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


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Submission + - DIY physics: A spark chamber on a steel tower, for no reason. (douglasruuska.com) 6

Gigabit Switchman writes: Slightly deranged hoopy frood (and artist) Douglas Ruuska has convinced a bunch of folks to help him build a functioning spark chamber (which is pretty cool, high voltage FTW), and then to put it on top of a steel tower covered with a few kW of addressable LEDs. Because that totally makes sense. The particle detectors will output data to a BeagleBone that controls the LEDs, and the design is being open-sourced once it's functional. Details at the story URL. (They're also running a Kickstarter, but I'm not linking that.)
Open Source

Submission + - Affordable, High-End, Audio-Responsive, Hackable LED Light (saikoled.com)

Gibbs-Duhem writes: SaikoLED is a little company in the US that after 5 years of engineering, feedback, and installation experience is launching a pretty awesome open source and open hardware audio-responsive RGB+W Arduino-based LED light for a starting price of $79. The light is being launched using a new crowdfunding platform, also launched today, called Crowd Supply where you can donate now if you like.

There is a wide variety of really neat stuff in the technical blog which is also being used as a repository for general Arduino and LED Lighting know-how. Some examples which generally include both media and code are autonomous audio responsive mode, a deriviation for how to convert from HSI colorspace to RGB+W optimally, how to get 4 channels of 16-bit PWM on any ATmega32u4 based device (including Arduino Leonardo), a color changing surface that can be used as either a beautiful table for artwork or as a neuroscience tool to study the function of flicker phosphenes in generating geometric hallucinations, how to get arbitrary color correction functions using HSI colorspace, cool shades that make neat patterns on walls, and a start to extensive documentation about how the device works from a low level to a high level. On top of it all, they plan to run events like the myki Challenge and donate a portion of their lights to schools and hackerspaces if they get funded.

Submission + - How to enable 16-bit PWM output on Arduino Leonardo (saikoled.com)

Gibbs-Duhem writes: "This tutorial has extensively commented code explaining how to configure the registers on the Arduino Leonardo to do 16-bit and 10-bit PWM outputs in hardware. It's directly written for the SaikoLED MyKi LED Light, but it's based on the Arduino Leonardo, so the same code will work for both. With 16-bit PWM output you have a lot of cool new options like arbitrary scaling functions to map an 8-bit value onto a different function, or to allow much dimmer LED light outputs. An example video showing red and blue LEDs side by side with 8-bit versus 16-bit PWM is shown here."

Comment Re:Enroll in Martial Arts (Score 1) 372

Totally agree with this. Martial arts are perfect for nerds.

I can't stand things like running because even though I know that in principle there are lots of details to pay attention to, I want something where both my mind and body are engaged.

Now I teach karate at MIT, and it's awesome teaching to people who are also nerds. My favorite is describing that a punch works like a torsional wave where an impulse is put into your spinal column at your hip level and then it moves up your body to your shoulders and then out your arms. Explaining karate techniques in terms of energy and momentum is incredibly helpful when you know those concepts. Learning kata and bunkai is also very interesting, particularly when you get to the point that you're making up new bunkai for existing kata. This is very interesting creative, intellectual, and physical work that nicely integrates all aspects of your fitness.

I also love that martial arts are the one sport where you continue to get better and better as you age. If I'm sparring with some 60 year old, odds are I will lose and lose badly. Sometimes hilariously badly. I like the idea of a sport where in 30 years I will actually be better than I am in my "prime" for most sports.

Comment Re:Babylon 5 (Score 5, Interesting) 409

This is basically the best reason to read the Honor Harrington series of novels. It blows every other science fiction writer away in terms of portraying reasonable space combat.

1. Always wear a space suit in combat. Duh.
2. You don't know where your enemy is until c*\Delta x has passed. This is both advantageous and disadvantageous.
3. Surprise! You can only decellerate as fast as you can accelerate! What? You mean I have to spend half of my time rushing at my opponent slowing down?
4. Laser beams hit at the moment you know they've been fired (not that they're used much, lasers are weak).
5. Lots of people die all the time. I think they killed billions of soldiers in a major war.
6. Yes, even your friends and main characters. Stray missiles suck.

It's fantastic.

Comment Re:What, specifically, did you do? (Score 1) 474

I don't know if this is the best solution for everyone, but I just have my computer generate pink noise. It's got roughly the quality of a rain storm, and after a few minutes I don't even notice it anymore. I can use it to sleep through anything, and I live with roomates who are, shall we say, active at night.

Comment Re:I'm still blown away (Score 4, Interesting) 536

This seems from the reports I've read to be pretty spot on. I would add an addendum to an earlier comment about this being why no nuclear plants will ever be built in the US again though; the current designs are generally "passive fail", meaning that unless electricity is supplied to the control systems, the plant will just... stop being just sub-critical and will go non-critical very quickly. For instance the pebble bed designs. My (somewhat, I'm probably giving myself a little too little credit) understanding is that these plants use nuclear fuel that just... can't react on it's own due to the sheathing materials. I think those are pyrolytic carbon still though, so of course there will still be problems with burning if they are exposed to air, the accompanying release of hydrogen, etc (I think).

This seems very honestly to be the entire focus of the nuclear industry -- designing plants which are safe to operate no matter what, which maintain reasonable cost-effectiveness. It's basically the holy grail.

I think the current problem is:
1. Natural gas is cheap, coal is cheap, they are cheaper to build and easier to maintain.
2. The regulatory process and validation work to get a new plant design is intimidating. Probably even intimidating as compared to the design of fighter jets.
3. Nuclear *is* scary to the vast majority of people. This is residual in large part from Long Island, and based in concerns over running reactors commissioned in the 60s still being operated. *That* part I am scared of. But as a scientist and engineer, I think that these are solvable problems so long as safety and the concepts of "fail safe" systems engineering based on the Therac-25 (http://en.wikipedia.org/wiki/Therac-25) which seem to have very permanently changed the way that people fundamentally think about how to do system engineering. These problems had not arisen and become understood when those plants went into operation. A current plant definitely would do a far better job of that.

Heck, it even effects me on a daily basis (at this point in my career I would classify myself as a systems engineer); I think all the time "What happens if all this equipment just stops working" and the answer is always "go to a safe operational mode". The are different ways to do that. You have the F-16 style of doing that, which includes crazy amounts of unstable control algorithms. But by *far* the preferred mechanism is physical. For instance, if I have a furnace I expect to go to 2000C, and monitor the temperature with one thermocouple while I use a single additional thermocouple as a safety, is not really enough. I would *far* rather have a thermal fuse that blows hard when a temperature exceeds some set ultimate super failure limit and shuts everything off immediately. I don't trust thermocouples to be reliable, and I don't trust the controls equipment to respond properly in an emergency.

But in one of these pebble beds, the sorts of controls they are integrating are way beyond "having power", by far the best safety integration is to design it such that electricity failing causes large physical things to happen. Dumping the pebble bed entirely, or dumping immediately a mediator into the reactor that is only prevented from triggering by constant electricity. Some of the designs I've seen literally place the reactor under a ridiculously large tank of water held closed by electricity. I don't know in what way that would fail, but it would be far superior to what happened in fukoshima.

Comment Re:Hard truth (Score 1) 217

This is true about patents in some areas, but not others.

I say this as someone who has easily worked around patents before, and nonetheless has heard repeatedly from investors that *some* IP is critical for self-defense and protection of core technologies.

Comment Re:Hard truth (Score 1) 217

Yes, absolutely. I'm a startup founder (not a web startup, which actually seems to make it much harder to find VC's nowadays, by the by).

I've got tons of impressive credentials, but if my startup idea sucks, or I've got serious issues with my team ability to implement it, of course I would want to know! I can't fix a problem I am too inexperienced to understand. If the solution is a search for a new co-founder who has more experience, then heck yes I want to know that so that I can do it! I want my startup to succeed, not flounder forever.

Even more, if it's got serious fatal flaws then it's worth me seriously thinking about whether or not I should be doing it. I can make an extremely nice salary at a regular job, even a job at another startup to learn really useful stuff at in terms of doing my own. Why would I want to waste tons of my time and spirit on a project that makes no sense, or won't work for team reasons?

I have gotten far more negative feedback than positive feedback from potential funders. I have never taken offense; the criticism isn't about me as a person, it is about me in the context of my startup. If the feedback I got was about how my religion or gender or sexual orientation or race was wrong or something ridiculous like that it would be one thing (although I am easy going enough that frankly I probably still wouldn't take offense). But if it's practical concerns about my skills with leadership, or my team's lack of expertise, or whatever that's incredibly valuable to know! Anyone that gets offended by free advice is doing it wrong. Yeah, the VC may be wrong, but you get what you pay for.


Submission + - Sound Responsive High Power LED Array (saikoled.com)

Gibbs-Duhem writes: "This video shows operation of a neat project using some new digital analysis chips to produce hacker-friendly sound-responsive LED modules similar to old color organs, but much smaller, easier to set up, and modular. As usual, the boards are open-hardware, and both eagle schematics and board layouts are available on the website. The artist is constructing a 10x10 array using a Kickstarter Project, which will put out 25,000 lumens of light for a display at Burning Man. This work is from the same artist who previously built wifi LED lighting for some pretty cool installations around the country and lights with ultra-high CRI and added ultraviolet LEDs for use with artwork display."

Submission + - Autonomous Sound Responsive High Brightness LED Modules (saikoled.com)

Gibbs-Duhem writes: "This is a cool little art project on Kickstarter where a bunch of individual high power LED lights are souped up with built in audio analysis hardware and an onboard microphone. At scale, these modules put out over 250 lumens a piece, and can be assembled into enormous arrays in any shape or size for under $50 per module for a modern version of the color organs built in analog back in the day. By the same artists who put together quite a few other cool audio oriented high power LED lighting systems, and other sophisticated lights for use with artwork previously covered on Slashdot. As usual, schematics, design information, and board layouts are available for the project."

Slashdot Top Deals

"It's the best thing since professional golfers on 'ludes." -- Rick Obidiah