NOAA: Earth Smashed A Record For Heat In May 2014, Effects To Worsen
Face it. You can't trust anything published on the internet or other media. Trying to dig to get the actual data is about hopeless. Most of the papers on global warming refer to other papers refer to other papers refer to papers by the same author based on a computer model. Scant attention is paid to validating the models against history. Actual data collected from the field gets locked behind for-pay firewalls. I've been tracking the CO2 issue since the 1970s when I was told we needed to solve the CO2 problem by 1980 or else its all over. I've witnessed reports on the internet reporting volcanic CO2 production revised to lower the production estimate by an order of magnitude without explanation and without acknowledgement it was altered. Much of the data you see on the Web is a lie on both sides of the issue.
We all have computers or else we wouldn't be reading /. Lobby your government representatives. If a scientist accepts money from the government they must digitally sign their data and make it publicly available. If they develop computer models, they must release it open source so they rest of us and review and run it.
Global warming is probably happening, but probably not for the reasons we suspect. One eruption in Africa released more CO2 than the entire annual industrial output of the United States. In the 1980s most CO2 came from "land use change" -- the burning of rain forests. The pollution of the ocean was still unevaluated so we have no idea how much CO2 is no longer absorbed into the ocean. We're stilling coming out of the last Ice Age so Earth is going to get warmer. Perhaps it is stupid to dump more CO2 when we're already getting warmer, but is that our only global warming problem? The science has become so politically obscured that I'm afraid we're missing other real dangers.
Cobalt-60, and Lessons From a Mexican Theft
A 1 minute dose close to 3000 Ci of Co-60 is intense enough to be 50% lethal within 2 weeks with medical care. The article mentioned a 30 second dose being 50% fatal -- its biology so it is not exact science. A five minute exposure is 100% lethal within 2-days. During that 5 minute exposure the victim would start vomiting, suffer seizures and spasms, and start losing mental faculties. This isn't a cancer in 10 year type of terror. This is almost immediate and monstrous.
I would not characterize the loss of Co-60 as Media FUD. There is some stuff on the planet that truly is scary.
A War Over Solar Power Is Raging Within the GOP
I love it when liberals say "Republicans think this or do that". Demonstrates a total lack of understanding of groups and susceptibility to propaganda. Republicans have their corporate welfare types, libertarian types, religious wackos, and atheists, and others. The subsets of the various groups makes for an intricate Venn diagram. When one of those yahoos speaks, they're not likely speaking for me.
How much a utility pays for co-generated power isn't a left versus right issue. You have a government regulated monopoly, the utility, which owns both generation and transmission facilities, arbitrarily setting the price for power from co-generators. Perhaps the transmission lines need to be a government regulated monopoly with every generator and co-generator paying for access to the transmission lines. This really doesn't appear to be a solar power issue but one about who controls the life blood of our society, electricity.
Support For NASA Spending Depends On Perception of Size of Space Agency Budget
Remember NASA was originally created through the cannibalization of military programs such as the X-15. Also recall that NASA actively discouraged the development of commercial space launchers to drum up business for its space shuttle. It purposely delayed permits for the commercial sales of Atlas and Titan rockets. I personally worked on reusable launchers and cheap satellites only to be told on multiple occasions to cease work or my company would never receive another NASA contract. If you're a space company, NASA was the only game in town. TRW created a million pound thrust engine for $40K (it had limited re-use -- they fired it 3 times on the San Juan Capistrano test stand). It went nowhere because of NASA.
Now also note that NASA's management and culture is incompatible with space exploration. We lost two space shuttles through wishful thinking and shortcuts. In this case it is hard to distinguish treason from incompetence.
Given NASA's history of actually impeding the development of space, their demonstrated incompetence, and our national debt, we're past due selling NASA off. We need to beef up other means of funding space research. Think of the Ansari X prize. NASA exists to perpetuate itself without regard to benefits to the nation.
Richard Stallman Speaks About Back Doors After NSA Documents Leak
Stallman is being silly again. The cloud is like a giant parking lot for data. If you leave your data unlocked, of course someone will steal it. Encrypt everything, including the unimportant, so you don't need to remember what's encrypted and what's not. Route your data through anonymizing servers, even your Amazon transactions.
Everyone needs to use encryption, not just the political activists and child pornographers. You have bank statements and health records to protect. No one needs to know what you buy but you. If the world knows what you buy and need, like your collection of garden gnomes, the price of garden gnomes will go up.
Just because you encrypt everything does't mean you can't share. Encryption keys can be split and shared. Some people are working on anonymous group keys so you can subscribe to the online New York Times without getting spam from 3rd parties. I know one fellow who encrypts his Facebook postings with his private key. When I want to read what he has to say I decrypt with his public key. He can use a shared key if he wants to communicate with a particular group of friends.
As for the threat of quantum computers -- its true that quantum computers can factor billions of times faster than conventional computers, but all that means that the average time to break my key has gone from billions of times the age of the universe to merely the age of the universe. I'm not worried yet. I note the U.S. military recommends 1024 bit keys for secret data and 2048 bits for top secret.
Stallman got one part right. You can't trust commercial operating systems. You need to be able to see what you're running to trust it. We don't have a secure cloud because commercial interests aren't concerned with your privacy. They want to track you so they can sell to you. They have not realized that if they are tracking us, they are being tracked. We don't have a secure internet because big companies don't want it. You need to do it yourself.
Encryption is the infrastructure for this century. Use it and insist that others use it.
Ask Slashdot: How Many of You Actually Use Math?
I actually don't use a lot of geometry in my job, but I use the basic skills of logic and proofs that I learned in geometry. I don't often solve algebraic equations but I find myself tutoring my employees in the logic of intervals. Calculus was a drudge and I almost never integrate a function, but the concepts I learned about functions I use every day. I'm not sure you can use functional programming without exposure to calculus. Almost every math subject has an ulterior motive. Geometry is there to teach you logic. Pre-calculus is to teach you about functions. Calculus is the first math where you're introduced to infinity and infinitesimal entities. Without that math I'm not sure your mind has expanded enough to work for my company.
I don't expect programmers to solve differential equations in front of me, but I doubt their technical literacy if haven't done it at least once. You don't have the tools to manage complex projects without knowing about diff Q. Likewise, you can't be a computer scientist without discrete math -- you just wouldn't have the mental tools to address the subject. Sure, you could stumble along, but you would just be depending on someone else who did the math.
Do Slashdotters Encrypt Their Email?
but I used to work for PGP...
Can Relativity Explain Faster Than Light Particles?
I'm a mathematician, not a physicist. In special relativity the Lorentz transformation has a singularity at the speed of light. Its perfectly defined below and faster than the speed of light. Richard Feynman suggested that unless the math prohibits it, it will be found in nature. Of course, if we see a particle hitting another before it was emitted, we'd likely interpret it as the target "pulled" the particle from the emitter. This explains the alternate view of physics -- all matter emits dark, and light bulbs and stars suck the dark in. Mathematically it makes just as much sense as emitting massless photons. Who's going to buy the idea of a massless particle that goes the speed of light? Its crazy talk.
This is just a reminder that like economics, scientific method is really a confidence game. Its all a matter of whether you believe the rules stated so far are consistent.
It's Time To Split Up NSA Between Spooks and Geeks
Where else is Google going to find someone that understands how to configure SELinux so it can be used in real life.
Oh, doh! I referenced real life in the same sentence as SELinux and Google in a Slashdot posting on the NSA.
Ethics of Releasing Non-Malicious Linux Malware?
We already know how to break into systems with buffer and heap overflows. We know how to do SQL injection into not-so-smart applications. If you work at it you can break into almost anything.
Absolutely no good purpose is served providing a toolkit that allows people to break into naively configured systems. Much of what you describe is akin to leaving the keys in your Maserati with the doors unlocked and the engine running. Please don't make things easier for joyriding teenagers.
If a site wants to know if they're secure, within the current limits of our knowledge, they can perform their own audits, and hire their own advisers to test their systems in a controlled fashion.
Applications, such as BOINC, have an unknown state of security review or audit. I doubt they applied the coding guidelines of CERT, or any of the Common Criteria levels. An administrator would only deploy such applications in the DMZ of their network. To call a Linux system, or Windoze system, secure means you've evaluated the risk of both the operating system and the applications on that system and decided it is good enough for you.
Multiple Home Pages with Google Chrome
Set up your tabs the way you like to start.
Go to the options menu on the basics tab, and under "On startup" click "Open the following pages" and click on "Use Current". Click on "Close". You're done.
Encryption? What Encryption?
If the laptop is at home or away from work it can attach to anyone's network, and the laptop is likely to be running an intrinsically unsafe operating system with its unsafe browser so in short order it will have a bot broadcasting everything.
Encryption? What Encryption?
Encryption is already becoming a de facto standard. Financial companies and banks used to get big fines if some bozo took home confidential data on their laptop, then let the laptop get stolen. Governments have decreed that if the laptop was encrypted, "no harm, no foul".
Now you can rest assured that your data on the laptop is now secure while the employee is blasting it over the internet.
Are Information Technology's Glory Days Over?
I still don't have a computer that can hold a conversation, effectively vacuum the floor, or clean the windows, or see when I'm running low on coffee. I still have hundreds of a little chores that require little intellect that a machine could do if they were just a little bit more intelligent. A typical desktop now has the power to simulate a nuclear explosion, just too bad I have no need for nuclear explosions. I just want something that do simple tasks like shut off the solar powered battery charger on a hot day when the batteries are getting too hot, or automatically know when to vent the hot clothes dryer exhaust inside or outside according to temperature and humidity.
Tom Siebel is a business guy, not an engineer. Maybe the big enterprise things are all done, but the millions of little things that need to get done will yield lots of people billions of dollars. Richard Feynman was right when he told us there was infinite room at the bottom.
What Questions Should a Prospective Employee Ask?
You're already on the right track asking about their version control and QA, so generalize the questions a little and find out how they feel about quality. Most companies just want to ship product, that is, "quality may be king, but schedule is god". Run away from those companies because they won't survive long enough to make your stock options worth anything, and you'll spend your time cleaning up other people's messes.
Ask them if they do agile-like things such as do they expect you to write unit tests. Ask if the developers write functional tests for each other (not pair programming -- I don't want you crowding me in my cubicle unless you're cute).
Ask how much time they spend on bug fixing. General rule of thumb is that if they're spending more than 50% of the time on bug fixing, the organization is going to crater within two years. Ask if they have a separate support organization that does bug fixing, which is a company's favorite way of hiding the true cost of their bugs.
Windows 7 RTM Reviewed & Benchmarked
User that know what they are doing can perform more tasks per unit time via a command line than poor old GUI users wandering through menus and dialogs. Objective studies have repeatably shown this. Just compare the time it takes you to copy a file via the old Windows COPY command versus selecting the file in Windows Explorer, right clicking to copy, then paste, then rename the copy. You don't have to be an expert to appreciate the command line.
Early Abort of Ares I Rocket Would Kill Crew
In 1980 I was assigned the task of analyzing the behavior of the IUS active cradle in the space shuttle during aborts. At the time the RTLS mode required separation from the main tank and early separation from the SRBs. Note that the IUS cradle functioned as designed during the Challenger disaster.
I'm perfectly aware of the current definitions of the abort modes:
Even a cursory engineering analysis shows those abort modes to be bogus. Given the SRBs have burned out around 150,000 feet with the shuttle only going 3000 MPH at that time, and presumably it is aborting because of a problem with the main engines, just where are you going to get the ballistic energy to get to Africa or Spain?
The point is that everything else must be going right if you must abort during launch, otherwise you will loose the craft and most likely the crew.
As for cross-range capability, you have cited no numbers. You failed to note that landing sites such as Zaragoza (ZAZ) or Morocco are available only for high inclination launches, i.e., a certain range of launch azimuths. The ISS at 51 degrees orbit is not a high inclination orbit. Gliding is not an option with the shuttle. At hypersonic speeds the glide ration of the shuttle is 1:1, and increases to 2:1 at supersonic, and finally 4.5:1 at subsonic speeds. 3000 MPH gives you a ballistic range of 50 miles. That leaves a long way to Spain or Gambia.
If you had actually had a clue yourself you would have actually read my post and realized that I wasn't suggesting Mexico City was a landing site. In fact the point is the shuttle can only land at specially prepared sites capable of handling a hot vehicle spewing hazardous materials.
As for the F-111 analogy, again you missed the point. So what if the F-111 only seated 2? Think about an alternate shuttle design in which the entire crew cabin separates, which in fact happened in the Challenger disaster, but had no means to slow down or recover. You've made unwarranted assumptions about weight, and in fact have accepted NASA's assertions without question. For some reason, NASA was able to design the Orion rocket with the ability to lift the entire cabin free and recover the cabin safely, but was unable to do it with the shuttle. The point is that the shuttle is fundamentally design flawed.
It doesn't matter how slow the shuttle the travelling or at what altitude the shuttle is at in respect to the CES. The astronauts are still behind a tractor rocket and still exit one at a time.
I don't appreciate words such as "bullshit" in posts, especially when the post'er has not done their homework.
Early Abort of Ares I Rocket Would Kill Crew
You've been drinking the NASA Kool-aid? NASA habitually has its astronauts train in impossible scenarios for PR purposes.
RTLS, return to launch site, requires separation from the main tank and SRBs. Despite NASA's design changes most of us old rocket scientists think that will detonate the tank. Separating from the SRBs requires a hard pitch down to avoid the SRB plumes, which in real life the shuttle does not have enough aerodynamic command to achieve. Simulation or not, RTLS is bogus and NASA knows it.
Also, the shuttle does not have enough cross-range capability to achieve a Transatlantic landing at any facility capable of handling it. An attempt at TAL will result in a water landing with loss of vehicle and crew. TAL is bogus and NASA knows it.
Abort Once Around suffers from Earth rotation and the lack of sufficient cross-range capability. All viable landing sites are just not available on a once-around. It would be interesting to see what would happen if an astronaut pilot did manage to land a fully laden shuttle at Mexico City, hot from re-entry and leaking hydrazine.
Abort to orbit really isn't an abort mode.
Now, about the Crew Escape System (CES), which are "tractor" rockets that pull the crewmen out one at a time. The shuttle reaches 25,000 feet in under a minute. At 35,000 feet its breaking the speed of sound. It seems the only way the CES will work is if you shut the main engines down. Now, ride the solids until you can separate, and then glide down until the crew can escape. Maybe that is how the TAL and AOA are supposed to work, but I still don't like the idea of being dragged behind a hot CES rocket plume wearing nothing but a space suit after waiting in line for the privilege.
Given our jet fighters now have ejection seats that can rescue a pilot flying upside down near the ground, CES seems like a real Rube Goldberg device. Back in the the 1960's, the F-111's entire crew compartment ejected.
Given our current state of technology, there is no need for intrinsically unsafe designs such as the shuttle or Aries rocket. I think I'll wait until I can buy a ticket on a rocket plane for my ride to orbit.
Early Abort of Ares I Rocket Would Kill Crew
It doesn't matter how the Challenger actually failed. The shuttle has no survivable launch abort scenarios.
Challenger carried an Inertial Upper Stage booster in an active cradle. NASA provided several launch abort scenarios to contractors and required the contractors to analyze the active cradle's behavior in the abort scenarios to demonstrate that the active cradle would not make a bad situation worse. The Rogers Commission found all of the NASA abort scenarios, such as early separation of the boosters, or an abort to Spain, or once-around abort, would not actually work, and would result in the loss of the vehicle.
Whether you think NASA conspired to cover-up the lack of survivable abort modes, or they just incompetently succumbed to wishful thinking, they will have no creditability until they actually do an expensive full-up test of a launch abort to demonstrate the safety system work. If the escape systems don't work, then NASA needs to re-design the rocket until it does.
Are Code Reviews Worth It?
Over dozens of projects I've seen 2/3 of code reviews fail to yield tangible benefit, but the other 1/3 of the code reviews were wildly successful in finding defects. Primary difference between the successful and failed reviews was homework. In the effective reviews multiple reviewers spent about an hour before meeting to review the code in respect to a checklist of items to watch for. Those checklists did not address formatting issues that can be corrected with a code beautifier. At the review meeting each person presented their found defects, which led others to find other defects in a synergistic effect. The checklist itself was subject to review. Because everyone, including the programmer whose code is under review are trained in the technique, there wasn't the usual defensiveness or the feeling everyone is ganging up on one person. The expectation is that defects will be found.
The reviews that failed, failed for lack of preparation, lack of formal standards expressed in a easily digestible checklist, and a lack of training so the sessions degraded into carping sessions about one person not liking another's variable names.
A multiple person code review is an expensive proposition. I'd use it only on critical code and on public facing documentation.