Slashdot Mirror

More than any other organization, Intel knows that multithreading is bad. Lots of smart people such as professor Edward Lee (the head of U.C. Berkeley's Parallel Computing Lab) have warned Intel of the disaster down the road. It is time for Intel and everybody else to make a clean break with the old stuff. There is an infinitely better way to design and program parallel computers that does not involve the use of threads at all. Instead of the Penryn, Intel should have picked something similar to the Itanium, which has a superscalar architecture. A sequential (scalar) core has no business doing anything in a parallel multicore processor. Intel will regret this. Sooner or later, a competitor will read the writings on the wall and do things right. Intel and the others will be left holding an empty bag. To find out the right way to design a multicore processor, read Transforming the TILE64 into a Kick-Ass Parallel Machine.

All I can find on that is some references to how the plants growing under the telescope are growing a bit weird because of the electrical fields. Can't find any pictures of the underside.

There are some pictures here.

Also are not the reflector panels in a radio telescope little more than thin wire mesh?

When arecibo was built, the original dish was made out of chicken wire. But as the observatory started using higher frequencies, the holes needed to get smaller and the figure needed to be controlled better than was possible with chicken wire. The current surface is aluminum plates with machined holes. Depending upon where you are and where the sun is, 30 to 50% of the light is probably getting through. (And more importantly, all of the rain.) The spherical surface is maintained to a few milimeters RMS by the anchor wires you see under the dish.

And no, I've never actually been in the pool.

This descibe the current state of the art in pulling sound off old records. Basically they use a microsope and camera to scan the groves. The sound is recovered from the scans. The result is better quality than when the record was new and played on then current players. See the URL below for more info.
http://sciencematters.berkeley.edu/archives/volume4/issue30/story1.php

These old wax recording were made with purely mechanical equipment. It's posable to recover very high quality sound, better then the people who made the records would have imagined. All that hiss and clicks came from defects in the media the optical scanner does not pick this up.

However, meat starts out at so much higher an energy density, the loss from cooking hardly makes a dent, relatively speaking. Meat-eating has been for some time pretty generally considered to be the thing that enabled humans to evolve large brains in the first place.

http://berkeley.edu/news/media/releases/99legacy/6-14-1999a.html
http://www.news.harvard.edu/gazette/2008/04.03/13-aiello.html

This cooking thing may be a more-or-less concurrent change in the same direction, too.

I've heard that even the best organic crops only deliver half of what regular crops do, so if we can produce food for 8 billion today (there's enough but not in the right places) then say we could grow organic food for 4 billion.

Then you have heard incorrect information.

A 22 year study by Cornell, a survey of research by Berkeley (the longest of which is a 150 year study), and a study by the University of Michigan all say that organic farming techniques are at least as good as conventional techniques in terms of yield and often better. In addition to better yields, the organic techniques required less energy inputs, used less water/irrigation, improved soil conditions over time, and retained additional carbon in the ground!

To present a balanced view, this is not true for all crops. Notably potatoes and certain fruits have better results with petro-chemical methods (organic potato yield is roughly 60-65% compared to conventional). However, organic yields are approximately equal for important staples such as corn, wheat and soybeans, as well as many others crops like apples and tomatoes.

On the low end of the studies, a 20 year Swiss study concluded that organic farms produce 80-90% yields compared to conventional farms. And all the studies show that organic techniques have greater yields during drought years.

Instead, machines have a cryptographically encoded conversation to establish both parties' authenticity, using digital keys that we, as users, have no need to see.

I've been doing that for years with SSH. Funny, that.

There is actually a stage in the game at which go becomes mathematically solvable (quickly). It is slightly sooner than a typical strong amateur finds things solved, but still rather late in the game.

http://math.berkeley.edu/~berlek/cgt/gobook.html

I'm not sure this is effectively so different than an endgame database. Certainly as the game progresses the moves become smaller and the game becomes more certain... well, except when you have a couple of dragons chasing each other across the board.

As far as I know, it's not possible to build up an endgame database for go. However, it is very amenable to mathematical analysis using combinatorial game theory.

In the endgame, go tends to break up into a set of smaller "subgames": Most of the fighting ends, except in a few islands dotted around the board. The standard text for mathematical go is "winning ways for your mathematical plays".

Here's a link to a short article expanding on mathematical go, and some other game theory:
http://math.berkeley.edu/~berlek/cgt/cgt-info.html

THEMIS launched in the first half of 2007. I remember because my plasma physics professor canceled class the day of the launch and invited us to the launch party...

The cause of the aurora borealis is something that has not been adequately explained up to now. It seems that magnetic reconnection phenomena in the tail are the trigger, but where exactly? That's what THEMIS was designed to figure out.

This is a very interesting result for plasma physicists and astrophysicists.

http://ds9.ssl.berkeley.edu/themis/flash.html
http://www.nasa.gov/mission_pages/themis/main/

What, no link? Major FTLC (Failure to Look Cool).

Should be Here .

The coastlines on the maps are the more or less modern coastlines, superimposed on the ancient plates, purely to help orient us. I think they assume we don't take the coastlines literally.

There are lots of interesting sites with graphics of continental drift in that period.

This one: http://www.scotese.com/Rodinia3.htm shows both what the coastlines might have been like, as well as having a key map of the modern shapes. And http://www.palaeo.de/edu/scotese and http://www.ucmp.berkeley.edu/geology/tectonics.html have some animations showing the continents moving through the period. Really awe-inspiring (in the meaning, not the quality of the graphics).

but there are so many free games, I simply cannot see the point in PAYING for games. There are free-to-play internet games. There are open source games. There is abandonware. Games, games, games. I could sit down and play a different game each day for MONTHS, all free, and not come back to the same game. So many free games to choose from..... What, exactly, do the game software people offer us? Well - for the latest game to run on my computer, I will have to replace my video card, if I want the full video effects. And, if I'm going to do that, I might as well upgrade the CPU. Ooops, the newest CPU will require a MOBO, and I might as well add some extra memory. Might as well just buy a new computer, huh? Oh, wait, I've fallen into the trap!! The games become more and more demanding of high-end hardware, and like a lemming, I just HAVE to have the latest game, which forces me to buy a new computer almost every year. IMHO, it's just a stupid fad that has hung on to long already. If/when the games makers come up with something new and different, I may take a look at it. Til then, I'll play free games. Actually, though, most of my gaming is online, and I have upgraded my accounts to paid accounts. The graphics may be nothing to brag about, but I play against real people, where intelligence is required, as opposed to fast reflexes, luck, or whatever else one may need to beat a "computer" game. I will also note, that one of man's greatest inventions is pretty much wasted, if all that it is used for is gaming........ Download BOINC, and put your machine to work, searching for cures for diseases, huh? http://boinc.berkeley.edu/

Are you on call for your job every hour of every day?

Odly enough, I am. I average between 400 and 450 hours a month and still find time to manage the companys web site. (a really small one with few hits though.)

And the staff members I know personally would rather drown in raw sewage than play WoW.

That was meant as a play on words targeted at the "WOW" signal, not the MMOG WoW. You know, sit in front of a screen (or printout as the case may be.) for a few years then at some random point in time shout "WOW", then go right back to staring at the screen for a few more years in hopes of scoring again. Get it? Funny shit isn't it?
I've been a member of the SETI@home for a number of years and fully support the effort and have had at one time as many as eleven of my own PCs running BOINC with SETI@home totaling 55,146 credits, but I draw the line at using my tax dollars to fund it.

Good. Maybe soon, all the BOINC users wasting time searching for non-existent aliens will move on to something useful!

World Community Grid is a boinc project , no Seti@home no boinc, no boinc no World Community Grid.

http://boinc.berkeley.edu/projects.php That's a list of other things you can do with BOINC and distributed computing.

...but I recall a set of defaults when I was at UCB which consisted of a black background, white foreground, & red cursor. The files you would need to setup the other colors (pattern matching and whatnot) are (for the moment) at 'http://www.ocf.berkeley.edu/~rchopra/tools/emacs.code-colors.rar'.

In our neighbourhood it's a a lot less dense than average.

Even taking the average of about 1 hydrogen atom per cc, if you had a tube 1 cm in diameter that stretched from here to Alpha Centauri, the total mass inside the tube would be 3e-12 grams.

So yes theres stuff out there, but it wouldn't ruffle your hair if you put the convertible top down on your spaceship.

http://www.icsi.berkeley.edu/~wooters/persistentX.html

I've not tried this but it looks good.

Some for vnc
http://www.novell.com/coolsolutions/feature/16011.html

With xinetd?
http://newsgroups.derkeiler.com/Archive/Uk/uk.comp.os.linux/2006-02/msg00109.html

Any @Home program, from the LHC to SETI to protein folding to Fluid Modeling would all help the world. The one to figure out how many ways to place eight Queens on a chessboard so they cant kill each other...not so much

BOINC runs the common software, and has a good directory of programs. http://boinc.berkeley.edu/projects.php

Clear, irrefutable proof that BSD is dying.

I know, it seems like only nine years ago it was a four-clause license, now all three major BSDs have gone to two-clause licenses. Within a decade it'll be a zero-clause license and BSD will finally die...

--Mike--

Patently false, unless you limit yourself to the retarded design we currently use. Using IFR technology, there is enough fuel for 100,000 years.

Yes, I get it. The burden of proof rests squarely on the other side, always and forever. And the reason is that if someone knew an unconventional truth they could make money off it. But isn't it even easier to make money with a conventional falsehood?

Are you aware of the current state of scientific opinion about Freud? He is at best "controversial", myself I would say he's been throughly debunked.

From an interview with Frederick Crews:

He was a charlatan. In 1896 he published three papers on the ideology of hysteria claiming that he had cured X number of patients. First it was thirteen and then it was eighteen. And he had cured them all by presenting them, or rather by obliging them to remember, that they had been sexually abused as children. In 1897 he lost faith in this theory, but he'd told his colleagues that this was the way to cure hysteria. So he had a scientific obligation to tell people about his change of mind. But he didn't. He didn't even hint at it until 1905, and even then he wasn't clear. Meanwhile, where were the thirteen patients? Where were the eighteen patients? You read the Freud - Fleiss letters and you find that Freud's patients were leaving at the time. By 1897 he didn't have any patients worth mentioning, and he hadn't cured any of them, and he knew it perfectly well. Well, if a scientist did that today, of course he would be stripped of his job. He would be stripped of his research funds. He would be disgraced for life.

There was discussion on Slashdot about the Soft Walls Project that did something similar. See the 1/04 and 7/03 discussions.

What I find interesting is just how vehement software engineers and pilots are about the idea, and yet everyone seems to trust fly-by-wire.

There is Soft Walls FAQ that covers common objections.

There was discussion on Slashdot about the Soft Walls Project that did something similar. See the 1/04 and 7/03 discussions.

What I find interesting is just how vehement software engineers and pilots are about the idea, and yet everyone seems to trust fly-by-wire.

There is Soft Walls FAQ that covers common objections.

The texts are too static and do not accommodate teacher needs; allowing the computer to become the active text is incredibly important.

No, it's not important at all. Until AI advances enormously to the point that software is as smart as human teachers, "active text" is a bunch of silicon snake oil, just a higher tech incarnation of programmed text. Haven't heard of programmed text? Think the non-fiction equivalent of a "choose your own adventure" book; after each little lesson you'd answer a question, the correct answer would take you to the next lesson, the wrong answer would take you to a review of the material. Was a fad in the late 1960s. Mostly useless.

The potential usefulness of OLPC is that it can cheaply put a bunch of books (e-copies of textbooks and works of literature) into the hands of many, and that it enables many-to-many communication on a global scale.

We all have ADD

(Rant mode on.) No, we don't. Many of us still retain the ability to actually sit the fsck down and focus our brains for more than five minutes.

Many - quite likely most - kids labeled as ADD also have this ability, but are not socialized to use it. I teach karate, and since "everybody knows" that martial arts training is good for kids with ADD, I've worked with a lot of them over the years. It's amazing how often simply setting boundaries, enforcing consequences, and getting children to get some gorram exercise "cures" "ADD".

Getting them the hell outside is another useful "treatment". The number of kids with an actual neurological problem is dwarfed by those who just need to run around outside and to be taught some behavioral boundaries.

The problem is that kids glued to game consoles or the intarweb are being trained that fast-shiny-jumpy is good, and slow, complex, and still is bad. (Yes, there are exceptions, games and sites and applications that encourage deeper thought. But they are the rare ones.)

I used to sit in a tree and read books. Maybe that's possible with an XO - I just bought one on eBay, I'll have to try that out when it arrives.

(Rant mode off.)

Absolutely correct. Excel does not follow accepted standards when performing decimal arithmetic.

Here are a couple of good links relevant to this issue:

http://www2.hursley.ibm.com/decimal/

http://mesa.ece.wisc.edu/publications/cp_2005-14.pdf?PHPSESSID=643d8caaab9a8736654674ed089aa4aa

http://www.cs.berkeley.edu/~wkahan/ARITH_17.pdf

BOINC does about 1200 TFLOPS (= 1,200,000,000 MFLOPS) atm.

=> BOINC probably burns around 20MW (assuming that the power used is directly proportional to the CPU time used even if it isn't 100%, which is wrong but an upper bound and probably not very far off).

1 KWh electricity = 0.43Kg CO2

=> BOINC generates 8.6 tons CO2 per hour or about 3100 tons/year (correct me if I'm wrong, I might be a few orders of magnitude off). That isn't very much compared to the 6b tons emitted by the USA anually, but still a waste...

What he's offering (it seems) is the infrastructure, the projects themselves should come from others. I'm a mathematician myself and often have calculations that are large enough to be impractical on a single machine, but not epic enough in scale to attempt setting up my own scheme like distributed.net or folding@home, if I even knew how. Fortunately I already have access to a cluster that I can throw such jobs at, but based on a conference I was at just last week, there are plenty of researchers who don't.

Although whilst there I also learnt of BOINC (and plans to use it for a crypto challenge), so we'll have to see if Nightlife offers any advantages - largely depends on how many machines it makes available, of course!

I have yet to see SETI@HOME actually have a 'event' worth the trillions of cycles wasted on it.

SETI@Home was a prominent beacon for grid computing for the masses, which spawned BOINC, which is being used to simplify distribution and management of other 'useful' grid computing projects such as protein folding and climate change prediction cores. Significant enough event, IMHO.

Here is the heat and energy considerations for running boinc. Boinc is a middleware for projects like seti@home and folding@home. According to that study the difference is roughly $3/month between idle and active computer usage.

http://boinc.berkeley.edu/

  "Use the idle time on your computer (Windows, Mac, or Linux) to cure diseases, study global warming, discover pulsars, and do many other types of scientific research. It's safe, secure, and easy"

  And you can do it NOW. With almost ANY computer.

He's either not done his research or he's an idiot.

BOINC

is a client that allows you to choose out of many projects like Folding@home or SETI. The client also runs on Windows, Linux and MacOS without problems.
There are many configuration options available to control the amount of CPU-power, cores, hard-disk space, RAM, the times it runs, how it should behave is someone else is using the system, etc. and the best is, anybody could set up a project that uses the client (although you'll probably have ahard time getting people to choose your project if it isn't something very interesting).

Check it out!

The Seti-at-home crowd, long ago, realized that it was more than Seti@home, thus created BOINC. So whats new here?

Pam Samuelson has a fantastic compilation of IP law papers.
http://people.ischool.berkeley.edu/~pam/papers.html
Good class, if you can get your hands on the reading list or notes.

when did you do that? IN 1998 they started getting a percentage.\
", the video store pays the studio a one-time fee of $2-$4 per videotape and then pays 40% of rental revenues. "

http://people.ischool.berkeley.edu/~hal/Papers/history/

While running a distributed computing project like BOINC (SETI@home, rosetta@home, FOLDIT@home, etc.).

That way you can feel like you are contributing it to something while you are wasting electricity. Not to mention how cool would it be saying: my picture frame is currently curing cancer.

As I said, there are things you can point to and say "that's evolution" in the same sense that you can point to a falling object and say "that's gravity".

Things that are seen as it happens, not just digging up a few bones and constructing a theory.

Those links are just the first two things I found from a quick internet search. However there is an abundance of such observations where evolution can be said to have been observed as a matter of fact.

As I said, there are things you can point to and say "that's evolution" in the same sense that you can point to a falling object and say "that's gravity".

Things that are seen as it happens, not just digging up a few bones and constructing a theory.

Those links are just the first two things I found from a quick internet search. However there is an abundance of such observations where evolution can be said to have been observed as a matter of fact.

the same Kurt Roeckx had to patch the original infamous May/2006 patch , because he put nested comments and md_rand.c didn't compile :

http://svn.debian.org/viewsvn/pkg-openssl/openssl/trunk/crypto/rand/md_rand.c?rev=173&r1=167&r2=173

This is the complete revision history of md_rand.c at debian, including the last fix by the same Kurt Roeckx that reads

"ssleay_rand_add() really needs to call MD_Update() for buf."

http://svn.debian.org/viewsvn/pkg-openssl/openssl/trunk/crypto/rand/md_rand.c?rev=300&r1=173&view=log

ooopss .... frightening

I have to repeat what this paper says:

" In a narrow sense, the security flaw we found in the Netscape browser serves merely as an anecdote to emphasize the difficulty of generating cryptographically strong random numbers. But there's a broader moral to the story. The security community has painfully learned that small bugs in a security-critical module of a software system can have serious consequences, and that such errors are easy to commit. The only way to catch these mistakes is to expose the source code to scrutiny by security experts.

Peer review is essential to the development of any secure software. Netscape did not encourage outside auditing or peer review of its software-and that goes against everything the security industry has learned from past mistakes. By extension, without peer review and intense outside scrutiny of Netscape's software at the source-code level, there is simply no way consumers can know where there will be future security problems with Netscape's products. "

http://www.cs.berkeley.edu/~daw/papers/ddj-netscape.html

Debian: learn from this !!!

and Kurt Roeckx, don't touch any code of security packages any more ( and yes, you can't nest comments if you want to build md_rand.c , perhaps you should slow down before pushing the "commit changes" button, at least until debian leaders start to put in place tighter and scrict testing-before-commit policies )

Found this post at openssl-dev list by Kurt Roeckx ( AFAIK the Debian OpenSSL Team member that made this RNG-clean patch )

http://www.mail-archive.com/openssl-dev@openssl.org/msg21156.html

Extract:

"What I currently see as best option is to actually comment out those 2 lines of code. But I have no idea what effect this really has on the RNG. The only effect I see is that the pool might receive less entropy. But on the other hand, I'm not even sure how much entropy some unitialised data has.

What do you people think about removing those 2 lines of code?

Kurt

"

BTW, i thought that Debian had some kind of policies about testing each package before committing changes in testing/stable branches. Also, the following paper, contributed by another poster, says interesting things about touching cryptographic code, we have to learn from this experience and have tighter policies !

" In a narrow sense, the security flaw we found in the Netscape browser serves merely as an anecdote to emphasize the difficulty of generating cryptographically strong random numbers. But there's a broader moral to the story. The security community has painfully learned that small bugs in a security-critical module of a software system can have serious consequences, and that such errors are easy to commit. The only way to catch these mistakes is to expose the source code to scrutiny by security experts.

Peer review is essential to the development of any secure software. Netscape did not encourage outside auditing or peer review of its software-and that goes against everything the security industry has learned from past mistakes. By extension, without peer review and intense outside scrutiny of Netscape's software at the source-code level, there is simply no way consumers can know where there will be future security problems with Netscape's products. "

http://www.cs.berkeley.edu/~daw/papers/ddj-netscape.html"

I'm sure the problem will be fixed if the developers acknowledge that the problem exists. Not a big worry.

Yes, it is a big worry because any keys generated with this package are now potentially suspect. This means that anybody who's used Debian or a Debian derived distribution like Ubuntu needs to go back and destroy all host and personal keys generated since 2006. All of those keys are potentially guessable.

And that's a real vulnerability. Early versions of Netscape's SSL implementation (the first SSL implementation) were trivially crackable because of just such a vulnerability.

The Tasmanian plant has been cloning itself for at least 43,600 years, so its genetic code hasn't changed. A part of the point of sexual reproduction is 'swapping' our genes so that our descendants are more readily adaptable to new or changed environments: in a sense, 'editing' our genetic code for each new generation. Although some single-celled organisms do reproduce sexually - in bacteria that's called conjugation - many (usually) don't. So... in the sense that each of us humans is the "same" organism even though we have new cells / some of our cells have died, some blue-green algae are at least a billion years old, some amoeba are more than a million years old, etc. If "continuously existing community of genetically identical cells" is how you define "individual," then some algae mats are awesomely ancient beings!

Baker already heads up Rosetta@Home , a BOINC project that has your computer fold proteins in its spare time. He's appreciated for keeping his journal up-to-date and being responsive to participants; Folding@Home is somewhat less responsive (and doesn't provide the BOINC option).

This is true, but TFA is still marginally useful in that the comments present a link to a more useful article which reflects the same general viewpoint as TFA but does make an argument and present alternatives:

http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-1.pdf

By Edward Lee of the EECS department at Berkeley: http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-1.html. Worth reading if you work with threads.