Why Tesla Is Going to Win in Autonomy | Hacker News Reader

Why Tesla Is Going to Win in Autonomy

medium.com

11 pointsgiacaglia6 comments

* The given description of Tesla's "Shadow Mode" has been bouncing around for a while, but it's not accurate (see https://twitter.com/greentheonly/status/1096325954375430144 for a description of what Tesla actually does).

* The description of LIDAR having "much more detail than other sensors" is confusing. LIDAR has more detail than radar, but LIDAR is much less detailed than cameras.

* The description of the 2016 accident is not correct. MobilEye provided the EyeQ3 chip on the car, not just the software. "Since then, Tesla added radars for these situations" is flat out wrong; Tesla has always had radar as part of AP, and radar didn't (and won't) prevent that sort of accident (https://twitter.com/artem_zin/status/748623772056428544)

* 'the newest Tesla Autopilot software 9.0 has the largest vision neural network ever trained' wut?

* 'The DARPA Grand Challenge began in 2004'...'the winner, Waymo' wut?

* 'systems based on if-then statements' ?

* 'Comma.ai continues to develop its software, and drivers can buy and install it in their cars' the software is free, it's on Github.

Furthermore, the thesis in the HN title is not supported at all in the post as far as I can tell.

Sobering to read that Waymo simply couldn't sell their software stack by itself even though that was their plan A, a la Android. The only reason that allows me to make sense of this is that old-line manufacturers just couldn't bring themselves to believe that software (albeit nets not code) could be quite valuable. Which may mean they're stuck in a very old-fashioned head-space.

Or perhaps they realize exactly how valuable Waymo is, and how dangerous it would be to ally with them. The contract probably stipulated that auto manufacturers cease efforts to develop a competitor to Waymo, meaning they'd be tied to Alphabet forever. Under Android, most OEMs have razor-thin profit margins and little ability to customize the stack.

If auto manufacturers think there's enough lead time for them to develop a viable competitor (Cruise), why should they knuckle under to Alphabet? Cue lobbyists fighting to keep each other from DMV certification..

I'm not assuming contract lock-in (there's none with Android, you can drop out at any time.) Android has created competition that reduces mobile hardware to a commodity, but that pressure hits competitors with different OSes, too. (For example, in China.)

PS - you are forbidden (strictly illegally IMHO - bundling law - but you are) from distributing Google apps with phones that employ the Android libraries but not the full Google Android OS proper; but if you avoid Google's libraries you're good to leave for any other OS and still OEM their apps.

I haven't been especially clear. I said "could be valuable" - I might also have said "could be hard to replicate." Nets are hard to create because they need masses of data to train as the article says. Paradoxically, at the same time they're just about impossible to effectively protect as IP; functional duplication of a net is (nearly) always possible if you're willing to put up the money. So why not pay less to rent the software now and wait 'till recreating the nets is far cheaper or already largely available as open source? I suspect the old-line car companies have tossed away a ton of money because they had zero understanding of neural net commerce or economics.

i seem to remember at least one other company doing software and hardware...

lol the author 'forgot' about Cruise/GM

but what's that compared to the copious knowledge gained by haunting the tesla subreddit

Reply on news.ycombinator.com