Your basic argument is that this has not been a problem in the past so it never will be in the future. I think this forgets the one of the fundamental points of technological development: to reduce or eliminate the need for human labor. Engineers developing automation are always striving for this goal, and always getting closer. What if a day comes when robots actually can eliminate the need for human labor, or at least for low skilled labor? I think the article is suggesting that this day is approaching, and considering possible consequences on our economy and society.

Now, our system might adapt. It might all work out. Sure, as is often pointed out here, our economic system has all sorts of flaws. But it also has lots of benefits that the average slashdotter does not give it credit for. It is very possible that the system will successfully compensate for the changes introduced by near perfect automation. But assuming that everything will work out because it always has is just naive.

Also, considering these changes as they happen and discussing ways of managing them is a critical part of a functioning democracy.

Yep, those are interesting. Thanks!

Engineering Disasters was published in 2005; it does not cover these specific rebuttals. When discussing an older rebuttal he points out that 11 U.S. helium filled blimps were lost by fire (more details needed on this point).

After a brief review of your links, my comments are:

1) To conclude the validity of the different sources would take a large amount of time. By the time I could possibly finish, this /. conversation will be long forgotten.

2) Personally, I think any analysis of the burn rate must consider that everything was burning; on this point I disagree with Bain. A hydrogen fire would light the skin, and a skin fire would, once large enough, ignite the hydrogen. I think it would be good to know more about the design to determine how easily a skin fire could ignite the hydrogen.

3) The linked papers basically say that there could not have been a spark due to the panels being interconnected. But if there were a spark, here is its power. Something had to set that thing on fire. An electrostatic spark is the prime suspect, but it could well have been something else.

4) I think the analysis of the potential power available for electric sparks to initiate the fire fails to take into account the complexity of the situation. It assumes that if there were a spark it would be initiated by the charge on one panel. This is used for energy calculations which conclude there was not enough energy to light the fabric. No other options are pursued. But, as noted, the panels were interconnected. If a spark occurred it would have been between a group of panels of which were somehow isolated. And it would take consideration of an arc caused by the ropes to the ground. Deterring details would require knowledge of the exact design of the ship and possible defects. How were the panels connected to the airframe? How were they maintained? How well did the connections between them hold up? Could something have caused the electrical isolation of a large section of the ship? Proper connection and grounding of systems to avoid sparks is very complicated, and the paper does not consider the details of the Hindenburg design.

5) Engineering Disasters also notes that the German investigation of the disaster concluded that the cover was electrically isolated from the structure (and that the Germans concluded it was a hydrogen leak).

6) The linked papers also comment that contamination from sea-salt particles would affect the conductivity. I will note that these would be mostly on the bottom of the ship - giving an even harder to predict electrical behavior to the skin.


Anyway, this is just a quick review. Further reading might provide more answers (and more questions).

Citation needed. And, while I agree with the XKCD take on mythbusters, they are not a suitable scientific or engineering source to cite.

To back up the claim that the skin of the Hindenburg initiated the fire I will site Engineering Disasters - Lessons to be Learned, by Don Lawson. ISBN 1-86058-459-4, pages 3 to 19. One quote from this section of the book:

"Addison Brian's tests
"In 1994 Dr Bain managed to get samples of the outer covering of the Hindenburg and carried out tests. The outer covering of cellulose acetate butyrate dope and fine aluminum powder was similar to the rocket fuel he was familiar with at NASA. He found that the aluminum and dope had bled through the fabric in places and had combined with the iron oxide on the inner surface. The resulting mixture is similar to a thermite fuse mixture used to achieve high temperatures in welding.

"Dr Bain tested samples of similarly doped fabric for their potential to be ignited by an electrostatic discharge. When an arc struck down onto the fabric samples, it only resulted in local damage. Airships struck by lightning had shown similar localized damage. When the arc was parallel to the surface of the fabric, the electrical energy was sufficient to ignite the sample, which was quickly consumed by fire."

The reference provided for the above quote was: Bain, A. and Schmidtchen, U. (2000) Afterglow of a Myth: Why and How the Hindenburg Burnt, DWV, January, www.dwvinfo.de

Other parts of this section go into further detail, including other hypothesizes for what caused the disaster. If you read Engineering Disasters, he does not claim that they mystery is completely solved. Indeed, It will probably never be solved with 100% certainty. But the theory of lightning igniting the flammable skin appears to be the most probable cause of the start of the fire. Obviously once the ship was on fire the hydrogen burnt, no doubt making the fire far worse. Who knows, maybe Helium would have put out the fire...

Do you have a source better than Mythbusters (and better than Wikipedia and Snopes)? If so, I would be interested....

Actually, there are budget models other than that of the U.S. government. Many of those models include saving money for anticipated large expenses, or for possible disasters. Some also include spending money where and when it is most needed - even if that requires saving it awhile. Instead of judging them on how quickly they spend donations, I suggest judging them on how effectively they use the money for their stated goals, and be sure to check a time period of at least a decade.

In my work experience I have found that a number of problems tend to crop up again and again:

1. Applicants for programming positions frequently lie about their programming skills
2. The average applicant for a programming position tends to do fairly well if given very explicit instructions, but are at a loss if something unexpected comes up. They tend to be less adaptable and nimble needed.
3. It ends up taking longer than estimated, even for simple projects.
4. Hand-holding and rework end up eating up all time and money savings.
5. By the time a programmer has been trained enough to actually be useful, they leave for a better position. (Especially true because they don't get raises corresponding with their increase in skill)

To my mind, managers need to be willing to provide good on the job training. They need to pay what someone is worth, with regular revaluations of what they are worth - and significant raises as appropriate.

Most problems can be found with U.S. or international programmers. Some problems specific to outsourcing:
-Their English is normally a lot worse than the average slashdotter. As I am working on learning a foreign language, I can really sympathise. And working on an international project I understand just how much trouble the language barrier causes.
-Spreading a project over a large physical area makes it even harder to communicate. The further away, the worse it gets. If you are in the U.S. and they are in India there will only be a few hours in which you can easily talk to them - if that. Most communication will take a minimum of a day.
-Despite the bad reputation of the U.S. educational system, the system in some other places is worse. This is reflected in the quality of the workers.
-Numerous cultural issues can cause even more misunderstanding. And, as unpolitically correct as it is to say, some cultures are better optimized for modern programming jobs than others.

Several pages is going beyond fair use. So is making a complete electronic copy when you do not own any copies (I am OK with complete electronic copies of books that they do own copies of - just not sharing them). And it is clear that copyright does not allow for someone to say that "I am copying all your works for commercial purposes. You have to opt out if you don't like this." Copyright even provides for punitive damages because this exactly how it is not supposed to work.

However, it does appear that the issue is not as clear cut as I originally thought.

So far as I have seen, Google is not pushing the issue of copyright duration. They don't appear to treat older works differently. They don't appear to be arguing about copyright duration in court. And they don't seem to have a massive lobbying campaign going on for shorter copyright duration. Or is all this happening and I missed it? - could be since I don't spend all my time following what Google is up to.

You make a strong point for why an author should want their works on Google. But I think a better analogy might be that the authors are like a starving dog biting the hand of someone who is tying to tie it up and shove steaks down its gullet. Google is not offering the authors a (possibly beneficial) service; it is doing everything it can to force it upon them.

The fact remains that Google is scanning large numbers of books and posting them freely on the internet. And, while it may not be their only motive, they are doing it for their own profit.

Should massive digital libraries be allowed? Yes. However, this should be take care of by two things: For slightly old works, reasonable limits on copyright duration - thus allowing anyone to create such an index. For new works, the copyright holder should be able to offer their works as they see fit. And the idea that they should have to opt out of Google copying their work is insane (even if there is some appeal to the idea of the RIAA/MPAA having to opt out of people copying their works).

What Google is doing is exactly the sort of thing copyright law is supposed to prevent. They are copying massive amounts of copyrighted works and using them for their own profit without permission from the copyright holders.

Sure, I love the idea of a computer search engine for books. And I can see how it could be very advantageous for an author to have their work on Google books. And I think that the term on copyright is far too long. Having said that, the courts should have put a stop to what Google is doing long ago. With something this clear cut, Google should have already been forced to pay damages. For that matter, they should have been the target of some of the government's anti-efforts.

No, ARCO should not be given the patent - nor should the researchers - because there should not be such a patent. Assuming the patent is simply on something they discovered then it should be invalid. Patents should be limited to a subset of human creations. Things you find in nature should not be patentable.

While your argument has merit when reviewing the 19th and 20th centuries, it is missing a key point. The entire goal of mechanizing and automating work is to eliminate the need for human labor.

The only reason to have a human do a job is because either we can't make a machine do it, or we don't want to. And in the long run, the people paying for the work will choose whichever is the most cost effective.

So I find the argument that there will always be more work to be done is questionable. Maybe there will and maybe there won't.

Personally, I thing ignorant people *should* be learning about and experimenting with radiation detection equipment. The lack of understanding causes a huge amount of trouble, and this is how people learn. And telling people that they are too uneducated to ever understand this stuff is exactly why the public refuses to trust us when we say that they don't need to worry about Fukishema unless they are next door.

What they need to be told at the same time they start to measure radiation is Don't Panic. The governments radiation monitoring is very very conservative. If the government panics you might consider being a tad concerned. Yea, a few people will still panic - but you won't stop them no matter what you tell them. And the rest will actually start learning about this stuff.


So, in answer to this ask slashdot:

1) Unless the the reactor explodes yet again (and 10x larger than last time), don't worry. Your child will be in far more danger from more mundane problems, like cars, falling out a window, pollution, and electricity - and you can do something about those problems. They are more likely to drown in the clogged gutter than get irradiated by it. But when they are a little older they should love playing in it.

2) There are not consumer radiation detectors like there are smoke alarms. Unless you work at a reactor, you are normally worried about small leaks with long term exposure. Professionally, those are normally measured by dosimeters. If you ever work at a facility like Los Alamos National Lab, you will have to wear one. You probably could get one to monitor your child, but the money would be better spend mitigating other, more serious, risks - or invest it today and it will help pay for their college...

3) Low cost hand held radiation detectors that give a live readout should be available from educational scientific supply stores. As others have noted, they are not all that accurate, and you really should not panic without a sold understanding of what is going on. But my high school class had fun when the teacher handed them out. Shortly thereafter the students found the uranium ore that he had stashed in a storage cabinet.

4) Understanding the risks with radiation does take a lot of research. You need to get a good idea of atomic chemistry. Just a few things you need to understand are the different types of radiation (electro-magnetic, alpha, beta, gamma, neutron, etc), what damage each type does, ionizing radiation, how it does biologic damage, and the ability of your body to repair damage. One thing to look into is how radiation danger is often evaluated based on the Linear No Threshold Model, and how that hypothesis is strongly disputed for low levels (in my non-expert view, disproved).

5) Sources: you can start by checking Wikipedia and google, but as you know they are not always accurate - and they are prone to getting overrun by alarmists. I don't know how much the Wikipeida editors actually know about radiation, vs just think they now.

One source I like is Enginering Disasters: Lessons to be Learned, by Don Lawson. He has several chapters on radiation, along with a good discussion of the Linear No Threshold Model. Some of his points are not as well supported as I would like, but still worth considering.

Another source to consider is entry level collage physics text books. Some of them should cover the basics of nuclear chemistry and radiation. For this topic, previous editions are very cheap and just as good (or better) as long as they were published within the past 30-40 years.

In short, you are saying that the HFT people make it easier for the prospective long-term buyers and sellers to come together to make a trade - both in time and location.

In the case of HFT with holding times under a day I disagree with there being any value added (and the ones everyone is upset about are down in the sub-second range). The system of people posting buy and sell prices allows time for a buyer to come along after someone decides to sell. As for bringing they buyer and seller together, that is the job of the broker. It is what they are paid to do, and they were quite able to do it back before HFT. (though computers make the brokers job far easier, and I don't blame brokers for using them)

I never said that computers have to be involved in HFT. That simply depends what frequency is considered high frequency and if human reaction time can meet your definition.

Market making: you could define HFT as market making with holding times less than X second(s). I claim that HFTs are not doing something useful because of the extremely short time they hold the stock. Just about any long term trader will be OK waiting a few seconds for a transaction to occur - and many of them would be happy waiting far longer. So they don't need HFT to provide for transactions in the sub-second range.

If you slow down HFT enough it could become valuable as market making (and it will no longer be high frequency). How slow that has to be depends on how patient the long term investors are when they decide to buy and sell.

Personally I think that stock purchases and sails should be based on a solid analysis of the merits of the company combined with the finical needs of the investor. The result being that stocks don't change price very quickly, and most people are happy waiting hours or even few days for a trade to take place. Thus, there should be little need for market making. Though there would be room for market makers operating with daily to weekly holding times.

Sadly, I suspect that trading speed will keep increasing well into the VHF and the UHF range.

It works something like this:
1) Someone invests money, holding the stock for many years.
2) They decide to sell it
3) N high frequency traders buy it from them, on average making a bit of money.
4) Eventually another long term investor buys it. On average, for what the first guy sold it for plus what the HFT people made.

Thus, the money they skimmed off should have either gone to the first long term investor or kept by the second. They are both doing something useful. The HFT guys are not.