Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:O(n^2) in memory slower than O(n) writes to dis (Score 1) 474

by grimmjeeper (#49336497) Attached to: No, It's Not Always Quicker To Do Things In Memory

In other news, Pope found to actually be Catholic. The Pontiff was quoted as saying. "I always knew I was Catholic from when I was a little boy."

This just in. Massive government study shows bears do defecate in the woods. Head of the $65M (£43.6M) government funded study, Dr. Hans Schmidt, describes the study. "Ve always knew ze bears did zeir business somevere but ve were never sure vere zey did it. But now it is confirmed. Zey do zeir business in ze woods."

Comment: Re:From another article... (Score 1) 341

He's been a successful entrepreneur, no doubt. But he really doesn't have the background in actually building safety-critical systems to fully understand the complexity of the problem. Sorry but I'm going to go with actual engineers who have done the actual work instead of the flashy business guy with no real experience actually building it.

That being said, I'm 100% in favor of him putting resources into developing the technology. It's good that he has many engineers working on the problem because the long term benefits of the work will save countless lives. It's just that he's wrong about it being an easy task. It's hard and it's going to take a lot longer than he thinks.

Comment: Re:From another article... (Score 1) 341

Yeah, I added that thought in the follow up to the first response to my comment. If we can get fewer accidents per mile driven than we get now, it's a win. The more we reduce accidents the better. Which is why it's good to see accident reducing technology showing up in cars long before the cars actually drive themselves. And even if that's all we ever really get out of the effort to have self driving cars, we're still better off

Comment: Re:From another article... (Score 1) 341

If it happens like the evolution of safety in aircraft systems, you'll see horrendous crashes that kill people being the impetus behind recalls and safety upgrades. There's simply too many permutations for anyone to accurately predict a complete set of potential hazards, much less find a way to get a computer to identify them all and come up with a strategy to deal with them. There will be some combination of issues that causes a crash where people die and everyone who has ever worked on autonomous cars will say "I never thought of that".

That being said, autonomous cars don't have to be perfect to be a success. They can still have some accidents due to unforeseen combinations of problems. But If we can get autonomous cars to the point where they cause fewer accidents than humans, we're ahead of the game even if they aren't perfect. And given how poorly so many people drive, the bar is lower than it should be. Nevertheless, the more accidents we avoid, the better off we all are.

Comment: Re:HUH (Score 1) 341

Trouble with flying cars is that they consume more energy than ground cars for every mile they drive. And while oil is cheap right now, it' still quite a bit cheaper to roll along the ground than hovering above it. Granted, things like maglev trains technically are more efficient but they're a special case with a great number of caveats that account for their efficiency. Something as versatile as a car that can go (almost) wherever it wants will always consume less energy if it doesn't have to float over the ground. And that efficiency is really what keeps flying cars from really taking off (pun intended).

Comment: From another article... (Score 4, Insightful) 341

jalopnik article

'"It's much easier than people think" says Musk, outlining how most of the sensors and systems available right now can handle self-driving duties on the freeway, something Tesla showed off late last year with its AutoPilot features.'

As someone who has spent a career working on safety-critical real-time systems, I can assure you that it's not in any way "much easier than people think". Quite the opposite. Sure, driving a car down a well marked highway on a clear sunny day with little traffic and no system failures is easy. But if you obscure the lane markings in any of a number of ways, add inclement weather, throw out random obstacles, random system failures, etc. the problem gets monumentally harder. Throw in an urban environment with all sorts of other issues just keeps making it harder and harder. And solving all of those problems takes up well over 90% of the effort when designing an autonomous system. Hell, developing something that can recognize the problem in the first place is hard enough. Being able to differentiate between sensor failure and sensors indicating a failure is a non-trivial task. He's full of it if he thinks we're anywhere near having a self driving car that's ready for public consumption.

Sure, there are self driving cars out there on the road. But they have huge engineering and support teams using them as an evaluation platform. And it's good that we have made as much progress as we have. I look forward to seeing the work continue and advance the technology. But it's not an easy task. It's going to take probably decades before we're really ready for a fully autonomous self driving car that's ready for public consumption. We'll probably see some of the technologies work their way into cars between now and then. And that's a good thing too. But it's not going to happen overnight because it's much harder than people think.

Comment: Interdasting... (Score 3, Insightful) 155

by grimmjeeper (#49279493) Attached to: White House Proposal Urges All Federal Websites To Adopt HTTPS
It's not a bad idea to run HTTPS. It makes it inconvenient to hack connections and makes people work for it. But I found this quote to be amazingly ironic: "Unencrypted HTTP connections create a privacy vulnerability and expose potentially sensitive information about users of unencrypted Federal websites and services."

Comment: Re:Eqaul Protection (Score 1) 759

Which is why the rich can break the law with impunity since the fine has no real impact on them. So what if you get a $200 fine when you make more than that in an hour? That's not a punishment. Making the punishment actually mean something to everyone would make the punishment equal between everyone. Points on your license which lead to suspension are more meaningful, especially if you get jail time and extended suspension for driving on a suspended license. I don't like the idea of community service because the value of people's time is different. The guy working 3 jobs just to keep his head above water would be impacted more by community service than someone who is comfortable in a 9-5 job. The idea of targeting a fine based on daily spending money works well. Figuring out a fine that would be a noticeable inconvenience seems fair. A lot more fair than being devastating to some while barely noticeable to others.

After Goliath's defeat, giants ceased to command respect. - Freeman Dyson