Slashdot Mirror

Ok...it's an improvement on size. But this is already demonstrated by UC Berkeley on a heavier scale

Get Clifford Stoll, he works for free!

From Socrates to Expert Systems.

It argues that rules based AI is a dead end. It also classified levels of expertise.

It would seem like this non-rules-based IBM brain simulation method would be one which could possibly go beyond the 'advanced beginner' stage that Professor Hubert Dreyfus proves that rules base systems are limited to.

From the ChurchStreet's FAQ.

10) Can the reconstruction software be purchased?

Not unless you are a qualifying intelligence agency. Our digitizing techniques and proprietary software cannot be purchased at this time unless your team is a high level governmental intelligence team. For others, ChurchStreet offers the reconstruction as a service, not as a product to be purchased.

Qualifying intelligence agencies should call ChurchStreet directly to get more information about our Reconstruction Software Suite.

This is just a simple(?) exercise in matching edges and colors at those edges to each other in all the piecess. This is how a standard jigsaw puzzle is assembled in 'meatspace'. ChurchStreet's software likely does this all inside the computer after the document shreds have been scanned in.

These guys are in the best postion to write/adapt such software and make it available to the public at large--not just government intelligence organizations.

P.S. For secure document destruction, burn it--it is the only way to be sure the document cannot be reconstructed. This applies to assorted forms of computer related information storage and processing--just toss the hard drives, CD-ROMs, floppy disks, RAM chips, memory sticks, motherboards, CRTs, etc., into the nearest (approved) incinerator and be done with it. It's an environmental/safety nightmare but the data in the destroyed media is now gone for good.

Want to give the ChurchStreet boys an 'impossible job?' Do the following:

1) Print up a document in English using a monospaced font.

2) Cross cut the document so that each character is in its own square 'cell' and is completly surrounded on all four sides by whitespace.

3) Hire ChurchStreet to reconstitute this document and send them the 'confetti'.

They won't be able to reconstruct the document because all the pieces are edgewise topologically identical to each other. The best they can do is use all the 'letters' and reconstruct all the words in the document. If they accomplish that, then they have to put them into the right order. If the document had 58 words on it, there would be so many message combinations that you could easily assign each one to every atom in the universe.

If their 'proprietary' document reconstruction techniques take into consideration the texture, grain, and thickness of the paper then they would stand a fighting to good chance of reconstituting such a 'challange document'. :)

They have

• Windows -- 89.5%
• Linux -- 7.8%
• Darwin -- 2.3%
• Other -- 0.4%

Now, this data is obviously skewed with respect to the total distribution, since the people who run something like SETI@home are probably more technologically inclined than the average computer user. This would mean that the percentage of non-Windows OSes is higher in this sample. On the other hand, the software for BOINC (SETI@home) is still somewhat Windows-centric, which would in turn increase the Windows share in the sample.

An interesting data point, nonetheless.

They have

• Windows -- 89.5%
• Linux -- 7.8%
• Darwin -- 2.3%
• Other -- 0.4%

Now, this data is obviously skewed with respect to the total distribution, since the people who run something like SETI@home are probably more technologically inclined than the average computer user. This would mean that the percentage of non-Windows OSes is higher in this sample. On the other hand, the software for BOINC (SETI@home) is still somewhat Windows-centric, which would in turn increase the Windows share in the sample.

An interesting data point, nonetheless.

I had to look him up on the internet. He has a web page here:

http://www.ocf.berkeley.edu/~stoll/

It is funny, because it is really simple and it is actually invalid. I had to view the page source to find the intended links.

It has the answer to one of the number puzzles that he mentioned in the book (but never gives the answer to if I remember correctly).

For everyone who uses Blender, take a look at the BURP project. It's a distributed computing project that runs on BOINC. BURP aims to develop a publicly distributed system for rendering 3D animations. Eventually, the goal is for people to be able to submit work for rendering, though they're not that far yet. The code is still in development, so workunits are just to clean bugs out of the system right now.

closest thing I could find.

http://www.cs.berkeley.edu/~daw/netscape-randomnes s.html

If you like that style, you may also like Dave Patterson's "How to have a Bad Career in Research/Academia".

There is such a thing as safe programming.
There are safe languages.
There exists formal methods.
There are best practices in programming.
There exists tools for source code verification.

If you program and don't care about any of these things, hey, guess what - you're 20 years behind in your programming practices and your reading list. Even if you program in C, you can adopt better practices (*).
90% or more of the problems related to software security spring from C/C++ hacking without any method of program verification for correctness. Just read a security site vulnerabilities list.
If only people were to program: medical; military; aerospace software like Firefox or IE programmers, the we'd all be dead one way or another by now.

(*) see OpenBSD for instance and compare their security advisories with Linux or Microsoft.

PS: Just one such example of a little used tool: CIL - Infrastructure for C Program Analysis and Transformation

. . . let's be alarmist about it, because the info didn't exist anywhere else.

http://www.berkeley.edu/news/media/releases/2005/0 5/images/buckwheat_fly.jpg

What kind of fly is that?!

Check out PlaceSite: http://www.sims.berkeley.edu/~savage/ps/ a project from SIMS @ Berkeley, all about getting people in a Cafe to not fall victim to the zombie effect and actually talk to each other!

specious argument. You are assuming, of course, that worms in the IPV6 world will crawl the way they do now, by generating a random IP address and trying to attack it.

Suppose they listen for broadcasts, and attack addresses they hear from?

Suppose they assume that a PC which has an Intel NIC in it, is in an organization that likely has more Intel NICs in it, and concentrates it's energies searching for other PC's in that 40 bit space?

Suppose the worm simply accepts the slow build-up of hosts; rather than having a Slammer-type worm that infects the vulnerable population in 10 minutes (http://www.cs.berkeley.edu/~nweaver/sapphire/), perhaps you end up with worms that take days to build up rather than minutes; the low rate of scanning would probably keep them below the radar of the network monitors until their exponential growth started creating a blip.

Perhaps the worms would start linking together to divvy up the address space to search. If you have two copies of the worm inside an organization, there's no reason at all that they should be attacking the same addresses. Imagine a giant P2P worm network that served to coordinate the attack.

Perhaps you have a worm that gets launched from 10,000 hosts simultaneously; we know that zombie networks can be this large. Perhaps the zombie PCs that launch the attack spend a week or two simply collecting IP addresses - off broadcasts, off the local DNS, off low-level PING requests. Each one could have hundreds or thousands of initial, valid, local targets to pinpoint before D-Day.

Perhaps you have a breakthrough worm that does all of these things.

Thanks, but I'll keep my NAT box.

Direct link to the article: http://www.icsi.berkeley.edu/~nweaver/login_witty. txt

One of the better worm analysis papers I've read was "Reflections on Witty" by Nicholas Weaver and Dan Ellis (of MITRE), published in the June 2004 issue of ;login, the Usenix magazine.

Rather than a dissection of the worm itself, the authors give a detailed analysis of the author/attacker of Witty.

Some insights about the worm author that Weaver and Ellis proposed:

• he was a fairly proficient programmer - there were no significant bugs in the code of the worm, he knew how to program x86 assembly and access the Windows API, he implemented a stack-overflow attack, and most importantly, he constructed a payload that was malicious to the host, but didn't significantly slow the worm's spread.
• he was quite clever at what he did - randomly padded packet sizes, randomized the destinations and port numbers, and he seeded the worm (rather than start at a single location, the worm started out from 110 different victims) -- prior to this no one had significantly seeded their worms
• he wrote compact code, Witty consists of 177 x86 instructions in 474 bytes (the rest is the buffer overflow and padding); with 177 instructions, he was able to construct routines to cleanup from the overflow attack, seed the RNG, propagate the worm, and execute the malicious payload (Witty slowly overwrites disks on the infected hosts until the machine crashes)
• he worked quite fast; the stack overflow in the ISS BlackIce products was published on March 18, 2004. Witty was released on March 19, 2004, less than 48 hours after the security advisory was published by eEye; it is possible that he knew of the vulnerability when eEye notified ISS on March 8, 2004, but the paper goes into why this is unlikely
• he probably tested the worm before he released it (cf. the lack of major bugs); this combined with the fact that he seeded on 110 hosts, means that he had access to a wide array of compromised machines -- it probably means he has access to the "hacker underground", to gain access to these machines in such a short time frame

The authors' conclusion is somewhat alarming, they reason that Witty represents a new generation of virus/worm authors: motivated, skilled and malicious individuals who are experts at what they do.

Well, the easiest fusion reaction to do is the Deuterium Tritium reaction (DT). That is, it is the reaction which requires the "lowest" temperature to ignite. Thing is, most of the energy released in this reaction is in the form of hot neutrons. The percentage of the fustion energy released in the reaction as neutrons is called the reaction's "neutronicity" and is something like 80% for DT. This really sucks because neutrons, as you may be aware, are absorbed into the nuclei of the surrounding structure material, transmuting its constituent atoms into radioactive isotopes (albeit with relatively short half-lives). Soooo, the best idea around these days is to create a vacuum target chamber with ...wait for it.... undulating "waterfalls" of hot liquid lithium or "filbe" (Lithium Fluoride Beryllium Fluoride mix). The Li absorbs the neutrons and is heated in the process, the heat is then sent to boil water/run turbines, and the usual. There is a bonus in this scheme though, the Li after absorbing a neutron is transmuted into more Tritium! More Fuel! This is called the HYLIFE II reactor design.

Your explanation is total crap. It doesn't even lead to the sorts of results provided by the study.

See http://www.ocf.berkeley.edu/~wwu/riddles/easy.shtm l#feministRule and http://www.ocf.berkeley.edu/~wwu/cgi-bin/yabb/YaBB .cgi?board=riddles_easy;action=display;num=1030742 212

Your explanation is total crap. It doesn't even lead to the sorts of results provided by the study.

See http://www.ocf.berkeley.edu/~wwu/riddles/easy.shtm l#feministRule and http://www.ocf.berkeley.edu/~wwu/cgi-bin/yabb/YaBB .cgi?board=riddles_easy;action=display;num=1030742 212

I've read the comments posted to this story, and decided again to look into Octave. A quick google pointed me to the categorized list of Octave & Octave-Forge functions.. This list is fairly complete, and extremely useful because it lists what's missing.

Unfortunately, its missing a lot of features I've grown accostumed to using in Matlab. switch...case and varargin / varargout were two that jumped out at me. It appears the functionality is provided, but not in a compatible way.

Ah well. Half of my work is done in Simulink anyway, and the libre equivalent I've seen most people point to stacks up about as well.

Wu :: Riddles

http://emlab.berkeley.edu/~moretti/dre.pdf

That was Ken Goldberg's (amongs others, of course)project, The Telegarden. It finally went offline in August of 2004. The physical garden started out at the University of Southern California, and was then moved to the Ars Electronica Center in Austria.

That was Ken Goldberg's (amongs others, of course)project, The Telegarden. It finally went offline in August of 2004. The physical garden started out at the University of Southern California, and was then moved to the Ars Electronica Center in Austria.

For those of you who are interested in what the patent applications are about but do not feel like sifting through them in search of actual information, I posted an analysis of the "Interdiction..." patent on the "Challenge of P2P" blog.

As an aside, and as discussed in the blog post to which I refer, what they characterize as hash spoofing does not require to break any cryptographic primitives. It merely refers to a form of false advertising.

And the obligatory response Software Fault Prevention by Language Choice: Why C is Not my Favorite Language (PDF).

Actually, to take the planks out of my own eye first, I probably ought to shut down the PC at 5:00p myself. (I'm at work) :-) The Macs at home (should) automatically go to sleep, though they haven't lately...

1. Genetic engineering: Gene therapy is currently a very promising field of study, and research on vectors is finally yielding some extremely promising results, both for viral (see some of the fourth generation or higher lentiviral systems) and non-viral (liposomes etc). As gene therapy becomes common, the same techniques can be applied to not just fix genes, but add or alter existing ones to give desirable attributes (better vision, etc).
2. Brain-computer interfaces: Once again, most current research takes place with the aim to provide superior prosthetics to people who have suffered from accidents. This is my personal area of interest. In principle, all the input and output going into the brain should be able to be intercepted and controlled. By doing that, a person could be transplanted into any artificial body desired. I feel that at the current pace of development, this will be a relatively (there is always risk with surgery) safe and well understood procedure within 20-30 years, assuming research isn't outlawed or anything like that.
3. Medical nanotechnology: Very speculative, I don't think anyone knows for sure whether is can actually be done or not. I'm listing it because it would be a different way to augment the human body from the previous two.

Some links:

University of California Neuroelectric Research Group. Some interesting information, with PDFs available, on BCI.

Gene Delivery Systems. A free quick intro (from a lecture/course) on some of the different vector systems being studied for gene therapy, and desirable characteristics.

Those of you with access to journals can go read a very interesting study published in the Journal of Cognitive Neuroscience 16(6):1022-1035. "Optimizing a Linear Algorithm for Real-Time Robotic Control using Chronic Cortical Ensemble Recordings in Monkeys," by Wessberg and Nicolelis.

http://emlab.berkeley.edu/~moretti/dre.pdf

l2718 writes "Two economists have just posted a paper online

Why not link to the actual paper, rather than a pay-to-view site?

http://emlab.berkeley.edu/~moretti/dre.pdf

Amazing. And they have internet connections there? I would guess that the latency would be astronomical. What I think would be even more ironic is if you ran Seti@home there. Do you?

Today, most basic and applied researchers are effectively standing on top of a huge pyramid, not just on one set of shoulders. Of course, a pyramid can rise to far greater heights than could any one person, especially if the foundation is strong and broad. But what happens if, in order to scale the pyramid and place a new block on the top, a researcher must gain the permission of each person who previously placed a block in the pyramid, perhaps paying a royalty or tax to gain such permission? Would this system of intellectual property rights slow down the construction of the pyramid or limit its height?

No problem. :) Have you seen "The 11 evolutionary stages of a programmer", BTW? The C++ examples in there don't need any comments, I think. :)

1. Brilliant researcher at Oxbridge University (who pays his salary) comes up with great idea, writes it up, submits it electronically by emailing it to an editor at the Snooty Journal,
   
2. The editor, a professor at Enormus State University (who pays his salary and has him teach a little less because of his prestigous editorship) thinks of an appropriate anonymous referee and sends off the article to be refereed. Snooty Journal may give ESU some money to cover part of the cost of a secretary, but does not pay his salary.
   
3. Professor at IviedHalls University (who pays his salary) receives the article to refereee, reads it, sends it back with comments after letting it molder on his desk/inbox for a bit.
   
4. Editor accepts or rejects the paper, possibly asking for modifications based upon the referee's recommendation, possibly some iteration at this step
   
5. Original author prepares the article in electronic format using LaTeX with Snooty Journal's style files and uploads it to their web site.
   
6. Snooty Journal staff typeset the paper, messing a few things up because they are not experts in the appropriate field, and send the "galley proofs" to the author to review.
   
7. Original author points out typos introduced in their typsetting process, sends back corrected galleys.
   
8. Snooty Jounal releases the article on their paid-subscription webpage and prints it as a dead-tree volume to send to libraries around the world that can afford it.
   
   As you can see, the hard part of the labor (writing, reviewing, refereeing) is not done by anyone at the publisher-- various universities pay the salaries of those folks and they pay again for the journal in dead-tree form.
   

   
   So you can see that there may be some objection to the arrangement. In the old days, the journal staff actually typset things and dead-trees were the only game in town, but most of the typesetting is done by the author.
   

   
   The choice is hard for some people that really need to publish in the expensive journals to get tenure, recognition, grants, etc. But for people who already have tenure, some are resistant to the journal extortion. Some may have a policy like mine- I do not submit to expensive journals or agree to referee for expensive journals, now that I have the advantage of tenure.
   

   
   There have been some successes of editorial boards that resigned wholesale, then started a free/inexpensive journal. Hopefully this becomes more common.
   

   • Some related links:
   • Journal prices in Econ
     
   • Rob Kirby calling for better pricing for math journals
     
   • Journal of Algorithms editorial board resignation to start a cheap prestigious journal
     
   • Knuth's letter about that resignation
     

forgot the obligatory plug for the Berkeley CSUA

here is a link to a mp3 of the talk if anyone cares.

oh and go CSUA!

more interesting to me, it could be used as a very high-bandwidth connection between a computer and me.

This has already been done several times, both at a low-bandwidth level (electrodes on the skull, done several years ago), and a high-bandwidth level (implanting an electrode directly in the neocortex, done in 2000)

If you're interested in this stuff, you should check out this journal article - PDF Reprint

Kennedy PR, Bakay RAE, Moore MM, Adams K, Goldwaithe J. 2000. Direct control of a computer from the human central nervous system. IEEE Trans. Rehabil. Eng. 8:198-202

Here's the abstract, if you don't want to wade through the PDF:

We describe an invasive alternative to externally applied brain-computer interface (BCI) devices. This system requires implantation of a special electrode into the outer layers of the human neocortex. The recorded signals are transmitted to a nearby receiver and processed to drive a cursor on a computer monitor in front of the patient. Our present patient has learned to control the cursor for the production of synthetic speech and typing.

None of us thought it could happen here ???

One clue .. http://www.msnbc.msn.com/id/4595173/

http://www.berkeley.edu/news/media/releases/2004/0 9/08_clarke.shtml

oh.. wait - he's just a "disgruntled employee" http://electromagnet.us/dogspot/modules.php?name=N ews&file=article&sid=67

Does that mean this event didn't happen in the US? http://news.bbc.co.uk/onthisday/hi/dates/stories/f ebruary/26/newsid_2516000/2516469.stm

Or this one? http://www.virtualboricua.org/Docs/hs01.htm

Are the groups identified here not terrorists?
http://www.latinamericanstudies.org/puertorico/dan iel-james.htm

How about the KKK or any group who has the word Aryan in their name?

Terrorism has been occuring in this country since its founding. The fear-mongering regarding terrorism is much more dangerous to the stability of this nation.

I just finished a course on this particular subject actually. A few fun comments.

In one biological study in Maine (Great Duck Island) it seems that the birds on the island they were monitoring had been attacking the sensor motes. In another case these devices offered the first look ever at night time migration patterns of zebras.(aka ZebraNet)

As far as military applications go the one that I am most aware of is DARPA's sniper net. It's a system of audio sensors designed to locate and pinpoint snipers based on gunshot triangulation.

There are some earthquake structural monitoring systems being built in California as well.

However things to be cautioned about. The smaller motes do not have very much in the way of processing power(ie can't even do floating point) so there's no need to get really paranoid about secret cameras. Most of the motes with cameras are big enough you'd probably notice them if you were looking. Primarily the motes are equipped with various sensor banks for things such as Light, Temp, Vibration, Audio, etc etc. Also if you're interested in working with the software for these things the primary OS people use is TinyOS. However a word of caution, if you want to muck around inside the inner workings of TinyOS you're pretty much on your own and some of the things are already legacy. The coolest part of sensor nets, in my opinion,is the ability to do in network data processing as the data is funneled through the network. Oh and there's already a Database system designed for use in these systems. It's name is TinyDB(surprising naming scheme I know)

Cheers

Yet again, investigation reveals that the primary cause of the incident is pilot error.

The thing is that the amount of devices and money put into this would make the vehicle well out of range for the average customer.

That's the main reason why automated vehicles/highway's aren't all ready in use. I work in the field of vehicle automation, and many of the major hurdles regarding the functionality have been crossed. One good website to go to is here.

To give you an idea, a fully instrumented vehicle capable of doing autonomous driving costs about $100,000. Similar to one of those new-fangled fuel cell cars. This doesn't include the cost of infrastructure, as you can't exactly equip all civilian vehicles with military grade GPS. The California PATH program actually uses cow magnets embedded in the highway (VERY expensive to do).

By the way, the main reason for automating highways is so that you can fit A LOT more cars on the road, and optimise the vehicle positions based on who is getting on and off where, and you can avoid traffic hold-ups from things like somebody tapping their brakes a mile up the road (called the "Slinky effect"). They found that the overwhelming majority of people would rather have the ability to own a car and use it whenever they want to on an automated highway rather than develop a really good public transportation system.

As for accident avoidance, many of the automated highway people (such as myself) are working on accident avoidance, since that is cruicial to automated highways being able to function. I'm not really sure how MS factors into my research, but I don't think that it will be good. Besides, their software will require a great deal of additional hardware, such as steer-by-wire, brake-by-wire, etc. (all ready well under development and will be in stores by around 2010). But I digress...

Subs generally utilize light water reactors instead of the more difficult to maintain liquid metal reactors.. Here is a page that gives a quick description of early Liquid Cooled reactors.

More links:

http://www.nuc.berkeley.edu/~gav/almr/01.intro.htm l
http://www.nuc.berkeley.edu/thyd/ne161/shir/projec t5.html
http://www.nucleartourist.com/type/metal.htm

Subs generally utilize light water reactors instead of the more difficult to maintain liquid metal reactors.. Here is a page that gives a quick description of early Liquid Cooled reactors.

More links:

http://www.nuc.berkeley.edu/~gav/almr/01.intro.htm l
http://www.nuc.berkeley.edu/thyd/ne161/shir/projec t5.html
http://www.nucleartourist.com/type/metal.htm

The second problem is that keeping water -- essentially any water, to 3 or 4 places -- out of the cooling system will require a higher level of (read: more expensive) engineering. The problem is that small amounts of water will lead to ionic compounds like NaOH, Na2O, etc. ... all of which will "slag" the sodium and prevent it from flowing or transferring heat efficiently.

The third problem is containing the molten sodium in a container which will not spontaneously alloy with the hot sodium.

Check out decent reactor cooling discussion here (note the "fuel can be bonded to container with liquid metal" line!)

Some of the other metals discussed here -- Wood's metal, mercury, etc., have varying levels of these problems. For example, a friend of mine got an unwelcome "addition" to her silver ring when she mishandled some mercury in a chem demo.

With all due respect, that is because you really don't understand the fossil record. Transitionary forms, or "links between the kinds" as you put it, are common in the fossil record. So, to respond to your statement "This pretty much rules out reptiles evolving into mammals, etc, unless you can find a way to reconsile that" here are some cut and dry references that reconcile this non-dilema:

http://www.talkorigins.org/faqs/faq-transitional/p art1b.html#mammTransition from synapsid reptiles to mammals

http://www.ucmp.berkeley.edu/diapsids/birds/archae opteryx.htmlArchaeopteryx lithographica

http://www.asa3.org/ASA/resources/Miller.htmlTaxon omy, Transitional Forms, and the Fossil Record

Fossils are also VERY rare, which explains that even though its hard to find fossils (very hard) and yet transistionary forms are COMMON in the fossil record. Case closed on this, they are all over the place and cover all the "kinds".

http://scienceviews.com/dinosaurs/fossilformation. htmlThe Formation of Fossils

So, you on board now?

Wasn't this paper co-authored with Joe Darcy .. now the Java floating point czar working on Tiger (Java 1.5)?

Much shorter version of the paper is here, and a good java floating point paper is also over here

oh .. and if you think that nobody at sun will admit java's weaknesses .. you gotta stop talking to the sales and marketing drones, and spend some time in targeted discussions with the engineers ..

If people wanted FORTRAN replaced, it would have been done so. Certainly, it has been tried.

http://titanium.cs.berkeley.edu/

Real programmers abuse the EQUIVALENCE statement.
Real programmers don't need structured code.
Real programmers write self-modifying code.
Real programmers use FORTRAN!

I thought Mathematica was the successor to Fortran.

Mathematica is (hopefully) mostly used for symbolic computations. In numeric computing, MATLAB and its extensions is quite popular (maybe even GNU Octave for those who rightly fear that proprietary software undermines freedom of research). I have no idea why the folks at Sun think that Fortran is their competitor. Maybe MATLAB suffers from a stigma similar to Visual Basic. Certainly someone inside Sun knows that their HPC customers frequently run MATLAB programs on Sun hardware.

Why don't they just improve the Mathematica calc engine for parallel/distributed supercomputing?

Why would they want to improve the product of a competitor on a government grant? Sounds like a stupid plan to me from a business perspective.

Anyway, language design suitable for numeric computing is not Sun's strength.

Fortress development team, hopes that Fortress will to 'do for Fortran what Java did for C.' Steele admits that Java isn't probably the best choice for numerical computing

So they finally admit that what Java did was break the IEEE floating-point specification, that was correct in C, as Professor William Kahan, of Berkeley (see How JAVA's Floating-Point Hurts Everyone Everywhere), had been shouting to deaf ears all this time?