Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?

Comment Re:communications system? (Score 1) 149

And if it's a kid running out into the road, too bad for it too?

This is where a human can make better value judgements than a computer. Is what just ran into the road a large dog or a small child? Different answers can mean different reactions.

Last I checked it was the car behind's responsibility to keep a sufficient distance to avoid a collision, they're never going to win in court that it was your fault for braking too hard.

There has been a series of public service spots running on US TV for a while now dealing with drunk driving. One shows a woman being rolled into a hospital with a gaggle of medical people trying to keep her alive. The husband/driver of the car she was in is following behind with a cop accompanying him. "Oh no, oh my, oh this is terrible, I was just a little buzzed..." he says. The gurney stops, the doctor in charge turns to him and asks "just a little buzzed? Why didn't you say so?" The medical people all step back, the woman on the gurney pops up magically healed and says "ok, let's go home then."

"Can't win in a court of law" doesn't fix the injuries the driver of the car that is rear ended suffered.

So yeah, bad for the people who are glued to your bumper I guess but I won't miss them too much.

Interesting new research just in shows that when two cars collide, sometimes people who aren't responsible for the collision are injured, too. Sometimes they even die from those injuries. "Better dead than red" was an old saying. I think the defensive driving advocates have a similar saying.

Comment Re:It's simple (Score 1) 452

You haven't given a reason why that should be prohibited but "third-party incrimination" (i.e. requiring answers from third-party witnesses) should not be.

Because without incrimination there can be no "third-party incrimination." What the third party is testifying to cannot be used to incriminate him. If it could, then the fifth amendment applies.

Look up the word "immunity". It's granted to "third parties" whose testimony might incriminate themselves but the information is important enough to waive those charges.

Comment Re:It's simple (Score 1) 452

But this is the entire question: Why is it an "abuse of government power" for the government to make the defendant answer questions, but not for the government to require third-party witnesses to answer questions?

Because the government is not seeking to punish the third party witness and is seeking to punish the defendant. The "innocent until proven guilty" concept plays a part here -- there is no "until proven guilty" connected to the witness.

It isn't a case of corrupt or incompetent courts. It doesn't have to be a case of rigging a trial. If I'm a prosecutor and I have enough evidence to convict you, and I can demand that you testify, I'll call you as witness number one. If you tell the truth and admit guilt, my case is done and you've saved me a lot of time and effort. If you lie and claim innocence, I'll present the same case I was going to in the first place and tack on perjury charges.

You know that, so if you're on trial and I call you, you can 1) give up and confess, guilty or not, 2) lie and get the extra time for perjury when I do get a conviction, or 3) bet that my case isn't that strong and lie, hoping I can't actually convict you. You might notice that this 3rd option is just like a plea bargain, but it carries the extra penalty of even more jail time if you lie and get caught, compared to refusing a deal and getting prosecuted. It's like handing a bigger hammer to the prosecutor if you allow him to force a defendant to testify. The hammer they have is already big enough (go to trial and spend a lot of money and time and maybe get convicted anyway at full sentence, or confess and get a reduced sentence). Turning that into "go to trial and get a full sentence for the crime AND a sentence tacked on for perjury, or confess and get the full sentence for the crime" (if there was going to be a plea, it would have already happened), under threat of option 3 "refuse to answer and go to jail for that, too", is an abuse of power.

It's also opening a door for a cross examination. If a defendant doesn't testify at all, you can't cross him and get him to slip up in telling his story. And if I can force you to testify, I can ask you more than "did you commit crime X", I can ask you about events leading up to the crime that maybe I can't prove but would convince the jury that you are guilty. "Did you have a serious argument with the victim prior..."? You now have to consider "can he prove I did?" before you lie, and if you lose that bet you get perjury charges even if I can't prove the whole crime.

What's an argument from first principles as to why you can't require answers from the defendant (who at least might be guilty),

Very simple. It's the prosecutor's job to prove guilt, not the defendant's job to prove innocence. That's the perfect world explanation.

but you can require answers from third-party witnesses (who are known to be innocent)?

If they are known to be innocent then there is nothing to punish them for. What they say cannot be used to put them in jail.

Comment Celebs (Score 1) 176

I only heard about this because Pee-Wee Herman shared the news of it, and of his application, some time ago. I'd like to say I hope he makes it, but then we'd be without Pee-Wee Herman here on Earth.

I forwarded the news to Jane Wiedlin (Guitarist for The Go-Gos) who said she'd be interested in going, but I never saw anything on her FB that said she had filed the paperwork.

I tried to get a couple of other celebs interested but so far all I know is Pee-Wee Herman put in his application. Wouldn't it be interesting if he wins the contest?

Comment Same performance as flash? Ha! (Score 1) 201

Putting 8GB of flash cache in front of a 5400RPM hard drive is not going to give you the performance of a pure flash drive. I don't care how good your caching algorithm is or how many rigged benchmarks you win (comparing only on sequential read/write doesn't count!), you're not going to be as fast. Particularly since flash scales performance with size - a 64GB SSD will be faster than an 8GB SSD of the same type, ignoring any hard drives it may be a cache for.

Will it be "SSD-like performance"? Probably, yeah. If their caching algorithm isn't complete shit, it'll probably be somewhere in the upper half of the two orders of magnitude that separate flash and disc. But "within an order of magnitude of" and "equal" are not at all the same thing.

Comment Glass the whole subcontinent for all I care (Score 1) 659

Normally I'm all in favor of supporting revolutions against unjust regimes, and Assad's regime is clearly evil enough to be worth deposing. Libya? Egypt? We should have done more to help.

The problem with Syria is that the rebels seem equally evil. It seems to be more about sectarianism than equality, and the fact that one side is supported by Russia and the other by several terrorist groups certainly doesn't help.

Had we come in earlier, supporting the rebels while the movement could still be turned towards "revolution against a tyrant" rather than "revolution against someone who follows a slightly different version of our religion", before Hezbollah and al-Quaeda could get too invested, then we could have done some good. But now? If it weren't for the substantial numbers of civilians, I'd say to just bomb the place to the ground. No matter who wins at this point, Syria and the world as a whole end up losing.

Comment Re:in USA FDA is butting in (Score 1) 314

I'm fairly convinced that the emerging brouhaha over e-cigs has really nothing to do with health risks.

I think the motivation by the FDA is over turf and a possibly legitimate desire to manage what is a drug delivery system. From what I've seen the chemistry of e-cigs isn't terribly complicated, but in the gold-rush kind of mentality associated with a product like this, it's not hard to see a new player trying to make up some margin with poor fluid.

Outside of that, I think the motivations are far more dubious and have more to do with the American gut instinct to ban something because somebody might be enjoying it. There's also the quesiton of money, government at many levels gets a ton of money from cigarette taxes and ongoing tobacco settlement taxes. A long-term switch away from tobacco by smokers will have a lot of states wondering where to make up the difference.

Comment Re:You can switch it off. (Score 3, Funny) 195

Nah, there's no reason to give them an entire continent, especially since there's a slim chance they could actually survive there. I'm thinking Ilha da Queimada Grande, colloquially known as "Snake Island".

Why is it called Snake Island? Well, there's a lot of snakes there. Just one species - the Golden Lancehead, which is extremely venomous. But legend holds that there are so many of them, they cover the island to a density of one snake per five square meters. Oh, and they can live in the trees. The island is so dangerous the Brazilian government (not particularly famous for caring about the safety of its people) has prohibited people from even visiting.

With all those snakes, I'm sure the politicians will fit right in.

Comment Re:Enough is enough. (Score 1) 417

Do you really think climate science is settled as much as gravity?

Come on.

The models they were certain of only 8 years ago turned out to need large corrections.
The extra storms were were supposed to see are lacking.
This is a theory which is undergoing large adjustments in real time.

It takes a long time for science to really settle down. At best, we've been looking at this about four decades. Prior to that there was serious consideration of the cooling trend. it wasn't until the early 70's that the majority of climate papers shifted around to warming and cooling papers pretty much stopped by 1978.

It's the insane certainty and the constant mistakes and misstatements and the excessive certainty which is a problem.

And it's a problem that several people who oppose climate warming are in bed with the oil companies and pretty skeevy. It gets some people emotionally invested in climate change. And people exploding and over reacting and making extreme statements always makes me a bit suspicious.

I'm not in a position to affect things in any way. and my votes generally go to the pro-climate change politicians.

But seriously- there is probably nothing we can do on this one. If it's humans, the 50% population increase we'll see during our life times is going to break all kinds of ecological systems. If you are serious- talk about getting the population back down to 4 to 5 billion. Otherwise, climate change people are really just pissing into the huge wave of oncoming humanity.

Comment Re:A screen (Score 1) 185

it's easy to claim you just came out of Walmart and somebody must have done it while you were in there. There's no way they can prove it's not true...)

You picked the wrong target. Walmart is one the most camerad places on earth. Loss prevention. They'll have the tapes that show your car was not approached by anyone but you, and since they'll want to keep cooperation with the cops for their loss prevention programs they'll be happy to cooperate with the cops for copies of the tape.

Comment Re:Deliberate stupidity [Re:Superstorm Sandy?] (Score 1) 417

So it turns out that the forecasting did work.

Kinda like the forecast "60% chance of rain" is right either way.

You do realize that all models can do reasonably well at forecasting if the forecast period is not very long, don't you? That the "validity" of a model is never judged by a truly significant forecast period, but by how well it hindcasts. And that every model that does long range forecasts has to ignore factors that it cannot predict, like solar output, right?

It's hard to claim that forecasts made 15 years ago are correct when it was about ten years ago that the satellite remote sensors realized there was a mistake in their calculations and the true surface temperatures were a couple of degrees off of what they were claiming. All the input data was wrong, so if you claim the output is right then the model has to be wrong. Or good at guessing.

Any model that is correct at 100 years is luck more than science.

Slashdot Top Deals

The first version always gets thrown away.