earlier quoted context omitted.

I'm curious where Tesla fits into this scheme. In terms of road miles, I reckon Tesla would have Waymo beat easily. They have far too many cars on the road and thus, way more data. Does this factor into the quality of self-driving tech at all?

I'm sure that Google gathers so much data from human drivers using Google Maps and Waze that they can build a decent model.

Why do you think Google maps data would be as useful as the camera and steering input data Tesla can collect?

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#304

Post by fixermark »

keeptrying wrote: post #293
jboggan wrote: post #271

Earlier quoted context omitted.

Waymo is going to win because of (1) better sensors and (2) better onboard computing power. Other companies can potentially compete in the former domain but literally no one in the world can go toe-to-toe with Waymo and siblings in the latter department. The ML powering Level 4 driving decisions is not cheap in dollars or power consumption and Google is already way ahead of the curve when it comes to custom fabricati…

The hardware will become commodity. The software is where the action will be. Imagine what would happen with fighter jets if they only had 3 main sensors. Lidar, Radar and camera.

The thing is: in the long run, yes; in the short run, Waymo is manufacturing its own chipsets that it has a patent on. It's not using COTS kit for on-board processing, which gives it a significant technical advantage that will be hard for competitors to match.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#305

Post by maxerickson »

repsilat wrote: post #302
maxerickson wrote: post #253

Earlier quoted context omitted.

A dictionary that fails to document usage is a lot worse than one that documents usage you don't like.

A good dictionary will also document widespread distaste for particular usages, though. It's a losing battle with any intensifier -- we've already lost "really" and "very" to this problem, and "literally" will pass too. That doesn't mean it's not worth trying to extend its life a little longer, though. ("Objectively" is also being used as an intensifier in casual conversation, though I think always with a touch of ir…

Did they add this bit today?

Should literally be used for emphasis?

Sense 2 is common and not at all new but has been frequently criticized as an illogical misuse. It is pure hyperbole intended to gain emphasis, but it often appears in contexts where no additional emphasis is necessary.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#306

Post by yohui »

fragmede wrote: post #282
neosat wrote: post #117

This could be a longer term death blow for Lyft, ironically. If Google has the self driving tech, the car & car partnerships, the money, maps, access to end consumers, what exactly does it need Lyft for? Lyft will be a vehicle for Google to kickstart and test this out with a current ride sharing network before Lyft is cut off ...

And when Facebook came out with Google Plus, we all switched away from Facebook, right? All the engineering genius in the world couldn't do enough for Google Plus (hint: it's not the engineering), why would they do any better with supplanting Lyft unless they copied Uber and lit money on fire by offering free rides for a year like they did with (domestic US) phone calls.

Google Voice still does offer free domestic US calls. They kept renewing the one year free offer until they just declared it free indefinitely.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#307

Post by sova »

fixermark wrote: post #304
keeptrying wrote: post #293

Earlier quoted context omitted.

The hardware will become commodity. The software is where the action will be. Imagine what would happen with fighter jets if they only had 3 main sensors. Lidar, Radar and camera.

The thing is: in the long run, yes; in the short run, Waymo is manufacturing its own chipsets that it has a patent on. It's not using COTS kit for on-board processing, which gives it a significant technical advantage that will be hard for competitors to match.

Elon maintains that with just cameras you can achieve super-human (10x) performance with self-driving cars. Fact I nabbed when he spoke about Boring at TED. https://www.ted.com/talks/elon_musk_the_future_we_re_buildin...

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#308

Post by ska »

aedron wrote: post #162
keeptrying wrote: post #77

Waymo will have the first self driving cars on the road. They are so ahead of everyone. I think everyone including regular car companies buy the tech from Waymo in the short term.

> Waymo will have the first self driving cars on the road Which are still way, way off in the future (if ever). I propose 'self-driving cars' are the 'flying cars' of our generation: Something that seems inevitable and just around the corner, yet here we are.

There is an important difference. "Flying cars" was always a media and popular culture thing - the vast majority of experts thought they were a long way off, if ever. This is not true of self driving cars.

Self driving cars are also a pop culture thing - and the most aggressive deadlines and comprehensive deployments predictions are unlikely.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#309

Post by amyjess »

nodesocket wrote: post #161
searchhn wrote: post #157

Earlier quoted context omitted.

Yes.. as a bay area resident can confirm this. I switched totally to Lyft the past few weeks and have no issues finding a ride promptly. Also all drivers that I met seem to be driving only for Lyft. PS: What convinced me to totally switch was this https://github.com/lyft/envoy , my way of saying thanks for the Lyft team.

This is bay area bubble mentality (I live downtown SF). Does not represent the vast majority of Uber's business and market.

I've observed a growth in Lyft drivers in Dallas, too.

When I first started using Lyft in 2014, there were many times when attempting to request a ride would return an error saying there are no nearby drivers, especially during rush hour (trying to get a ride at 8am was simply a no-go), and it wasn't uncommon for the nearest driver to be 20+ minutes away at various times of day. I actually heard a lot of drivers complain to me about how Lyft sends them pings from people ridiculously far away and saying that it would be better for all involved if Lyft didn't try to match people that far away and just returned an error instead.

Now? Neither of those situations have happened in recent memory. I can't remember the last time I got the "no drivers available" error, and at the absolute worst, I might get a driver 15 minutes away in the height of rush hour.

Edit: Forgot one more thing. I used to see drivers who had two phones for both Uber and Lyft all the time. Nowadays I don't see anyone doing that anymore.

Re: Lyft and Waymo Reach Deal to Collaborate on Self-Driving Cars

#310

Post by maxerickson »

speedplane wrote: post #58
maxerickson wrote: post #16

Earlier quoted context omitted.

New York City cabs seem to stay in service about 6 years, with 400,000+ miles on them at that point.

Based on the NYC taxis I see roaming the streets, I'd guess that they last longer than six years. Plenty of cars look 10+ years or older. I'd guess that that the average age of a NYC taxi is around six years.

The NYC 2014 Taxicab Factbook (a real thing) gives the average age as 3.3 years:

http://www.nyc.gov/html/tlc/downloads/pdf/2014_taxicab_fact_...

It could be somewhat higher now but I doubt it has doubled.

(I had just glanced at https://www.quora.com/How-many-miles-does-a-NYC-taxi-do-in-i... for my previous post)

Post reply on HN