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

 



Forgot your password?
typodupeerror

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: finger pointing (Score 1) 396

by Goldsmith (#49351433) Attached to: Millennial Tech Workers Losing Ground In US

I think at this point everyone agrees that the STEM job market in the US is screwed up. Right now we're all pointing fingers at eachother blaming millennials, gen X, baby boomers, immigrants, business owners, politicians, civil servants, the whole government, high schools, colleges, testing services, misogynists, political correctness, investors, people who don't invest, Obama, Bush...

Anyone have any ideas on what to do about it? How about we work on that now.

Comment: doesn't make sense (Score 2) 149

by Goldsmith (#49320785) Attached to: Obama To Announce $240M In New Pledges For STEM Education

There are billions poured into STEM, and encouraging early career scientists through programs at NSF, NIH, DARPA, etc. None of that is working (less than 50% of people trained in science stay in science). When I was still training students, the best of them generally ended up working in finance, not physics. An additional $250 million is not going to make a notable difference. We need a cultural and structural change in how we train and retain good scientists and engineers, not a meaningless bandaid.

Comment: anyone actually beeen here? (Score 1) 415

by Goldsmith (#49315971) Attached to: How 'Virtual Water' Can Help Ease California's Drought

If you've been through California's central valley, you know that it's not a desert. We're talking about some of the most fertile farmland in the world. More than half of the USA's fresh produce comes from California. Just the almond market alone is $2.8 billion a year. Despite that, California's central valley is also one of the poorest, least educated populations in the USA.

Given all that, is "screw the farmers" really the best solution here? Maybe we should make more fresh water? This isn't theoretical. The largest water desalination plant in the western hemisphere is being constructed in San Diego. It "only" took 18 years of regulatory and legal wrangling and $1 billion of financing. We need about another dozen of these plants to make a real impact on the statewide water supply. Now that the regulatory and legal framework is set, increasing the cost of water to construct additional desalination plants and related infrastructure would make more sense than choking agriculture out of the state.

Comment: they're missing something (Score 1) 112

by Goldsmith (#49288817) Attached to: How To Make Moonshots

It always seems like these guys are missing something. Bell Labs had this figured out. Arguably, IBM has done a better job of this in recent decades. I get the impression the people at Google just kind of heard about a bunch of failed DARPA projects and decided to try and fix them up. Self driving cars, enhanced reality headsets, balloon based networks, nanoparticle diagnostics, jetpacks, neural network enhanced computer vision... all legacy military development projects... all very cool, but not really lightning strikes of inspiration.

Is it that they don't have the right people? Are their projects too fast? Are they too structured? Maybe they're purposefully trying to only do things others have already tried and failed for some reason. Perhaps it's that they've forgotten the difference between innovation and invention. Innovation builds successful companies, but they'll need a hefty dose of invention if they really want a "moon shot."

Comment: Re:Space for solar hasn't been much of a concern (Score 1) 437

I realize from the rest of this discussion that you're in Alaska. Down in San Diego, solar covered parking is fairly common. We tend to have acres of parking lots at big businesses, malls, car dealerships, etc. These have been prime locations for solar parking shades.

You can apply some financial/government subsidy wizardry to mitigating the cost of solar. Down here (for some houses), if you're willing to commit to paying your current average power bill every month for the next X years, someone will come and install solar on your house for free.

Comment: Re:it's the system (Score 1) 320

Intended for? You mean the motions of the planets in the solar system, the first example of where Newton's laws broke down? Newton's laws don't work where they were first used. Even Newton knew that: he couldn't predict the moon's orbit properly without a fudge factor. Take a deep breath, no one is saying mathematics goes bad. My point is simply that gathering better data, and producing better ideas is how science works.

Scientists try to understand why something happens, not just how it happens. Getting close enough for practical applications and predictions is enough for engineering, but in science you have to get the "why" right.

Comment: Re:it's the system (Score 2) 320

Don't be absurd. Of course I've heard that F!=ma. That's well covered by the time you finish a Physics PhD.

Newton's laws are perfectly good approximations for most cases, but they're not always valid. Newton was wrong, that was the point of relativity. And yes, we're looking forward to correcting relativity when we do figure out dark matter and energy. Einstein had the intelligence to know he was wrong when he formulated general relativity; no one has figured out how to fix it yet.

Comment: it's the system (Score 1) 320

There are two answers to this, the first is the easy answer:

Science is often "wrong." This is how science works: you come up with a theory or some measurements, support it as best you can, but expect someone to do it better in a few years. Often "better" means results so different from what was seen before that the prior work is now considered "wrong." As we get better at science, this happens faster.

The second answer is a bit more complicated and acknowledges that there is a real problem.

To me, this is real and it's due to the recent loss in prestige and ability in government/industrial labs combined with the emergence of the internet. This led to the use of journal publication metrics to arbitrate scientific disputes instead of government or industrial validations. (This is different from the problem of sponsored research.) Using publications to "decide" scientific rightness instead of independent validations has also put immense stress on the peer review and publishing systems. Use of fast-but-incorrect techniques, shortcuts, and repetition of boilerplate language is very effective at rapidly generating publications, and thus is more "scientifically correct" in the current system. This is happening while the public has more access to this content that should not be reasonably expected to contain absolute truth.

Comment: the math doesn't work (Score 1) 245

by Goldsmith (#49133871) Attached to: The Peculiar Economics of Developing New Antibiotics
In TFS, it's agreed that it costs $1B to develop a new antibiotic.

The success rate for drug development is about 10-15%.

Now, you're probably spending $1B cumulative on all the failed drugs to get one hit. They key here is that you're not actually guaranteed to get a drug that works. You could easily spend more than $2B on a program like this, with a little bad luck.

Let's look at this differently. About 250 million antibiotics prescriptions are given out in the US every year. Let's have every one of those pay $10 over cost of manufacture and marketing (for example) to the drug companies who have developed new antibiotics in the prior 10 years (that collective effort helps all the antibiotics companies). Now you're spreading around an "extra" $2.5 billion every year, not just once. That's going to compensate for higher risk approaches more quickly and contribute to a longer term solution for this.

Comment: this doesn't help anyone out (Score 3, Informative) 681

by Goldsmith (#49108749) Attached to: Bill Nye Disses "Regular" Software Writers' Science Knowledge

Ok "regular software programmers." Go actually read the article, and then come back and read the summary again.

Now, Nye was trying to say that our technical work force is not trained in enough science. Maybe that's right, and maybe it's wrong, that would be a better discussion for Slashdot. Nye (or the reporter) obviously did a bad job here. At the same getting offended at being called less scientifically literate than the top tier of scientists doesn't help either.

Comment: the particle physics culture problem (Score 1) 89

by Goldsmith (#49063081) Attached to: Scientists To Hunt For Supersymmetric Particle In LHC

Particle physics did an excellent job building a multidisciplinary, international, scientific workforce. As a field, they are largely independent of the world of 12-36 month grants and frequent peer reviewed publications the rest of us live in. More scientific fields should look to particle physics for guidance on self-organization and priority setting.

However, in the process, particle physics has separated itself from general physics. Outside of some cosmologists, there are not many other physicists who can (or care to) work with particle physics colleagues. We were on board for Higgs, but I think the physics and more importantly, the culture, has veered off so far from what we're used to that it's going to be hard to justify discoveries as "fundamental."

Comment: an old discussion, with new jargon (Score 1) 226

by Goldsmith (#48948775) Attached to: There Is No "You" In a Parallel Universe

This is a rehash of the Boltzman's Brain paradox, which doesn't require quantum mechanics, just infinity and statistical mechanics. It's a line of thinking in physics that goes back at least 80 years and probably back to the late 1800s. This doesn't mean it's wrong or bad, just that generations of physicists have thought about this (usually with a beer or two) and there's not a hard physical answer to the question: do I exist somewhere else in the universe?

It comes down to one little bit in that article: the universe could be infinite, and may have been infinite since before the big bang. The rest is the same line of reasoning about the improbability of growing toward infinity (gravitation at first, limits on inflation now) that we've been looking at for many generations. We're pretty sure we're not growing into an infinite universe. We still have no idea if the universe started off infinite. Addressing that is a bit outside of what we can currently do.

Comment: press release creep (Score 1) 38

Nanotechnology is particularly bad about press release creep. That's when the author of a paper publishes

"The proximity-induced ferromagnetic order in graphene can lead to novel transport phenomena such as the quantized AHE which are potentially useful for spintronics."

and it becomes

Graphene: Reversible Method of Magnetic Doping Paves Way For Semiconductor Use

Comment: Re:Hang on WTF? (Score 1) 191

by Goldsmith (#48856701) Attached to: Japanese Nobel Laureate Blasts His Country's Treatment of Inventors

Things may be different in Japan, but you do not understand how this works in the US. Investment in research doesn't mean you own the work.

Having a solid IP assignment agreement with a scientist and a strong cultural and political expectation of ownership is what determines who owns IP. Without a legal IP assignment contract (wording which has survived a court challenge, and an agreement in which both parties benefit - this is where investment comes in), the work IS owned by the inventor.

In terms of investment, you have a physicist who has invested $500k in specialized training (the current estimated personal cost of scientific training over 10+ years). In the US, the government funds the majority of training and early research (~$2-4M) through universities (who on average come out ahead financially in this arrangement). In the last stage, a company funds the final research leading to development (~$500k-$1M).

Who owns the patents to the work? In the US, the fundamental patents are owned by the university that trained the scientist. Universities generally don't invest in research, they get someone else to pay for it. Big universities generally don't count lab scientists as employees anymore (they're all contractors and 'visiting scholars' now). But, they do secure solid IP contracts from every contractor, student worker and professor who works on campus.

There's a very good argument that all of these government funded university patents should be owned by the government. Government grants generally include clauses claiming some ownership of IP generated. It's too bad that politically, that's not something that can be enforced.

Is it possible that software is not like anything else, that it is meant to be discarded: that the whole point is to always see it as a soap bubble?

Working...