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

 



Forgot your password?
typodupeerror
×

Comment Re:More Republican corporate welfare (Score 1) 248

sure, and we didnt have the tech to send us to the moon in the 60s...but we did it

So what are you trying to say here? He3 is only useful in a fusion reactor and we don't have a working design. People have been working on one ever since they invented the H-bomb and come up short, we have enough He3 here on earth to experiment/test with. Maybe we should see if we're able to do something useful with it before we spend billions trying to build a moon mining operation?

I completely agree with you and it is sad to see this tired old argument every time there is a moon story. There are plenty of good reasons to go to the moon, He3 isn't one of them. There is no reason to even bring up the subject given the numerous other reasons to go to the moon.

Comment Re:Holy Jebus (Score 1) 220

It is one thing to use such techniques to test something in a quiet area like a testing chamber, and something completely different to use this technique not only in a rocket that is accelerating at 32 m/s^2 but also has a whole bunch of other noise going on from nine operating turbo pumps, the rocket engines themselves, and other things rattling around tied to that whole system. In addition, to be able to locate a cause while the whole rocket is undergoing massive unplanned disassembly (aka an explosion releasing as much energy as a small nuclear bomb). Also doing that remotely off of recordings made on less than ideal microphones sharing bandwidth with many other more critical data functions over 30 miles away from where you are at.

As something SpaceX can and perhaps should be doing as a part of their Q/A analysis, no doubt using some sort of sound probing to detect faults is going to be done. I just don't know if this has been done on something like a black box of an airliner to perform fault analysis post-mortem on a vehicle failure.

Submission + - SpaceX Falcon 9 CRS-7 Failure: Broken Strut on Helium Tank (parabolicarc.com)

Teancum writes: In a press conference held by SpaceX CEO Elon Musk, he released some findings about what caused the explosion of the launch vehicle carrying the CRS-7 Dragon spacecraft on its way to the International Space Station.

Apparently a Helium tank was held in place inside of the Liquid Oxygen tank on the 2nd stage that failed while it was under going the nearly 32 m/s^2 acceleration (about 3.2 time Earth's gravity acceleration). This part was manufactured by an undisclosed 3rd party contractor for SpaceX and was rated to being able to hold up to 10,000 PSI, but failed at 2000. In the past week, SpaceX has been "testing an enormous number" of this exact strut that is currently in their inventory intended for future flights, and confirmed that at least one of them failed in a similar fashion where metallurgical analysis has been performed trying to identify potential defects.

It was also confirmed in this press conference that the SpaceX Dragon spacecraft was tracked after the explosion and remained intact until it could no longer be followed by the tracking station due to it falling below the horizon. The Dragon could have survived the explosion and been recovered, except that the on-board guidance computer was not programmed to deploy the parachute during ascent. In the future, SpaceX plans on having this parachute deployment as a standard procedure even on cargo missions in the event of a rocket failure.

Comment Re:First thing I thought of (Score 5, Interesting) 446

The first thing that came to mind when I heard of this site is "This is a prime target for a hacking/blackmail scheme." The only surprise here is that it didn't happen sooner.

As someone who has data in there (out of curiosity), it couldn't have happened to better people. The people that run AshleyMadison are worse than the lowest spammers. Not because they sanction marital cheating, but because they are exceedingly scammy in every aspect of the way they operate their business. They make Paypal and Stamps.com look like saints.

Comment Re:What could possible go wrong? (Score 5, Informative) 120

I'd like you to point out any launch site for orbital rockets that is anywhere even remotely close to tall buildings or even aircraft flights? The FAA routinely makes a pretty large exclusion zone around any launch activity. With the recent launch disasters from SpaceX and Orbital-ATK, I think such warnings should be well heeded even for ordinary Kerosene fueled rockets, much less something with an exotic propulsion system like this. It sure isn't going to be launched out of Central Park or any other urban center.

Besides, the CEO addressed this specific issue in an interview recorded a few months ago. Not only is the launch going to be far from cities, it will also need to happen in an arid region in part due to the microwave power being absorbed by water in the atmosphere. In other words, it is likely that this won't be launched from KSC in Florida either.

Comment Re:Hard to believe (Score 2) 116

I find that hard to believe. I have had 4 legal experiences in my time.

1) a divorce - (family law) 2) a labor dispute over a layoff - (labor law) 3) a private investment - (securities law) 4) A copyright filing - (intellectual property law)

In every case, there were some areas that could have been algorithmic, but in many dimensions on each one there were things that came about from advice from the attorney on how to position myself and under what laws I could make a case, which has a lot to do with language parsing and the definitions of the words used and their context. Unless this was paired with something like Watson which can determine meaning from context, I don't see this as being anything more than a paralegal replacement, but not a lawyer replacement.

Yes, but the vast majority of cases are fairly straightforward. Laws are nothing but a set of rules, and computers are great tools to track rules and figure out which apply. Precedents are set which further define what happens when the law falls short. Law (at least US Law) is chock full of "tests" which are fairly easy to apply. They come in the form of "If this AND this AND this, then $ruling". Unless you are in a precedent-setting case, which is rare, then I absolutely believe that a computer can be fed the results of a bunch of yes/no questions, asset values, and come up with the right answer with very high accuracy. If the two parties can agree on the answers to the yes/no questions and the asset valuations, want to reduce costs, and are not at each other's throats, then why not use a computer?

A computer doesn't have an interest in wasting time and accumulating billable hours like a lawyer does. No matter how much honesty and integrity the lawyer has, getting paid is always going to be on their mind.

Comment Re:as always.... (Score 0) 204

How would insurance save money? Another middle man to pay. The only justification for insurance is when you need to smooth out the bumps in your spending - an individual may not have $30,000 sitting around to replace their crashed car. NASA can almost always slip a schedule; self insurance makes a lot of sense for them.

f SpaceX had to pay the premiums for each launch separately, but NASA had a contract that didn't allow cost increases due to insurance premium increases, then insurance would be a great idea.

As it stands, NASA shoulders the costs for SpaceX's mistakes. The only reason that situation is allowed to stand is because it is common in government work. It shouldn't be.

Comment Re:Statism vs. Libertarianism again (Score 3, Interesting) 123

There's a world of difference between an adobe flash exploit and the availability of a gun that can mow down a large number of people in a matter of seconds.

There is not. Shutting down NYSE [slashdot.org], for example, cost billions of dollars. At $10 mln per life [wikipedia.org], that's hundreds of lives right there...

Are you making a serious argument in comparing people getting shot and the NYSE shutdown? This is the hill that you're going to make your stand on?

It's a very poor example but a valid point. A much better example would be fraud [identity theft], ransomware, spam, etc. With computers you can easily steal time from people on an unimaginable scale.

Suppose someone hacks me, and I get off relatively "easy". I may spend 1 hour of my time canceling a credit card, activating the new card when it comes, and changing all the passwords of all the accounts that the credit card number is associated with. That's probably on the very low end of what a hack can cost an individual.

The hacker doesn't stop there. They repeat their act 1,000,000 times. That's a fairly successful and prolific hacker, but not unheard of, espeicially if the attack vector is a business. At just an hour apiece per victim, 1 million victims is 114 total man-years spent cleaning up. Nobody died, but an entire lifetime has been stolen.

The Target hack(s) affected "up to 110 million people". If we take that figure at face value, and each victim spent only an hour dealing with it, that's 12,557 years or roughly 148 lifetimes. Even if I count injured people, I can't find a mass shooting that comes anywhere near 148 lifetimes.

Comment Re:Algorithm (Score 3, Interesting) 233

The ads that Google shows you are based on your search terms most of the time.

Except when it's not. Which in this case clearly indicates there's a profile that's made up of more than just search terms.

The search terms were identical for all profiles, male or female. The authors of the paper admit in the abstract that they don't know who is responsible for the different results, but since the only difference was the "gender" setting it is clear that at some point in the chain (Google, advertisers, recruitment companies) there is a rule that says "favour males", just like there is a rule that says "favour females" for tampon adverts.

Right, confirming that it's not just search terms. So we agree, there's a profile involved, not just search terms.

The difference between those two examples, and why one is a problem, is hopefully obvious.

It's really not obvious. Are you suggesting that advertisers shouldn't be allowed to target ads? Are you suggesting freedom to engage in advertising should be modified by rules? You're implying that. On what basis do you justify telling corporations how to spend their ad money?

Google generally shows ads that they think you want to see. They learn from feedback- which links you click and which you scroll by immediately. They aggregate that data, then slice it and dice it into different personas (or profiles). I am sure they have categories which all people fall into 2 broad categories, and they have a separate profile for every user. All their data mining and AI research result in a weird reflection of humanity. If that results in women not seeing certain ads, I can only conclude that that is because women generally don't want to see them, or prefer to see other types of ads instead. Perhaps the majority of women prefer to see ads for jobs with more schedule flexibility. That would be a reasonable conclusion since only women can carry fetuses to term, and doing so requires some amount of schedule flexibility. More than 50% of women have children, and determining who does and does not want children is probably not easy- even people with very strong opinions on the matter (like myself 10 years ago) do change their mind suddenly, for a variety of reasons which may defy profiling.

Comment Re: Fear (Score 2) 364

The DOW Industrials are at a P/E of 16.2, historical averages since the 1880's is 16.6, there's no huge bubble or crash coming unless it's an international contagion from Greece or China that halts world economic progress.

The DOW is not a good benchmark for investigating if there is a bubble or not. It is comprised of just 30 companies, mostly huge conglomerates and industry titans. There are also little to no "new" businesses on the list. The tech companies on the list are very mature, and include Apple, Intel, IBM, and Microsoft. As far as companies that are trading at a "fair" price, I would say the 30 companies in the DJIA are priced very fairly because of all the eyes on them.

If there is a bubble, it is almost surely not reflected in the DJIA. Let's say, for argument, that there is a bubble (you don't have to agree, just for the sake of argument). Where would it manifest? Technology companies founded in the last 10 years? Tech companies founded in the last 3 years? Those are the likely candidates in my opinion, but they are not represented in the DOW 30 AT ALL.

Maybe you think the next bubble will again be in banking? In that case, only Goldman Sacks, Visa, and Chase represent the banking industry on the DOW 30. I have the opinion that if there is a serious disaster brewing, those 3 companies can keep it from affecting their balance sheet until the last possible moment. They were fairly successful in doing so in 2009 so I have no doubt they could and would do it again.

My point is that if you are looking for a bubble, looking at the DJIA is a complete waste of time. P/E ratios may not be the best indicator either- in 2008/2009 the P/E ratios didn't make alarming moves until *after* everybody knew there was a big problem. You're relying on every other investor to tell you that things are OK (by assigning a fair Price), but every other investor may well be stupid. The best indication I think is what companies are paying when they buy other companies. Are they paying reasonable prices that will allow them to earn a profit on their investment? If they are buying a company to protect market share, is their investment at least as much as the potential losses if they hadn't bought the company? If the answer to either question is no, that's a big problem. It means they have so much money that they don't know how to manage it, or it means that they are basing their decisions on emotion, and not numbers. Either of those is a recipe for disaster.

Slashdot Top Deals

Say "twenty-three-skiddoo" to logout.

Working...