Slashdot Mirror

Dude, you forgot the links. This is actually a very interesting idea: put the wind generators where the wind is much faster and consistent (though not completely).

Magenn makes a rotating blimp thing.

Sky WindPower uses the turbine to fly as well as generate power. Here's an article at Stanford

Is this a good idea? I don't know how the economics really works out, but certainly there is a much higher energy density at altitude.

The thing is Wind is Flaky, Personally I like to have power all the time, even when there is no wind.

There are two solutions to this problems:

1. Giant Batteries/ Flywheels/ Water storage hills 2. Gas Supplement.

3. Interconnecting Wind Farms http://www.stanford.edu/group/efmh/winds/aj07_jamc.pdf but I agree: in the years to come we still need "conventional", reliable, energy, the cleaner being the nuclear one

"Knuth had his day."

Wow. Just wow.

First, I want you to write a work that tops TAOCP. Or at the very least show your check from Knuth for finding an error. Oh, wait, I highly doubt you've done either.

I'm betting 0x$5.00 that he doesn't have one.

At which point, prepare for this graph to become an understatement.

[The inventor of CS, Emmanuel] Candès...

The way I understand it, there is actually a bit of controversy over whether Candès or David Donoho "invented" compressed sensing. It seems to me that Donoho was actually first, but Candès ended up getting most of the credit.

Your right, besides video games, servers, scientific research, movies, increased productivity, and probably a dozen things I haven't thought of what do we need more processing power for?

Of course efficiency is good. Computers have been becoming more efficient since day one.

There is a place for tiny, low power ARM chips and 150 watt 8 core server chips.

Where's the incentive for the developer to keep cost under control?

The "plus" part is usually small, smaller than the profit that could be had under other types of contracts. That makes "cost plus" contracts not always interesting. As I recall, they pay just enough to keep you afloat, but not much more. Also, the "cost" part is audited frequently; you can't go out, get a yacht and charge its upkeep to the contract - and get away with that.

Heaven forbid that you double check the facts. Not to mention that there is no worldwide measure of the quality/accuracy of any academic papers released.

You're right. There are hundreds of crap conferences. The World Multi-Conference on Systemics, Cybernetics and Informatics, is particularly notorious. They accepted both, Rooter: A Methodology for the Typical Unification of Access Points and Redundancy, and my personal favorite, David Mazieres and Eddie Kohler's seminal work, Get me off Your Fucking Mailing List. So let's just limit ourselves to a top conferences, shall we?

SIGIR is the top information retrieval conference in the world. The acceptance rate was 16% last year, which makes it an "extremely selective" conference in the research world. The acceptance rate has held around 15% - 17% for decades now, and in fact tended decrease as the number of submissions have increased. It accepts submission from worldwide and from both academia and industry.

This analsys from 2007 of papers over the previous 30 years shows that China has moved into 5th overall in number of accepted papers. This is in no small part to Microsoft Research Asia.

So yeah, there are a lot of people just copying stuff around, but there's also a lot of people actually doing extremely good work. You're a fool if you fail to recognize this do your jingoism and racism.

The Unicode standard is 18 years old. Why does Slashdot not support it?

Because it's an English site. ASCII supports every character required.

LOCKSS - Lots of Copies Keeps Stuff Safe:

http://lockss.stanford.edu/lockss/Home

It was Persi Diaconis, Professor of Statistics at Stanford University.
Read the report here: http://news.stanford.edu/news/2004/june9/diaconis-69.html

Theora as good as h264? Yeah, sure. Sorry, VP3 (which Theora is based on) is previous generation codec, comparable to h263. There is no way for it to be as good as h264 unless you use crappy encoder or wrong settings. I like it how Theora apologists compare YouTube videos encoded to achieve balance between size, quality and decoding speed to Theora on maxed out settings and twist it into "they are comparable". Here is more realistic comparison: http://www-cs-faculty.stanford.edu/~nick/theora-soccer/ which shows that Theora requires 60% more bandwidth than h264 for similar quality.

I'm glad you mentioned the MIT open courses. They are also not the only ones that do it! I started my way through their beginning courses in CS but I found it rather dry. After looking around a bit I found courses at Stanford as well: http://see.stanford.edu/ Found the lectureres more interesting and they have some sweet semester long courses on AI (from language processing to robotics). If you have the time check em' out. The resources available nowadays are amazing..

The irony of military robots is that we are using them to enforce a global economic system that is based on forcing humans to do labor in exchange for the right to consume the fruits of industry. Why not just build robots to do the work directly instead? Why not use global networks to freely share information about how to make the world a better place that works for everyone? The same is true for nuclear missiles intended to fight over oil and land instead of using the same technologies to build nuclear power plants (or solar ones and wind ones) or to create self-replicating space habitats or seasteads for endless new land. We need to start thinking in 21st century terms now that we have 21st century technology. Otherwise, we will likely accidentally kill ourselves with the tools of abundance.

As Albert Einstein said:
    http://rescomp.stanford.edu/~cheshire/EinsteinQuotes.html
"The release of atom power has changed everything except our way of thinking...the solution to this problem lies in the heart of mankind. If only I had known, I should have become a watchmaker."

Or further:
    http://www.aip.org/history/einstein/nuclear1.htm
"""
"Concern for man himself must always constitute the chief objective of all technological effort -- concern for the big, unsolved problems of how to organize human work and the distribution of commodities in such a manner as to assure that the results of our scientific thinking may be a blessing to mankind, and not a curse."
"""

Or more on how Einstein was more than the disconnected absent minded professor he is made out to be:
    http://www.wsws.org/articles/2002/sep2002/eins-s03.shtml
    http://www.sacred-texts.com/aor/einstein/einsci.htm

It is not the nukes and drones that may kill us all eventually, it is the unrecognized irony.

AGW skeptics don't ignore long-term trends like the Early Medieval Warm or the Little Ice Age in setting up models. Skeptics don't select a small number of tree-ring samples that Just Happen to fit their ideas while discarding the majority that don't. Skeptics don't add arbitrary, ad hoc adjustments to their data to hide the fact that the data doesn't fit their theory. That's what the alarmists do, and have been caught doing, repeatedly.

Yeah, instead skeptics pretend the MWP and LIA are more than local events. And the bit about "Skeptics don't add arbitrary, ad hoc adjustments to their data to hide the fact that the data doesn't fit their theory"? Are you fucking serious? Even if we ignore non-scientists like Monckton aka Sir Münchhausen with his host of completely made-up graphs - how about the papers from Friis-Christensen and Lassen? Here's one paper that not only rips them apart, mostly by showing where they manipulated and ignored data that didn't fit their theory: http://stephenschneider.stanford.edu/Publications/PDF_Papers/Laut2003.pdf The funny-sad part is, even after they have exposed a hoax, they are still widely touted as the truth (TM) by all skeptics.

Go read about the plagiarism detector used at Stanford. It works on abstract syntax trees inside the compiler, and it essentially gives no false positives while seeing past most obfuscation techniques.

Yes, it is possible, and if you were a compiler expert (like Professor Aiken, who's PhD thesis was this system) your disagreement might mean something. It is very difficult to modify some code enough to get a different-enough syntax tree yet still preserve the core algorithm - I actually believe that knowing how to do so requires MORE knowledge than the solution to the original problem. The idea that somebody could be skilled at obfuscating existing code but not have the ability to write original code is absurd.

Realistically, a professor will not care if you pick up a similar solution, understand it well enough to write your own implementation of the core algorithm, and clearly cite where you got the inspiration. A good professor will either compare the original source to see that you made intelligent modifications or adaptations, and/or will call you in to explain the code to show that you understand the original well enough to have implemented it yourself. Happens all the time, and anyone clever enough to find another source without blatantly cheating is clever enough to demonstrate that understanding.

I've been TAing every semester since I got to college, and every semester we tell people that we run their submissions through MOSS (the canonical code plagiarism detector, hosted at [and perhaps developed at?] Stanford). We exhort that it's really not worth their trouble to try to get their code past it

I always wonder how many people thought they "cheated" by copying code but then ended up changing so much they actually did end up making their own unique code and learning a lot. I know, probably rarely would happen, but just thinking aloud.

I'd really like to see the penalty for cheating to be an immediate failure in the course, if not expulsion.

Now I don't want to advocate cheating but I don't know how well it does for a person to give them a "one strike and your out" deal. As someone pointed out earlier, the pressures of certain universities can be quite high as well as the pressure from just your family and friends. This can push even the best students to think about cheating and I'm sure some have at some point. I think a fat zero on that assignment is mandatory and a firm warning of "we told you before not too" can scare most people straight and not totally screw them over. I know I know, some would say a zero for the course should be a good scare as well but that can sometimes also totally destroy people which I don't think is helpful. A zero on the assignment and a firm warning I would think is good, after that if they do it again, yes, fail their ass cause they won't learn.

In a sort of back asswards way I'd think that cheating and getting caught is almost a learning experience on it's own. You're constantly testing them throughout the semester on this test. They get the material ("Don't cheat, we will catch you"), take the test ("Hmm, maybe I could cheat"), and learn that that was wrong ("Uh no, don't cheat"). Give them a zero for that test and if they fail it again, then they're gone. Well, you're not really ever giving them the test, but I think you get where I'm going with the idea.

But I'm not a teacher, just thinking aloud.

I've been TAing every semester since I got to college, and every semester we tell people that we run their submissions through MOSS (the canonical code plagiarism detector, hosted at [and perhaps developed at?] Stanford). We exhort that it's really not worth their trouble to try to get their code past it, and that they really ought to just contact the course staff if they're in a bind, as there's really nothing worse for them than getting caught cheating. And every semester, we find several pairs of students who have copied each others' code. Sometimes it's a literal, word-for-word copy (comments too) with the name changed (or occasionally without!); sometimes it's the same structure with different comments, suggesting they just sat side by side and wrote the lab together.

I'd really like to see the penalty for cheating to be an immediate failure in the course, if not expulsion. The idea that honest students spend hours working on an assignment, and then someone who didn't plan their time well, or doesn't get things as well, or is too lazy to ask for help thinks they can just not do the work and get the same grade is offensive, and cheaters should be punished accordingly.

Small addition to the above--personally I like observing the sun through the eyepiece with a solar filter--the experience is more personal and you can see more detail.

But as others have pointed out elsewhere on this thread, you can use your telescope to project an image of the sun very easily.

This has a couple of advantages:

  1. No extra equipment needed at all

  2. Several people or even a small group can view simultaneously

  3. You can project the image onto a piece of paper & the sketch the location/shape of sunspots directly

Some more ideas & tips here:

  http://www.skyandtelescope.com/observing/objects/sun/3304766.html
  http://solar-center.stanford.edu/observe/

Also as the single overall best tip for figuring out how to use your telescope: Subscribe to Sky & Telescope:

http://www.skyandtelescope.com/

After a year or so of reading that magazine each month you'll have a much better idea about what to do with your telescope and how to use it.

Or a projector setup for the telescope. You project the image from the eyepiece on a small screen.
http://www.astrosociety.org/education/publications/tnl/05/stars2.html
for more info http://solar-center.stanford.edu/observe/
Actually Solar observing looks like a good class activity.

My thesis advisor rewrote his textbooks every five years or so. But bemoaned losing the original copies of figures and being able to regenerate them in ever-improving computer/print media. So he started the requirement of reproducible documents : the computer programs, both scientific & graphic along with raw data much me assembled together for every figure in your thesis and scientific paper. These would be assembled into makefile-like system to create whatever portion one needed. Then they were archived for posterity. The ultimate test was to "burn" your figures, i.e. erase them from the upcoming document. Then the programs would be run to regenerate them. In practice there are "degrees" of regeneration. Sometimes the figure data is the output of a multi-month supercomputer run. So just the document formating programs would actually be run on the raw output data.
This system was just mentioned in Science magazine Jan 22 2010 p 415. (no free link)

The researchers have programmed Shelley to handle like a racecar by using a set of computer calculations called algorithms

See what happens when you let Liberal Arts majors playing journalist direct the public's understanding of technical things?

Soon: "John's car rolled out of his driveway all by itself and hit a fire hydrant, honey! He should sue General Motors for faulty algorithms!"

You're ridiculing the author for clearly and correctly defining the terminology? If the intended audience of the article is unlikely to know what "algorithm" means, don't you think a concise definition is in order? Now, I'm sure that writers assuming an Ivy League audience would reasonably expect people to know what the big words mean.

Not to mention the fact that there are many Computer Science and Engineering majors who are also capable of effective communication; this is not the sole domain of "liberal arts" majors. Journalism majors... well, make your own judgements.

The researchers have programmed Shelley to handle like a racecar by using a set of computer calculations called algorithms

See what happens when you let Liberal Arts majors playing journalist direct the public's understanding of technical things?

Soon: "John's car rolled out of his driveway all by itself and hit a fire hydrant, honey! He should sue General Motors for faulty algorithms!"

You're ridiculing the author for clearly and correctly defining the terminology? If the intended audience of the article is unlikely to know what "algorithm" means, don't you think a concise definition is in order? Now, I'm sure that writers assuming an Ivy League audience would reasonably expect people to know what the big words mean.

Not to mention the fact that there are many Computer Science and Engineering majors who are also capable of effective communication; this is not the sole domain of "liberal arts" majors. Journalism majors... well, make your own judgements.

The fastest code that I know of for AES in CTR mode is Kasper-Schwabe. It does 8 128-bit encryptions at a time, so it also should be suitable for, say, PMAC if you doctor it. I believe that it does not handle decryption (outside of CTR mode where it's the same as encryption) or other key sizes. Modes other than CTR lose some optimization, and should be ~20% slower. It should be available on Kasper's homepage. It requires SSSE3 and reportedly achieves 6.9 cycles/byte on Nehalem for CTR mode.

My code is available here. On Nehalem, it achieves ~9.4 cycles encrypting, ~11.1 cycles decrypting in essentially any mode. It is suitable for encryption or decryption, and supports all three key sizes (longer keys are slower, of course). A newer (unreleased, experimental) version makes slight performance improvements (maybe down to 9.1 cycles encrypting on Nehalem) and implements an optimization for CTR mode that brings it down to ~7.5 cycles. Email me (mhamburg AT cs DOT stanford DOT edu) if you want to try the experimental version. However, my code fundamentally requires SSSE3, and it performs quite poorly on Conroe.

Also, Dan Bernstein (homepage) has somewhere a fast conventional (not timing-attack resistant, but not requiring any sort of SSE) implementation of AES for several processors, and I've heard Crypto++ is pretty fast too.

I believe that all of the above libraries are public-domain and patent-free.

Out of curiosity, what's your application? Can you just get a VIA Nano or Intel Westmere core and run on that?

while it's clearly tangential from the case at hand (textbooks, etc. which do require a lot of work on the part of the publisher), I find Knuth's writings on the avarice and economic insanity of remaining with Elsevier as a journal publisher to be particularly destructive to the sorts of arguments of alleged publisher value-add that are normally presented.

http://boscoh.com/science/how-the-scientific-publishing-industry-began-to-eat-itself
http://www-cs-faculty.stanford.edu/~uno/joalet.pdf

It would be more noticeable to the people using it if more of them actually were interested in understanding what other people are saying.

It boils down to blaming the tool or blaming the tool. I'm in favour of blaming the tool.

Burden of clarity has been posted here before, today is my turn.

At the high school level, much of the educational process involves jumping through hoops. Students making it into Waterloo have obviously done something right. Top departments there have entrance requirements in the CMU bracket. I recall a year in the 1980s where the median GPA out of high school for entrance to Systems Design Engineering program was 93%. More than a few of these flunked the English assessment test.

Who created the hurdles where you could gain entrance to an elite program such as that and then flunk a basic test of composition and grammar? The adults. And somehow, every generation, the adults get stupider. I have a friend who went to an engineering school in Ontario in the same time frame who had an instructor who promised the class "you'll all thank me later for learning how to neatly hand-letter your engineering blueprints". Twenty years later, it's still too soon to tell. Maybe "later" meant at some point in the aftermath of peak oil and the evacuation of Bangladesh.

If you train your muscle memory to generate "cuz" when you mean "because" it's not the easiest thing to suppress in the heat of the moment when they spring the assessment exam on you in your frosh week. Stopping to correct your hands will interfere with your composition process, which will also be graded negatively.

Another fallacy in play here is Paul Collier's"bottom billion. Fifty years ago the bottom billion was a quartile. Now it's much less than that. Meanwhile, the bell of the income curve has shifted significantly to the right. So despite the fact that the bottom billion has made little progress, there's reason to be optimistic about the bottom quartile.

Notice the effect doesn't show up in an elite university whose intake funnel has not widened to the same degree as the post-secondary education in general.

Clive Thompson on the New Literacy

For the record, I've posted that link before. A rare data point in a sea of whinging is worth posting twice.

Lunsford is interviewed about her own writing process at How I Write which is an excellent resource for those us who would rather walk around the office with our fly undone than pen "your" as a verb in business correspondence.

At 90wpm I'm about 90% at fielding your/you're, their/there/they're, it/it's on the fly. At that speed, I have a poorer track record on than/that or dropping negatives (the last part of the trace to fill in) and the ed/ing problem, which comes to the fingers so swiftly the hands outrace the mind.

I recall that Knuth once brought in Mary-Claire van Leunen as part of a minicourse in mathematical writing, for which I found a fragment: Mathematical Writing by Donald E. Knuth, Tracy Larrabee, and Paul ...

IIRC, she gave one piece of advice I've never been brave enough to try: compose an essay using a crayon to gain insight into your mental processes. That would certainly throw a wrench into my autonomic nervous system. I'm sure it would illuminating if I survived the process with sanity intact.

That's effectively what Waterloo did when I was there: handed me a speed crayon (Bic pen) and wondered why my composition process suffered. I was one of those young people who just weren't as good as my predecessors.

It would be more noticeable to the people using it if more of them actually were interested in understanding what other people are saying.

It boils down to blaming the tool or blaming the tool. I'm in favour of blaming the tool.

Burden of clarity has been posted here before, today is my turn.

At the high school level, much of the educational process involves jumping through hoops. Students making it into Waterloo have obviously done something right. Top departments there have entrance requirements in the CMU bracket. I recall a year in the 1980s where the median GPA out of high school for entrance to Systems Design Engineering program was 93%. More than a few of these flunked the English assessment test.

Who created the hurdles where you could gain entrance to an elite program such as that and then flunk a basic test of composition and grammar? The adults. And somehow, every generation, the adults get stupider. I have a friend who went to an engineering school in Ontario in the same time frame who had an instructor who promised the class "you'll all thank me later for learning how to neatly hand-letter your engineering blueprints". Twenty years later, it's still too soon to tell. Maybe "later" meant at some point in the aftermath of peak oil and the evacuation of Bangladesh.

If you train your muscle memory to generate "cuz" when you mean "because" it's not the easiest thing to suppress in the heat of the moment when they spring the assessment exam on you in your frosh week. Stopping to correct your hands will interfere with your composition process, which will also be graded negatively.

Another fallacy in play here is Paul Collier's"bottom billion. Fifty years ago the bottom billion was a quartile. Now it's much less than that. Meanwhile, the bell of the income curve has shifted significantly to the right. So despite the fact that the bottom billion has made little progress, there's reason to be optimistic about the bottom quartile.

Notice the effect doesn't show up in an elite university whose intake funnel has not widened to the same degree as the post-secondary education in general.

Clive Thompson on the New Literacy

For the record, I've posted that link before. A rare data point in a sea of whinging is worth posting twice.

Lunsford is interviewed about her own writing process at How I Write which is an excellent resource for those us who would rather walk around the office with our fly undone than pen "your" as a verb in business correspondence.

At 90wpm I'm about 90% at fielding your/you're, their/there/they're, it/it's on the fly. At that speed, I have a poorer track record on than/that or dropping negatives (the last part of the trace to fill in) and the ed/ing problem, which comes to the fingers so swiftly the hands outrace the mind.

I recall that Knuth once brought in Mary-Claire van Leunen as part of a minicourse in mathematical writing, for which I found a fragment: Mathematical Writing by Donald E. Knuth, Tracy Larrabee, and Paul ...

IIRC, she gave one piece of advice I've never been brave enough to try: compose an essay using a crayon to gain insight into your mental processes. That would certainly throw a wrench into my autonomic nervous system. I'm sure it would illuminating if I survived the process with sanity intact.

That's effectively what Waterloo did when I was there: handed me a speed crayon (Bic pen) and wondered why my composition process suffered. I was one of those young people who just weren't as good as my predecessors.

...the N900 is an amazing platform. I know it from a computational photography class at my university: http://graphics.stanford.edu/courses/cs448a-10/ It runs a full Linux distro, has a 5MP camera, and now with FF 1.0 I consider it the first phone with a real browser. (IPhone/ITouch/IPad doesn't count because there's no flash and they don't support any browser extensions. Once I can run Flash, Firebug, and Adblock, then it's real.)

I think it deserves a shoutout especially because
*) Nokia is truly awful at promoting their products
*) a certain company that's great at marketing is making all sorts of splash with the antithesis of this phone. it's called the iPad; it runs a Unix derivative, but is an affront to the Unix philosophy. it somehow manages to be three times the size of an N900 with a tenth the functionality.

I think that N900 + FF Mobile is a real tool in an ocean of toys.

This is proven? Really? You have citations? What if - bear with me here - just imagine that people's PERCEPTION of their ability to multitask were proven to be unreliable?

http://news.stanford.edu/news/2009/august24/multitask-research-study-082409.html
http://blogs.discovermagazine.com/80beats/2009/08/25/multitaskers-are-bad-at-multitasking-study-shows/
http://www.google.com/search?q=multitask+study&ie=utf-8&oe=utf-8&aq=t&rls=Swiftfox:en-US:unofficial&client=firefox-a

Go. Browse. Study.

While it is true that women in general are better multi-taskers than men are, NO ONE is such a great multi-tasker that they should risk their lives on it. Reading the myriad of real studies done on the subject, you will learn that those people who THINK that they are great at multitasking are less likely to be good multitaskers than some of us who DO NOT think we are so great at it.

Personally - I'm a male, and I resent when people want me to do 6, 12, or 100 things at once. I want one task, I want to focus on it, I want to complete it to the best of my ability, then I want to move on to the next task. And, the STUDIES show that I complete more jobs, with better results, than the multitasker who THINKS that (s)he is being productive.

Multitasking. There was a term for that long, long ago. Women who wouldn't focus on a single task, and flitted from one thing to another, were called "flighty".

As for testing people - what planet are you from? Here on earth, we routinely test people for driving skills, and award driver's licenses to people who have zero driving skills. You wish to test them, and award a higher class of license? Get real.

Some people also think that they can drink copious amounts of alcohol, and still be sober enough to drive. Such people are suffering from impaired judgement.

People who think they can multitask efficiently are also suffering from impaired judgement. Such people should NOT be permitted to drive, as they put lives at risk.

I decided to go ahead, and I know exactly why I made that choice based on scientific data. If someone else is informed of the scientific data and chooses against circumcision, I fully respect that and have no problem with it.

I am struggling with this decision for my son; can you please provide the scientific data?

Having viewed the circumcision video posted elsethread, right now I'm thinking no.

...Stu

The procedure is by no means a pleasant thing to view, but that video is far from the truth I witnessed twice. That video DOES in fact look pretty barbaric. I think I might have throttled that doctor myself if I'd been there. It's no wonder the anti-circ crowd wants to use THAT as their propaganda. You want a run-down of the procedure itself?

Diaper removal. Doc hands me a small creamer-sized container of sugar water, into which I intermittently dip my (gloved) pinky, and my son happily sucks/drinks the sweetness. Puts him at ease a bit, during which time the pediatrician uses a small needle to administer an anesthetic nerve block on the groin area once on each side of the base of the penis.

Give a minute or two for the nerve block to do its thing, while giving sugar water. Baby sucks/drinks, but never cries (neither of them did). See the last paragraph of this section:
http://en.wikipedia.org/wiki/Circumcision#Pain%20and%20pain%20relief
It describes exactly what procedure my sons received. Maximum pain relief possible. And the procedure my boys had looks a whole LOT more like what you can see here:
http://newborns.stanford.edu/Gomco.html

Keep in mind that I had a very experienced pediatrician who's done hundreds in the duration of his practice, and each procedure was less than 10min.

As for the science, you can ask your pediatrician or doctor about the benefits in terms of infections and STD compared with those of uncircumsized individuals. I can't quote numbers, because it's been 4yrs since I had that discussion with my pediatrician and can't recite the data from memory. For us, it was also a matter of the boys feeling confident in themselves when they know they look "the same as daddy" (I expect all who have never made this decision to balk at the thought).

I definitely won't tell you what to do. Get a doc that you trust. Ask for his professional input to make your own decision. Whatever you decide, you're the boss, don't forget it. Best of luck to you!

Many of these biology experiments require very expensive machines, such as microarray machines, as mentioned by the article. I don't know if purchasing refurbished machines is a wise choice since we don't want data quality to be compromised.

A microarray is pretty expensive yes, but a lot of DIY biology could be done with just a computer and or a secondhand PCR machine. Used PCR machines apperantly can be had for under a grand. Even less if you can service a broken one yourself, which many of these DIYers seem capable of. Probably won't have all the fancy options of a higher priced one either, but our academic lab has an expensive cycler with many options that we never use.

Data quality with many of these things is less tempermental than a microarray too. The secondhand PCR machine in this case might not be good for sequencing, but it would be a great tool if you were, say, making a plasmid to make glowing bacteria, using it to identify species of plants, making in-situ hybridization primers. There are a lot of things you can do with a basic cheap PCR machine.

As far as microarray data goes, an affymetrix premade microarray chip goes for about a thousand dollars. Obviously it's not feasible for most people to do many of these out of their own pocket, but not everyone does. Say you want to find out what genes are expressed more in dog breed A than dog breed B. If you were wanting to publish that data in a peer-reviewed journal, you'd probably need 6 chips, it seems like most people I know who do microarray do triplicates. If you were just wanting to find out for yourself, like to find canidates for which genes produced trait X that was in breed A, you could do just two, one for each, and hope it wasn't wildly innacurate. You could then focus your search based on that, taking it with a grain of salt until you confirmed it through other, less expensive means.

If you were going to be doing many microarrays, this website appears to be a guide for making your own microarrayer. The price tag for building it exactly as that lab says to would be about $24k. Again though, many DIYers are mechanically inclined and could cut corners for their own purposes.

Another issue is gathering the samples. If you're collecting yeast, that would be simple. Arabidopsis, other small plants, mice, or other small animals, you probably need quite some space.

I don't see that. Our lab studies chicken embryos. An egg incubator is pretty small. C elegans can be grown wherever you've got space. Arabidopsis can grow in the yard, you don't need acres. A research-grade mouse colony would be expensive yes (maintaining a genetically pure mouse colony in a sterile environment free of variation is harder just obtaining mice from the street). If you need other model organisms, there are farms. It can be a limiting factor, yes, but when is that not true? You can't exactly use elephants as a model organism in really any lab in the world.

Humans? That won't be simple at all. You have to clear privacy issues, getting the research review board to sign papers, etc.

Which research review board? If I'm comparing gene expression in human blood samples in my garage, without using public grant money, the "review board" is whatever poor saps I sucker into giving me their blood.

You can always resort to publicly available data. But chances are that you won't be able to impress scientists much for going that route. Also, most of the important discoveries are already done on this data.

I reject both of those claims. Real scientists recognize valid results independant of the professional nature of the researcher or his lab. Hell, most of us "p

Many of these biology experiments require very expensive machines, such as microarray machines, as mentioned by the article. I don't know if purchasing refurbished machines is a wise choice since we don't want data quality to be compromised.

A microarray is pretty expensive yes, but a lot of DIY biology could be done with just a computer and or a secondhand PCR machine. Used PCR machines apperantly can be had for under a grand. Even less if you can service a broken one yourself, which many of these DIYers seem capable of. Probably won't have all the fancy options of a higher priced one either, but our academic lab has an expensive cycler with many options that we never use.

Data quality with many of these things is less tempermental than a microarray too. The secondhand PCR machine in this case might not be good for sequencing, but it would be a great tool if you were, say, making a plasmid to make glowing bacteria, using it to identify species of plants, making in-situ hybridization primers. There are a lot of things you can do with a basic cheap PCR machine.

As far as microarray data goes, an affymetrix premade microarray chip goes for about a thousand dollars. Obviously it's not feasible for most people to do many of these out of their own pocket, but not everyone does. Say you want to find out what genes are expressed more in dog breed A than dog breed B. If you were wanting to publish that data in a peer-reviewed journal, you'd probably need 6 chips, it seems like most people I know who do microarray do triplicates. If you were just wanting to find out for yourself, like to find canidates for which genes produced trait X that was in breed A, you could do just two, one for each, and hope it wasn't wildly innacurate. You could then focus your search based on that, taking it with a grain of salt until you confirmed it through other, less expensive means.

If you were going to be doing many microarrays, this website appears to be a guide for making your own microarrayer. The price tag for building it exactly as that lab says to would be about $24k. Again though, many DIYers are mechanically inclined and could cut corners for their own purposes.

Another issue is gathering the samples. If you're collecting yeast, that would be simple. Arabidopsis, other small plants, mice, or other small animals, you probably need quite some space.

I don't see that. Our lab studies chicken embryos. An egg incubator is pretty small. C elegans can be grown wherever you've got space. Arabidopsis can grow in the yard, you don't need acres. A research-grade mouse colony would be expensive yes (maintaining a genetically pure mouse colony in a sterile environment free of variation is harder just obtaining mice from the street). If you need other model organisms, there are farms. It can be a limiting factor, yes, but when is that not true? You can't exactly use elephants as a model organism in really any lab in the world.

Humans? That won't be simple at all. You have to clear privacy issues, getting the research review board to sign papers, etc.

Which research review board? If I'm comparing gene expression in human blood samples in my garage, without using public grant money, the "review board" is whatever poor saps I sucker into giving me their blood.

You can always resort to publicly available data. But chances are that you won't be able to impress scientists much for going that route. Also, most of the important discoveries are already done on this data.

I reject both of those claims. Real scientists recognize valid results independant of the professional nature of the researcher or his lab. Hell, most of us "p

I mean, I kind of agree with you, many people shouldn't be driving and we pay a 55,000 person a year toll in deaths from accidents. But what's the alternative?

Automated cars like Stanley.

E.O.Wilson wrote an essay in the late 90's suggesting China is the test case for humanity's attempts to find solutions to environmental and population problems. China as a traditional agrarian male dominated culture has moved from a practise of female infanticide to using technology to abort female foetuses. From this practise a sex ratio imbalance has arisen that some see as of little current or historical importance. The nation's one-child policy could leave 24 million bachelors by the year 2020. My own readings in history have taken on views more in line from what has been learnt from the last few decades of research in primatology. Chimpanzee behaviour favouring, figuratively speaking, male oligarchies restricting access to resources maps clearly, in my mind, onto all three, still widely practised, Mediterranean death cult religions promulgating male dominated societies. Based on China's current sex ratio imbalance the questions to be addressed probably can be set in historical, anthropological and primatological contexts.

Personally I suspect China flirted with democracy, but as is nearly always the case, power structures are not given to relinquishing dominion. Recently /. ran a story that the Chinese government replaced the movie "Avatar" with a biography of Confucius. The works of Confucius are only known by way of reconstructions, but his core message seems to have been one of a familia philosophy, strongly patriarchal, and, in that light, like the Christian, Islamic and Judaic cults that I find map well onto Chimpanzee behaviour. The core mandate of such power structures is submission and tradition. I suspect the Chinese government, if not the Chinese people, are moving away from democracy and into a tradition bound version of Confucianism, but at best it's only a superficial reading.

The discussion can go on and deeper but one current salient point should be made. Chinese society is observed to be much more family orientated than our western societies. A recent rampage killing in the international press was reported on as having happened in western societies because the killer was deranged, whereas the Chinese feedback suggested the man went on a killing spree because his family wasn't there to support him. Western society is strongly vested in the rights of the individual, China not nearly so much. If the West and China and, perhaps much of Asia, are to achieve an equilibrium than we're going to have to bridge this core cultural divide from both sides.

just my loose change.

I did my best attempt at an unbiased comparison which shows Theora to have about a 30% disadvantage, although it uses a slightly older version of Theora: http://www-cs-faculty.stanford.edu/~nick/theora-soccer/

also, photocopying a book is established to be a violation of copyright when the book is still protected by copyright and the copying does not meet the conditions of fair use (in the US).

e.g.
http://fairuse.stanford.edu/Copyright_and_Fair_Use_Overview/chapter7/7-d.html
http://www.kasunic.com/article1.htm

even some "educational" uses are not fair use, as was found in the more recent Kinko's case discussed in the second link.

No way, they are/will be using the unmanned submersible robots (http://www.liquidr.com/ and http://www.stanford.edu/class/ee380/Abstracts/091028.html) for their high risk payloads. The logistical chain must be unbroken for the profits to come. This thing is used after a hard day at the cocaine plants and shady offices to get really high.

To you. Are you a doctor? Are you a virologist? What the fuck do you know? Nothing. Do you think anybody in his right mind is going to risk the lives of hundreds of thousands of people by paying attention to you?

...

The 1918 flu caused 650,000 deaths. Nobody really knows why. We could have another epidemic like that any year. When the new flu comes up, nobody knows until it's all over whether it's going to be the big one until it's all over.

Ordinarily, I am courteous and tend to cut people some slack. I will, however, make an exception in your case: You don't know jackshit. Fool.

The 1918-1920 influenza pandemic killed tens of millions. That's MILLIONS. Big difference. This reference says 20 to 40 million. This says 50 to 100 million. Going by the conservative estimate, that is several times the number of people that were killed in the First World War itself. Maybe you should acquire some courtesy until you succeed in your quest for omniscience.

There's already such a project underway from Stanford: http://qcn.stanford.edu/. Doesn't extend to phones yet (as far as I know, though they're probably working on it) but will use the accelerometer in MacBooks, Thinkpads, other laptops that may have them, or you can get a USB attachment that has one for desktop computers. Works essentially as you describe.

Phones might be tough because they move around a lot with the person who carries it, but would be especially useful because if they've got an accelerometer it's likely they also have a GPS unit and a way to tell which direction it's pointing, which you don't have on laptops. If people were in the habit of leaving their phone on their desk while they're sitting there, it could certainly be useful - you could just ignore those with a lot of motion because obviously the person is moving around with the phone.

I have a MSCS from Stanford (1985), and the field has changed since then. Back then, it was all about discrite math - number theory, combinatorics, mathematical logic, computability, and proofs. There was no number-crunching at all in the curriculum. Of course, back then, an FPU was an extra-cost option on a PC. I've actually done automated program verification work. But outside of IC design (where formal methods are routinely used), there's not much of that going on now. Now, number-crunching has come to the fore.

In the 1990s, I spent several years on what turned into ragdoll physics for games. That's all about differential equations and number-crunching. I had a hard time switching over. But I finally got used to deterministic number-crunching. I have no mathematical intuition for it, though; I took it up too late in life.

Now, the leading edge of computer science is probabilistic number-crunching. Take a look at Stanford's CS229 - Machine Learning class. That's the technology that's driving AI now, and it's working across a broad range of domains. The logicians are out, and the statisticians are in.

This has been tried before. See the "Clean Slate" program at Stanford. Basically, it's a plan to redesign the Internet to put telcos more in control. The emphasis is on identifiable "flows", allowing the endpoints, bandwidth, duration, and traffic statistics of a flow to be identified. Visualize an Internet that allows cell-phone like billing and you have the telco dream.

Read the OpenFlow white paper. The basic idea is that, every time a new "flow" appears, the first packets are forwarded to Master Control, which decides what to do about them. Deny? Wiretap? Throttle? Report? It's all up to the "Controller". See page 3, col. 1. This is implemented by making ordinary routers "OpenFlow compatible". Most routers already have flow tables. Currently they're mostly caches for routing info. OpenFlow puts them under centralized control.

With relatively minor mods designed into existing router FPGAs, (or software - there's a Linux implementation for test purposes, and downloads for some Linux-based routers) they can be OpenFlow compatible. They can act like ordinary routers until a controller contacts them and takes them over.

The hype is about "enabling innovation in campus networks", but the reality is that it puts a central controller fully in charge of, and fully aware of, each user's connections.

To put things in perspective, a circa-1972 Xerox Alto workstation would be about $388,000 in 2009 dollars, but I can't imagine anyone preferring one to today's $399 laptops (about $77 in 1972 dollars)! :-)

We currently only use a small fraction of a percent of the energy in our Uranium fuels, and we only use a small percentage of Uranium (U235) in the majority of our current reactors. With full reprocessing (something that no nation currently does, as it's more expensive than just mining more even with the costs of storing un-reprocessed waste) we have enough fissionable fuels for breeder reactors to produce our current global electricity demand for roughly a billion (or milliard, i.e. 109) years. That's many times longer than the human race has existed thus far.

Granted, energy demand is constantly increasing, and a lot of the fuel is hard to extract (there's lots of Uranium in the oceans, but it's in very low concentration). Even so, we have decades of fuel available at current technologies and current price points. Thorium reactors would extend that by a factor of about 4, without even getting into breeder reactors or reprocessing. After a couple of centuries, we'll either have working fusion , vastly better mining technology, or vastly better reprocessing capability (this is already being worked on, via several approaches). That's not a major crisis, by any means - that long ago, we were still using coal-fired steam.

I suppose it depends on how you define "sustainable" though; eventually the sun will burn out and then your solar, wind, and hydro powers won't work anymore either. At the current rate of consumption though (roughly 15% of global electricity usage) we would *still* have plenty of fissionable fuel. That's sustainable enough for me.

Sources:
http://en.wikipedia.org/wiki/Fission_power#Conventional_fuel_resources
http://www-formal.stanford.edu/jmc/progress/cohen.html

I remember reading papers on this during my AI classes in the mid 90's. I don't see how this is impressive nearly 15 years later.

Here's the first link I found on G.P. Music from '98 which actually had the computer rate some of the music.
http://graphics.stanford.edu/~bjohanso/papers/gp98/johanson98gpmusic.pdf
If you look at his references, people were doing this in the '80's.

No, I didn't RTFA. I didn't even read the article I linked in this post, so don't get upset if they aren't completely related.

but I think you're missing my point

No, having been a psychologist in a former life :) I just wanted to give a hint that 'psycohistory' may (and will) have a bright (or dark, YMMV) future — though with a different concept of large samples.
Which then will further complicate the notion of 'free will'.

CC.

nobody is quite sure as to how public bank account numbers ought to be

Everyone who writes checks are, they'll give their numbers to anyone. Just ask Knuth.

Having lived in Switzerland for a while and experienced the cheque-free banking system there, I can say that cheques suck on so many levels. Handing or mailing someone an IOU in the form of a cheque is stupid when you consider the alternative.

In Switzerland, and I believe in most of Europe, payments are pushed rather than pulled. The receiving party sends the paying party a standard slip with the receiver's account information and amount being billed (or the payer could fill out a blank slip manually). The payer feeds the slip to his own bank's ATM and authorizes the payment. Or, he keys in the information to his bank's e-banking website. Alternatively, they payer can take the slip to any post office and pay with cash. The transaction clears the same day.

Compare that with a cheque-based system:

• The receiver's bank has to demand money from the payer's bank, and typically imposes a hold period on that money.
• The payer doesn't know when the receiver will deposit the cheque; the possible delay makes reconciling accounts a bit messy.
• The receiver doesn't know whether the cheque will bounce -- he's just getting an IOU.
• The payer can easily overdraw his account, through carelessness or maliciousness, and be penalized by both his bank and the receiver.
• The receiver can claim that the payment wasn't received on time, due to mail delays, hold periods, etc.
• The payer can claim that the "cheque is in the mail", when of course it hasn't been sent yet.
• The payer has to worry about whether the receiver has tampered with the cheque (e.g. altering the amount).
• The bank has to authenticate the cheque by verifying the signature, which probably doesn't happen properly in most cases.
• Because the authentication system is basically based on trust, the payer is exposed to massive cheque fraud! Sending a cheque means giving out your account information, which is just as bad as giving out your credit card number. A receiver-pull system is inherently less secure than sender-push. If everyone agrees to do sender-push only, there is no risk involved in revealing your account information.

There are only two advantages of cheques that I can think of:

• Giving someone a casual gift. You can easily write a cheque as a birthday or wedding gift, knowing just the recipient's name. In those situations, it could be socially awkward to ask for the recipient's account information.
• Paying someone who doesn't have a bank account. I understand that many poor people (illegal immigrants?) in the U.S. don't have bank accounts. They end up taking their paychecks to some check-cashing place that charges a hefty fee. This is a rather weak "advantage", since checks are a sub-optimal solution anyway -- possible sane solutions would be to stop hiring illegal immigrants, or let them have bank accounts, or pay them in cash.

In summary, a cheque-based banking system is so completely backwards and broken, it's amazing that such a system could exist in the modern world.

> That's because at my core I am a greedy self-serving animal. The only difference between me and the apes is that I horde money instead of food. You'll never change this.

FTFY. You speak only for yourself. You are also dead wrong.

I used to be a libertarian too, with a card and everything. Then I realized that I don't live in a vacuum, my peer group is actually the entirety of mankind, and there is no "other". Communism as implemented was flawed, to understate it, but to hold capitalism up as the best we can manage while simultaneously portraying humans as incapable of evolving beyond our basest instincts is perverse, to make another understatement.

If you're going to reference Sci-Fi, at least quote a good author:

"The saddest aspect of life right now is that science gathers knowledge faster than society gathers wisdom."

"The Earth should not be cut up into hundreds of different sections, each inhabited by a self-defined segment of humanity that considers its own welfare and its own 'national security' to be paramount above all other consideration. [...] There are no nations! There is only humanity. And if we don't come to understand that right soon, there will be no nations, because there will be no humanity. "

"Anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that 'my ignorance is just as good as your knowledge.'"

                    -- Isaac Asimov

So many questions would be answered if the actual science was available.

Too bad PNAS charges for people to see that. If you find Yi Cui's site at Stanford and look under "Publications," you may find a relevant pdf.

*sigh* if only the editors knew how to use the internet... or is it that they don't know science is peer reviewed and not press released?

http://stephenschneider.stanford.edu/Publications/PDF_Papers/MobergEtAl2005.pdf - so now cite a peer-reviewed paper saying the MPW was warmer than today. Not warmer than in the 70s, but today.

The last decade has seen a revival of various hypotheses claiming a strong correlation between solar activity and a number of terrestrial climate parameters: Links between cosmic rays and cloud cover, First total cloud cover and then only low clouds, and between solar cycle lengths and Northern Hemisphere land temperatures. These hypotheses play an important role in the scientific as well as in the public debate about the possibility or reality of a man-made global climate change. I have analyzed a number of published graphs which have played a major role in these debates and which have been claimed to support solar hypotheses. My analyses show that the apparent strong correlations displayed on these graphs have been obtained by an incorrect handling of the physical data . Since the graphs are still widely referred to in the literature and their misleading character has not yet been generally recognized, I have found it appropriate to deliver the present overview. Especially, I want to caution against drawing any conclusions based upon these graphs concerning the possible wisdom or futility of reducing the emissions of man-made greenhouse gases.

My Findings do not by any means rule out the existence of important links between solar activity and terrestrial climate. Such links have over the years been demonstrated by many authors. The sole objective of the present analysis is to draw attention to the fact that some of the widely publicized, apparent correlations do not properly reflect the underlying physical data.

Gee, where is the uproar about this Sungate from 2003?