Slashdot Mirror

PS: I should have mentioned an example when I wrote this post. See problem 2 in this homework solution from a numerical analysis class. -- Paul

What amazes me is how many tools are out and available online regarding this sort of pattern recognition development. Since a lot of people know Java, I'm would encourage you to use the Java Media Framework (free from Sun). Once you have those libraries installed, it's quite easy to start editing sound, images & video. You might need to grab and install codecs if you're doing video analysis but I think almost all image codecs are supported.

I'm not going to lie, the video computation can be quite heavily but thankfully that framework is implemented such that the entire video doesn't have to be loaded into memory, just a one frame buffer analysis can be used if you want.

The last thing you would need is simply the know-how on programming these analysis algorithms. There are sites out there with a large wealth of up-to-date algorithms. An example would be the text book style site of pattern recognition or image processing. While this doesn't teach you how to do things, it does contain the raw resources and algorithms. General resources like the computer vision homepage exist that serve as links to all kinds of resources. Unfortunately, I know of no real solid books that contain everything out there because this field is so rapidly developing. My professors taught me from hand printed slides in a large compendium they had accumulated over the last couple years.

The last piece missing is the data to analyze. While you might not have the ultra high resolution Van Gogh images to do this yourself, it may be possible to visit museums with 6 MP cameras to obtain your own data. Failing that, there are repositories online that sometimes contain image information you can start with. While this may not satisfy your specific needs, it sure is great for the lazy developer like myself.

Lastly, I will mention citeseer and Google Scholar for cutting edge papers that you might want to try implementing. Distributing these algorithms and building a good GUI can be tricky but really anyone can build the backend. I heavily recommend experimenting with this if it interests you.

Of course, musical accompaniment systems have been around since the 80's and earlier. It's been a research topic in artificial intelligence nearly since its inception!

Robert Rowe published a book on the topic in 1992, on his Cypher system. Here's another good article on the topic by Chris Dobrian. For an open-source system, check out Bob van der Poel's MMA.

Additionally, there's been plenty of work done on robotics for playing instruments, particularly for percussion.

So, admittedly, this is the first time I've personally heard of a project combining the two, so I'll give it that credit for innovation. But I'd be sorta surprised if it hasn't been done previously. When you think about it... all these musical accompaniment systems react in real-time to MIDI input. Simply make a couple of motors respond to MMA's ouptut, for example, adjust timing according to latency and inertia, and you could probably have this project done in a few days.

Not to play it down, I always love to see fun projects like this.. :) And the physical design is quite beautiful for a drum-playing robot.
But "first".. well, give credit where it's due. I think the summary is over-reacting. (I scanned the article.. don't believe it makes any such claims.)

Its about scale. Web-scale to be exact. RESTful architectures (like HTTP) are proven on that scale, and pretty much nothing else is. VNC and the like are nice and have their uses, but you don't want to build an application for the web in general with them. You shouldn't be thinking about bandwidth, but rather about things like maintaining state for thousands of simultanious users, naming of third party resources and how to make your application benefit from caches.

The REST article on wikipedia is rather terse, but can be used as a starting point. I personally found Roy Fielding's dissertation (http://www.ics.uci.edu/~fielding/pubs/dissertatio n/top.htm) pretty readable and convincing.

How is this news. This all sounded really familiar and old, and of course, the first site I went to on a google search took me to this from 2000. There's tons of stuff from 2004 on studies done with pidgeons. But this stuff is definitely not new.

Totally by accident, I attended a major Physics conference (in preparation for the CERN LHC/Large Hadron Collider coming online next year, very exciting!) & did some tests with New Mediums : video-podcast, Sony PSP, LiveWebCast over a mobile-blog. This was done with the approval of the Program Chair, who's a young UC Irvine physics professor who understands the value of Technology.

Through a major Physics blog, a USC physics prof (string theorist) mentioned the SUSY '06 conference (4th International Conference on Supersymmetry and the Unification of Fundamental Interactions). I contacted the Program Chair, & he invited me down to do some "New Medium" tests:

http://www.jumplive.com/susy06/index.html

[ incidentally, that USC prof had a meeting last year with other profs to discuss the new "blogging technologies". There is the USC Annenberg Center, which addresses technology & communication. So, USC is "with it" ]

I recorded lectures, plenary-sessions on HD (high definition) video & other video devices (digital cameras w/video capability). I put them up on a video-blog (& its corresponding video-podcast over iTunes Music Store, just do a search on "SUSY")

http://susy06.blogspot.com/

Some of these videos are really LONG, like 240mb. I also delivered them over a Sony PSP (another big-market portable video-player, 12 million out there). Some of the videos were delivered on site, within 15 minutes of taping..near-live as iTunes video-clips. There are some QTVR panoramas of some conference events. There was a LIVE delivery of pics/videos at a Textamerica.com mobile-blog:

http://susy06.textamerica.com/

[ there are some video interviews, & some hi-res pics of talk presenters ]

There were 2 Nobel Laureates in attendance (Burton Richter/Stanford & F. Wilczek/MIT), & many big names from the world of theoretical/experimental particle-physics. Some of them were on that NOVA episode on String Theory (Brian Greene/Columbia host). Being an Elec Eng PhD, it was exciting to experience a technical conference in another field. I was given recognition at the conference, & links from their website here

[ I am currently looking for a business-entity to take my "Proof of Concept", & deliver this to next year's SUSY '07 Conference in Karlsruhe/Germany. And, to ALL technical engineering/science conferences. Please contact me ]

The purpose was like that of my target Market ("Offroad Racing", see http://www.jumplive.com/

"A better informed Public is more likely to appreciate/understand, & therefore publicly fund Science"

Physics (& Science in general), like Offroad Racing, suffers from an image problem. It's a niche-market, & the general public just isn't aware of their "activity/events". As the result, they both suffer from Funding issues (in racing, it's known as "Sponsorship"). Offroad Racing has been termed "Our Little Wonder in the Desert". Similary, Science could be termed "Our Little Wonder in a World of Idiots". You may recall the SSC (Superconducting Super Collider) that was cancelled in the 80's, which was a major blow to US program in particle-physics. There was NOT a public outcry, like you see now of the HST (Hubble Space Telescope) being de-commissioned. If the Public really understood/appreciated particle-physics, perhaps the SSC could have been resurrected. Science really is getting the "shaft" in USA, & I think the Slashdot crowd is concerned about this.

I realized halfway thru my project, that these lectures over video-iPod could have value as a Research Tool. The conference attendees could re-view the lectures, especially the Plenary sessions. I even talked to a Harva

Their research seems to deal mostly with the third problem, which is one of the biggest barriers to use in real life. Many of the algorithms used on these types of problems are NP, or require ridiculous amounts of (expensive) labeled data to train from. Also there are problems with generalization and overfitting.

There is no freeware software that can compete with this type of algorithm under these conditions - over 300,000 articles in just a few hours.

All that being said, I'm waiting for the paper, along with more technical specifics, to be released so I can really see what this is about

Are you sure? Have you read the paper, or just the over-simplified press release? Here is the paper: http://psiexp.ss.uci.edu/research/papers/i si2006.pdf

http://psiexp.ss.uci.edu/research/papers/isi2006.p df

While that's certainly LDA code, it's actually from a lab different from the one discussed in the story, and I think they use some slightly different techniques. For topic-modeling code from Mark Steyvers' lab, who produced the paper in question, here's the link:

Matlab Topic Modeling Toolbox

1. Defining the problem.
2. Getting an accurate result.
3. Getting it as fast as possible.

• basically, the model assigns a probability distribution over topics to each document
  i.e., documents aren't assigned to a single topic (as in latent semantic analysis (LSA))
• topics are learned from the documents automatically, not pre-defined
  this means, incidentally, that they're not automatically labeled, although a list of the top 5 words for a topic generally characterizes it pretty well.
• the technique can learn which authors are likely to have written various pieces of a given document, or which cited documents are likely to have contributed most to this document
  side benefit: you can also discover misattributions (e.g., authors with the same name)

• basically, the model assigns a probability distribution over topics to each document
  i.e., documents aren't assigned to a single topic (as in latent semantic analysis (LSA))
• topics are learned from the documents automatically, not pre-defined
  this means, incidentally, that they're not automatically labeled, although a list of the top 5 words for a topic generally characterizes it pretty well.
• the technique can learn which authors are likely to have written various pieces of a given document, or which cited documents are likely to have contributed most to this document
  side benefit: you can also discover misattributions (e.g., authors with the same name)

Actually that happens all the time in nature; there were tons of mass extinctions when the isthsmus of Panama formed. All of a sudden there was a huge exchange of animals between the two continents; particularly from North America to South America, the latter of which was isolated since the breakup of Gondwanaland.

From a relevant link:

The Invasion of South America. In the upper Pliocene, 3 million years ago, the isthmus of Panama reappeared as a result of changes in the earth's crust. This was a disaster for many of the animals that had evolved in isolation in South America. South America was invaded by deer, camels, raccoons, tapirs, horses, mastodons, bears, peccaries, rabbits, shrews, cats, dogs, weasels and rodents. For some reason these animals were able to displace many of the South American species, driving many of them to extinction.

Some of the new arrivals (e.g. mastodons and horses) survived only for a brief period. Others were very successful, for example the camel family which has given rise to the vicunas, guanacos, alpacas and llamas. The camels as well as the horses subsequently became extinct in North America where they originated.

I think you're right. The install I saw that was the very closest to what was postulated further up the chain was one where the sector antennas were on the corners of a building, and the heliax ran from the center of the rooftop like spokes on a wheel out to the 4 clusters of antennas. The rest had the cabinets either at the base of the tower, whether it was on the dirt or on top of a building, or the sector groupings had their own cabinets, and the heliax runs were measured in single digits of feet (well, ok, there were only 2 like that). Then there were the water tower installations, the tubular tower installations, and the other oddballs where the antenna runs were encased completely in a grounded metal tube, save a few feet on either end to connect to the racks or the antennas.

UCI agrees with us, by the way. Sector antenna sites are very safe, and the antennas used have incredibly good pattern control.

Sometime back I happened on this article by Gregory Benford about how we can communicate across tens of thousands of years to our descendants in the form of a warning ... as in don't open this up.

I particularly like the suggestion near the end about making the site scary so that it would become a place that was taboo just because it was so creepy. His suggestion was shaped stone monoliths that would wail in the wind. Anyway worth reading.

Go read "DEEP TIME", this subject is nothing new...

http://www.physics.uci.edu/~silverma/benford.html

Skulls and bones were actually discussed in the original WIPP study. They were rejected as being culturally defined, ie today's usage on medicine bottles being unrelated to the original (pirate) meaning. Generally, they assumed that while human beings would not change much physically in 10'000 years, almost no abstraction would survive (stay understandable) that long.

The proposal was to replace artificial symbols with pictorials. Disgusted faces, stick people displaying sickness, with symbols which tie sickness to the actual monument.

Related link about how the monument was planned (long but interesting reading):
http://www.physics.uci.edu/~silverma/benford.html
The process involved all kind of scientists, including a physicist/science-fiction author.

The original WIPP report is also available somewhere as a PDF (sorry, no link). They accepted two or three alternative solutions for everything, from physical layout to warning labels.

Hmmm, it's an interesting idea, but it seems to give a lot of false positives. (So naturally, it will detect fake papers, if it thinks every paper is fake.)

First thing I tried was some pages on computational oncology website, in particular, my cancer primer, which I wrote in not a short time. Everything I fed was determined to be inauthentic. Perhaps I just write like a robot. :-) I figured that perhaps the detector was more primed for real papers, so I figured it wasn't too big of a deal.

So, next I tried my most recent research paper, and it, too, was determined to be inauthentic, and in fact with less authenticity than my website. So much for the theory of being primed for scientific papers only. This thing is starting to look pretty bogus to me ... but an interesting idea, nonetheless. -- Paul

Hmmm, it's an interesting idea, but it seems to give a lot of false positives. (So naturally, it will detect fake papers, if it thinks every paper is fake.)

First thing I tried was some pages on computational oncology website, in particular, my cancer primer, which I wrote in not a short time. Everything I fed was determined to be inauthentic. Perhaps I just write like a robot. :-) I figured that perhaps the detector was more primed for real papers, so I figured it wasn't too big of a deal.

So, next I tried my most recent research paper, and it, too, was determined to be inauthentic, and in fact with less authenticity than my website. So much for the theory of being primed for scientific papers only. This thing is starting to look pretty bogus to me ... but an interesting idea, nonetheless. -- Paul

pine schmine. real geeks use nmh. for an example of why, see: http://www.ics.uci.edu/~mh/book/mh/finpic.htm

There is a large bit of hand waving here. Why are neutrino oscillations and neutrino mass inseparable?

I hate when people act as if a complicated issue is simply true. So, as a public service to the Slashdot community:

Here is a site that attempts to explain it.

My quantum physics knowledge isn't teriffic. Any particle physicists know of a better source?

nonzero rest mass for photons, neutrinos and other things previously considered massless are being explored in theory and experiment. Relativistic energy also is a gravitational charge source.

Non-zero rest mass for neutrinos is a near certainty. Non-zero rest mass for the photon is not considered a serious possiblity. I do not know the gravitational effects of moving a moving mass compared to one at relative rest. If gravity inceased with speed then a fast enough moving object would inevitably collapse to a black hole as it approached the speed of light, no matter the mass.

Kind of - think about it: 1 + 2 = 3. The mouse travelling at the speed of a horse under its control will always outrun a sleepy-cat. It's the only option the mouse had looking at the circumstances, and given Oracle must be the mouse in the story, it's in the winning situation.

But they don't report that. From the article, I quote: "Will it work? It is too soon to say."

No difference was seen in the frequency or intensity of flashback episodes.

Now, to pick a nit, it is incorrect to say that no difference was seen in the frequency of flashback episodes. The article clearly states there was a difference, but that the difference was not statistically signifcant. Given the very low number of participants in the study, it would be relatively surprising to have seen a statistical difference, those folks need to do a signficantly larger study to make any sense of it. I agree with you that, by itself, this is not cause for optimism, but nor is it a direct cause for pessimism. In light of the authors previous studies, I think it's still an interesting and possibly productive line of research.

Blah blah blah Bremsstrahlung losses, yada yada. Go read the Rider's paper, Todd Rider's, "Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium". Here, have a link.

Aneutronic fusion is "impossible". 3He is silly. I just see little reason to go fusion at all if its going to be heavily neutronic, significantly more neutronic than fision. Just built some molten-salt fision reactors and start burning the nuclear waste we've already got. I dont think we'll run out anytime soon.

I'm obviously not a nuclear engineer, but I got my aneutronic fusion hopes basically dashed. I still havent seen anything to give me hope.

I thought we already knew this - that bees fly because little ridges on their wing roots act as vortex generators, breaking up the airflow above the wing? This design technique works for whales as well - at least for swimming in water. They have tubercles on their leading fin surfaces.

Hardware-software codesign is nothing new and revolutionary. It has been taught for years at berkeley and around the country. A bunch of links can also be found here

The numbers: 1188 Macs, 3453 Windows, 948 Linux *, 545 Unixes
* 530 of the 948 Linux machines are dual boot with Windows. Over 90% of users choose to boot with Windows versus Linux
** 600 machines are unknown exotic OSes floating on campus. BSDs, QNX, student brewed OS, honeypots...etc.
***1000 laptops unregistered with depts floating around with students and researchers which only increases the number on Windows and Macs.

You want high fidelity? I got your fidelity right here World's Highest-Resolution Display Wall and 200 Mega Pixel HiPerWall

I've never seen someone fellate their boss more than in that interview. It was like watching the whole thing unfold through a bad Scooter Libby novel, only without the fiction.

Personally, I find this workflow* to be a more likely scenario for how the recent Star Wars were made.

* the link is correct, image name includes original site, and posting date of image, for reference.

The last part of that sentence indeed summarizes the chief obstacle to longevity of any monument.

Incidentally, this is not the first time that such a time-scale has been deliberately studied. A while ago the U.S. Dept. of Energy actually commissioned a study into the problem of marking a long-term nuclear waste repository (WIPP in New Mexico, Yucca Mountain if it ever opens) so as to prevent unintentional intrusion and possible spread of contamination.

Physicist and SF author Gregory Benford was on the team, and his account appears as the first chapter of his book, Deep Time. The book is, it seems, out of print, but still available on Amazon. There is a slightly garbled copy of that chapter online, minus the cool illustrations of several marker concepts. Some illustrations appear in the excerpted report of the WIPP Marker Panel. Fascinating and slightly unsettling stuff.

There have been many studies, but like geese, fish also take advantage of swimming in groups:

Scientists show how fish save energy by swimming in schools

And there have been many studies into dolphin and whale motion.

Hydrodynamic study into whale flippers

The use of rotary motors only seems to occur at the bacterial level (flagella bacteria)

would not surprise me to learn that a campus counsel or some such wouldn't let a library give away rights to content that UC held the rights to (like a library's special collections holdings)

So in other words, the public domain is locked away. The PD consists of OLD books, which are largely in special collections.

Here's some policies I digged up. It's worse than the policy though. They say write a letter explaining your needs and they ignore you.

• UC Irvine: royalty fees required for PD books and other nonsense
• UC San Diego: royalty fees required for PD books and other nonsense
• UC San Diego: 20-page limit policy and no scanning policy
• Can't find UCLA's policies online, but they also prohibit scanning and wanted a $200/page royality!

Let's see. You've got your power/energy units mixed up. Watts is the power unit, and to qualiy that in terms of energy used, you need to know time. The convinent unit for our application is the kilo-Watt-hour (kWh), which is just an average of the power applied over an hour. No big deal, but let's try to see what's really going on, because I think your average Watt usage is about an order of magnitude off, just guessing. Personally, I have a mini-machine shop complete with 200 Amp 480 service, welders, compressors, plasma cutters, lights that turn night into day, enough 120V to run the entire block if I wanted and all sorts of other good stuff that I use often, and I don't come close to using 20kW on a consistent basis, or basically ever. I can only use one machine at a time, afterall. Unless I splurge on some good CNC equipment, anyway :D

From UCI, they say that the average home in 1999 used 866kWh/month, it probably hasn't changed drastically. If we call a month 30 days, that's 866/kWh/720 hours. Hours cancel out, we get 866/720kW, or about 1.2kW average over 70 days. With AC and fridges, that seems entirely reasonable, if a bit low. Also, computers rarely used the full rated output of the power supply, so if you're a geek with a bunch, you've got to take that into consideration.

I've had a hard time with finding exactly how much energy is contained in an average lightning bolt, I must admit. I've seen anything from 5,000 Amps at 2,000,000 Volts (which sounds reasonable) over 200 miliseconds to a hundred or a thousand times that (which dosen't very sound reasonable) Watts=Volts*Amps, so my reasonable sounding lightning strike will discharge 10 GigaWatts over 200ms. With 3.6 million miliseconds in an hour and a bit of division, it looks like our bolt will do 2777kWh if entirely captured, which is enough to run our average house for 3 months and some change, not bad. Shame there's no way to capture it.

Truthfully, I have no idea how close that figure is to an average lightning bolt, it seems that most of the numbers out there people just pulled out of the air. I could be off very far either direction, and likely am, for all I know. Nature is pretty fantastic, though, isn't it?

If that were true, then the same would be true of CO2. Yet environmentalists tell us that we're having a significant impact on the CO2 ratios. And scientists are already warning about the effects of a sudden increase in vapors:

"The widespread use of hydrogen fuel cells ... would cause stratospheric cooling, enhancement of the heterogeneous chemistry that destroys ozone, an increase in noctilucent clouds, and changes in tropospheric (lower-atmosphere) chemistry and atmosphere-biosphere interactions," scientists from Caltech and Jet Propulsion Laboratory in Pasadena proposed in the journal Science in 2003. Noctilucent clouds are eerie high-altitude clouds whose abundance, some scientists suspect, is influenced by climate change.

See? You can't have it both ways.

BTW, there are no oceans in Arizona, Colorodo, Kentucky...

At an already 600fpm and wanting to improve, am I the only on thinking that Wal-mart should have designed the Denver International Airport Baggage System?

The proton-boron method using a laser reminds me of colliding beam fusion, which I first heard about in 1997. Interesting thing here is that energy capture occurs electromagnetically using a "decelerator." Read about it at:

http://fusion.ps.uci.edu/beam/introb.html
http://www.sciencemag.org/cgi/content/full/278/534 2/1419?ijkey=A.zNwOzIwyrKA
http://www.sciencemag.org/cgi/content/full/281/537 5/307a
http://www.stormingmedia.us/01/0116/A011653.html

I did find this...

http://today.uci.edu/news/release_detail.asp?key=1 261

about the UCI study.

I should have posted this to begin with, but I was at work and did not have my bookmarks.

Coincidentally, I was working on an article about women in the technical field (I am a Network Tec and Data tec) and I happened to run across the UCI link this afternoon.

I actually found it quite interesting.

However, as with all open source software, caveat emptor. We ourselves (University of California, Irvine's TeamXAR - I'm the team leader) found that CARMEN does not fail well, and in fact had to give it a few kicks each time we started it up.

Note also that the finalists (and possibly semi-finalists) will need to issue team technical reports, so while that probably won't include code, key details will emerge.

I should also state before I get a flurry of questions that we ourselves did not make the cut for the 40 semi-finalists for the September 2005 National Qualifying Event.

We had built a ton of software since our team's inception in July 2004, but did not touch a real car we got a message from DARPA effectively stating "Hi! We'll be there in about a month for a site visit. See you then!"

As to be expected, as we (a core group of 6 undergraduate students plus a few extra helpers) pulled nearly a month-straight of all-nighters to get something to show the reviewers from DARPA, not only was Murphy's Law present at our site visit, but we learned Murphy's Corollary: Murphy likes to work overtime. Nothing worked.

Three weeks later, the car was tooling around our site visit course, but of course, three weeks too late.

If you're interested in following our progress, hop to our website at http://www.ics.uci.edu/~darpagc/ or http://www.teamxar.com/

- Phil

Anyone who has ever read 1984 knows that this is one of the hallmarks of a controlled society.
As soon as a book can (untraceably) be edited much objectivity is lost.

This is already happening, and it is indeed scary.
They just don't quite have the untraceable part down yet.

About a decade ago, Time Magazine published an essay by Bush Sr and Secretary of Defense Scowcroft on why they chose not take out Sadam during the first gulf war. A lot of the points they made have been proven true today.

Time DELETED the article from their online archives. It was as if it were never written, URLs that once worked are now road-kill on the information super-highway. Not only that, but significant changes were made to other articles in that same issues as compared to the print version.

Fortunately it wasn't quite so untraceable and has been widely reported (not widely enough IMNHO). Here is one take on the story, you can find plenty more by googling for bush scowcroft "reasons not to invade".

http://eee.uci.edu/programs/comp/39c/google/hesket h.html

After reading your post and staring blankly at the screen for a few moments in disbelief, I can only come to the conclusion that:

A) You're a troll
B) You're drunk
or
C) You work with a bunch of saboteurs that intentionally crash their Treos to get paid downtime.

I've had my Treo 600 for about 2 years, now, and have experienced none of the problems you've described. The GP's description of no more than 1 crash per month is very accurate. Anytime it has crashed, it quickly boots right back up. I've never had to send it in to be serviced.

For those that think that a Palm is just an "organizer" and a PokcetPC is a "pocket computer", don't buy into stereotypes. I use my Treo as a computer. I have an ssh client installed that I use frequently to work on some servers I admin for. The thing came with a capable web browser, but I have many options to install something else, if I want. I also have a Samba client that works great, an FTP client, a VNC Client, and an Instant Messenger. Somebody already mentioned the superb movie player TCPMP, but that's not all, I also have a Video recorder that makes use of the Treo's built-in digital camera. I use a perl script I found to convert the video to mpeg1. I use a Photoshop-like image editor that has support for complex things like layers and blending modes. My Treo is also my mp3/ogg player and I use it to listen to podcasts in the car. I read ebooks and even /. using Plucker. I take audio notes using SoundRec. I even have a Python interpreter, and can code native apps in C right on my Palm. I won't even bother to mention all the games that are available. You can google for them yourself. I've seen apps out there for viewing/editing Word Docs and Excel files, but having never had a need for that, haven't installed them.

As for the choice of Quicksort - Most likely, they chose it because just about every C library out there has an implementation of quicksort. And while personally I prefer heapsort (in the worst case, quicksort has Q*O(n^2) behavior, while heapsort always takes only P*O(n log n), But P >> Q), I'll admit that for almost all unstructured input sets, quicksort finishes quite a lot faster than anything else.

Choosing to quote only references that support your thesis, given that there are good arguments against it, is quite dishonest. This criticizes the 'proof' you mention, showing that it is not nearly as general as you would have us believe.

Also see page 30 of Dell feb 2000 migration

If Apple was *really* smart, they would have perfected some way for executables to run on any processor (they had 10 years to do it since switching to PPC), which could have been easily deployed by now. Then the switch to Intel could be nearly transparent from both the user and the developer's perspective. And they could continue to use PPC or Intel or whatever, depending on which worked best for the target market.

It could have been their answer to .net.

But no, they have to be locked into one processor. And thus alienating most of their current market.

I really dont understand why we are still locked into any particular architecture anymore.

It's the kind of thing that makes me want to pack my bags and move to a parallel universe where things actualy make sense...

There is one method (that I know of) which is capable of direct conversion of nearly all the energy of the fusion plasma into electricity. It is not possible in inertial confinement laser fusion though. It is the P + B-11 reaction. This involves the fusion of protons (normal hydrogen) and Boron 11 in a magnetically confined fusion reactor called a colliding beam fusion reactor. There is nearly zero neutronicity in this reaction (.001) and the product is 3 He atoms. In this scheme there is sort of a scrape off layer at an end of the reactor where the product He ions are directed into a magnetron cavity where they are decelerated and produce microwaves, the microwaves are collected with an antenna and rectified to DC power. Also they use something called an "inverse cyclotron" to do essentially the same thing at lower frequencies. It is estimated that a 90% fusion power to electricity output could be achieved using this scheme. Cool eh? What's the catch you ask? Well, its generally thought to be impossible (or VERY nearly so). This is because P-B11 fusion does not take place at the tepid 100 Mega Kelvins of DT fusion but instead at a blistering 1.4 BILLION K (123 KeV)!! This obviously complicates things immensely.

There is one method (that I know of) which is capable of direct conversion of nearly all the energy of the fusion plasma into electricity. It is not possible in inertial confinement laser fusion though. It is the P + B-11 reaction. This involves the fusion of protons (normal hydrogen) and Boron 11 in a magnetically confined fusion reactor called a colliding beam fusion reactor. There is nearly zero neutronicity in this reaction (.001) and the product is 3 He atoms. In this scheme there is sort of a scrape off layer at an end of the reactor where the product He ions are directed into a magnetron cavity where they are decelerated and produce microwaves, the microwaves are collected with an antenna and rectified to DC power. Also they use something called an "inverse cyclotron" to do essentially the same thing at lower frequencies. It is estimated that a 90% fusion power to electricity output could be achieved using this scheme. Cool eh? What's the catch you ask? Well, its generally thought to be impossible (or VERY nearly so). This is because P-B11 fusion does not take place at the tepid 100 Mega Kelvins of DT fusion but instead at a blistering 1.4 BILLION K (123 KeV)!! This obviously complicates things immensely.

My research is currently looking into approaches to related areas (as a user, not necessarily as a cryptographer), you may wish to look into "secret sharing", where given a secret (e.g. a private key), a set of participants, and what the literature calls an access structure which is a collection of subsets of participants that you wish to be able to easily recover the secret (called a qualified subset), establishes a two stage protocol:

1. Share - a trusted entity called the dealer takes the secret and encodes it into a set of shares, securely awarding each participant a unique share.
2. Reconstruct - some subset of the participants presents their shares, if the shares are valid and the subset is a qualified subset, the secret is recovered and securely distributed to that subset of participants, otherwise the secret should not be revealed.

There are proactive variants that periodically recut the shares to prevent accumulated leaking of shares over time from forming a qualified subset.

Also there are verified secret sharing schemes which support a verify operation, where a share can be checked for correctness without trying to reconstruct the secret (so that bad dealers can be caught and that at reconstruct time invalid shares can be found prior to reconstruction).

Finally there are "cheating immune" schemes. A cheater is a participant who gives a bogus share at reconstruct time. If they know something about the reconstruction step and can assume the other participants are giving valid shares, some schemes may allow the cheaters to learn something about the secret. In cheating immune schemes, this is prevented.

Finally there are schemes that use verifiable threshold schemes and verifiable secret sharing for digital signatures.

If you are interested in some references, Doug Stinson's bibliography on Secret Sharing (he has some recent work too). Tal Rabin has done some good work, as has Markus Stadler. Recent work by Stanislaw Jarecki has caught my eye.

• Increase developer productivity by buying them faster machines.
• Always attempt to solve performance and scaling problems with more hardware. Don't give programmers an optimization problem unless more hardware completely fails to improve the situation.
• Don't attempt to fix any problem where hardware is or is suspected to be the problem. Replace the hardware.
• Don't attempt to fix any software problem that could be solved by replacing the hardware (OS borked -- throw out the whole machine).
• Pay a premium for hardware with better reliability.

Campus Crusade for Christ
UCI