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

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Re:Nope (Score 1) 195

No matter the subject, as AI grows, its capabilities will become exponential.

Except it can't, really; the universe is bound by physics and can't support exponential growth that way.

You can't exceed Carnot efficiency.

You can't defeat the square-cubed law (for things that involve getting resources/waste into/out of a particular volume (including heat - where do you think this AI is going to get all the energy it needs to do all this stuff?).

So while AI might be able to do certain things efficiently, it can't grow without bounds - the universe just doesn't allow it.

Now, that said, AI will probably indeed handle most of the deterministic things in the world - it will probably equalize a lot of sub-optimal things.

But we aren't going to have AI playing sports, we aren't going to have AI taking all creative jobs, we aren't going to have AI "replacing" tourism - you can't "AI" a trip to the grand canyon for instance. You can't "AI" having a family.

So if we let the AI figure out how to distribute all the resources it creates, rather than letting people do it, we'll all be fine. But that is going to be the trick - actually letting the AI do it.

Comment Re:Nope (Score 1) 195

And what happens when one no longer needs humans to do the work since they provide so little value?

Why doesn't someone just use one of these astonishing AI programs to figure it out for us?

Seriously, if the AI is going to do all the work for us, why don't we also let an AI figure out how to transition people to post-scarcity society without massive bloodshed?

Comment Re:Surprised (Score 1) 474

The conference produced data showing that there are approximately 100,000 jobs available in Minnesota that can't be filled because of lack of skills or lack of interest.

That sentence is missing one clause: "At the salaries/wages/benefits being offered for those positions."

A "skills gap" makes no logical sense - if there is sufficient demand for products that you could hire that many people, then there is sufficient demand that you could train people on the job and still afford it. The other possibility is that it's not a skills gap, but a certification/licensing gap, which means you need to work with your certification boards to start allowing more people through the certification process by opening more schools and/or funding more people to get those certifications. Or in some cases, reducing certification requirements. I'm not even talking about medical or emergency services or anything: beauticians for instance - have you ever seen how many hours they have to put in to be allowed to cut hair and apply makeup (granted, sharp objects and potentially nasty chemicals, but sill...)?

Either way, ultimately there is no "gap" - it's a mismatch in labor supply and demand at some price level.

Comment Two Million Man-Years? (Score 1) 292

I don't understand how the costs of this can approach that magnitude (using $100k / man-year as a generous number). The linked article was very sparse on numbers, so it's unclear how many people are being compensated, but even if you compensated ten thousand people 100 years worth of income each, that would only be half the cost, and I don't understand how any huge civil engineering project could cost 1 million man-years of effort. The Hoover Dam apparently only cost $700M in today's dollars - what is involved in the cleanup of things that has the equivalent cost of about 100 Hoover Dams? $200B is also roughly equivalent to the entire Apollo space program.

Mind boggling... that's just how big $200 billion is.

Comment Re:Energy storage and HVDC transmission (Score 1) 275

Right, I understand why people say that a real-time spot market for energy makes sense from a physics standpoint - but I didn't ask if it made sense. I am honestly interested in which is better from a society standpoint - although I guess 'better' is subjective, so the more precise question would be, which system ("overproduction and/or excess storage" vs "real-time spot pricing") is better for society in terms of ability to plan (from a consumption standpoint), deal with disruptions, and have lowest pricing impact (without subsidies or other redistribution mechanisms) for the least-advantaged population.

Comment Re:Seriously? (Score 1) 88

>No, it's that it should be something a child is actually drawn to, not an activity forced down their throat to perform in lock-step with thirty other classmates.

Why single out computer science here? Why mandate English, math, science, etc. for students?

Because the sad reality is that a student has to apply for a major in college *prior to taking classes at that college*. So they need to be exposed to every subject they might be interested in in the K-12 system, and maybe they don't know that they'll like or dislike a subject until they actually take it.

Most people who become biology majors like biology in high school. Most physics majors took physics in high school and thought it was something they could do.

Most computer science freshmen go into CS because they like video games.

Comment Re:Clicks are all that matter (Score 1) 158

>You are experiencing false nostalgia. There was never a golden age of "real" journalists.

The media has always been biased in favor of the establishment in America. (Which cuts deeper than the left-wing bias, really.)

That said, in the 80s there was a distinction between News and Opinion on TV. News reporters would make at least a token effort of presenting both sides (equal treatment principle + fairness doctrine), and opinion pieces were often not found at all in news programs. Newspapers separated out the News section from the Editorial section (and would often run pro- or con- columns on whatever was the pressing issue of the day).

Now we have news "personalities" getting choked up on live television when Trump wins, and the New York Times completely taking its mask off and just running "FUCK YOU, AMERICA! I THOUGHT WE WERE BETTER THAN THIS! I HATE YOU TRUMP, WARRRGHGHGHBBLBLBLE" headlines all over the front page. I took screenshots. The schadenfreude was delicious, despite my boy GJ not winning.

Now they're pushing the narrative of "fake news" being the reason Trump won, because, naturally, self-reflection would be too painful.

Comment Re:Nice to see we'll be in better hands (Score 1) 313

>Opinions that all turn out to be true

??

Did Bush actually turn out to be Hitler in disguise? Because that was their main bit with him.

Democrats only have three possible responses to any Republican in the world:
1) He's Hitler.
2) He's racist.
3) He's a fool.

(Replace He with She for any female Republican, such as Sarah Palin - guess which one they picked for her?)

The American electorate woke up to this fact a while back, and aren't buying it any more. Bill Maher said (about a week before the election) that maybe it was a mistake to call every Republican candidate Hitler, because *now this time he's really Hitler*. Lol.

Comment Re:It's not rigged, you're just LOSING (Score 1) 297

The statement being copied went something like "Stein raised more money this weekend than she did down the whole home stretch of her campaign" -- the point being not to compare how much, but rather how FAST it was raised. And that is indeed a good point. If she can raise $3M in one holiday weekend, how could she not raise similar amounts that fast at least occasionally all last year?

That's why I think the guy who pointed out it's probably a guess-whose-bot doing the donating is likely correct.

Comment Re:Lets call Bullshit (Score 1) 297

Someone noted that the pace at which the money is coming in indicates a bot, not real humans, so is probably almost entirely from a single source. I leave it to your imagination who that that might be...

Also odd how ...curiously nonspecific... is her statement about what will be done with any leftover funds. I translate this as "convenient way to refill my war chest".

And funny how the whole notion didn't get serious until it was almost too late to do anything about it, even if a recount were called for.

I had regarded Stein as honest, if politically misguided. I may have to reassess that.

Comment Re:Energy storage and HVDC transmission (Score 1) 275

Please tell my why extremely variable prices are better than stable prices?

Trying to manage "instant" pricing also requires massive communication infrastructure and generally puts poorer consumers at a disadvantage. (Not necessarily poor monetarily - poor in ability to respond to the price variation.)

I'd much rather have a system where prices are essentially fixed and stable where we have enough system storage capacity to address any supply/demand variability and not require every point-of-use to have to adjust behavior due to spot prices.

Comment Re:Beating by only 15% is not a breakthrough (Score 1) 29

Capacity is great, and so is total life, but where's all the breakthroughs (potential or real) to improve recharge time? I want a battery that can charge fully in 5 minutes. You could have vehicle-battery ranges no better than today (250-ish miles) but with 5 minute recharge times, you eliminate 95% of the complaints about electric vehicles.

Realistically I just don't see this happening - trying to charge even just a 50kW-hr battery in 5 minutes requires at least 600kW electrical power; a nice 100kW-hr battery would require at least 1.2MW. Mechanical battery swap systems seem like a non-starter, although I suppose they could be more feasible if maybe cars took more, smaller battery packs?

My hopes are still on liquid-air batteries for transportation, because you get the benefit of high-efficiency electric motors, the higher energy density of only having to carry the fuel (not the oxidizer, at least initially), and the benefit of fast recharge because liquids can be pumped quickly without high voltages and currents.

Slashdot Top Deals

For every problem there is one solution which is simple, neat, and wrong. -- H. L. Mencken

Working...