Slashdot Mirror

Basically, there are two ways to preform replication: "lazy replication" and "eager replication". "Eager replication" means that all updates are atomic across all nodes and that transactions are serializable. However, the problem with "eager replication" is that as you increase the number of nodes n, the probability of deadlock increases on the order of n^5. The "solution", such as it is, is to remove the expectation of serializabilty, using timestamps for concurrency control, only allow commutative transactions on your data, and use two-tier replication. This works for banks and others whose database applications consist mainly of commutative transactions, but won't for many others: YMMV. (Gray's paper also details the differences between having a single "master" node that "owns" all db objects and having each node own several objects.)

IIRC, the way Notes does it is by queuing updates at the local node and using an optimistic concurrency control mechanism when the local node connects to replicate. This is great for the application domain that Notes caters to: I "own" my own calendar, and if I'm out of the office (and have my node -- notebook -- with me), you can't schedule me for an appointment until I come back. However, for many application domains, this won't work.

In any case, that's why Notes does it -- because it can, thanks to the nature of its data domain -- and why most people don't -- because it's hard/impossible for the general case.


~wog
PS -- If you can't get into the ACM Digital Library, check out these lecture notes from Stonebraker's anthology at Berzerkeley.

802.11 wireless and X? Grab your favorite Palm computer... locate the existing wireless solution for your hardware. Download VNC if you have a Windows Ce. If you don't have CE, than go get PalmVNC and wirelessly compute! Geez, what kind of geek ARE you if you couldn't figure that out?!

PalmVNC is here. If you really want to control an X session or Windows box using a 160x160x16 display, be my guest.

UCMP Raptor site

Those Weta machines were busy crunching Seti packets. There's at least 8 of them in the top 1000 for New Zealand, a damn sight faster than any other machine down here....
--

It makes me sick to see things like this. According to a previous article, Bush is predicting a 4.6 trillion dollar surplus over the next ten years. From that, they can't find a measly little ten million dollars to keep this valuable piece of equipment in the sky for another decade? Let's see, 4.6 - 0.00001 = 4.59999. Wow, they're really gonna miss that money.

By the way, check out Berkley's slide show on EUVE. It's fascinating. Describes the equipment on the satellite, as well as it's discoveries very well. Even good for a non-techie.

There is also an ns version 2 available.

I don't know if this is exactly what you're looking for, but it might help.

The paper which has this experiment was just published in Science. You probably need to be connecting from somewhere that has subscribed to Science.

A good intro reference to DFSs is here.

dabacon

Don't RAM chips already have some internal ability to map out bad areas, like the way that IDE/SCSI hard drives come with spare sectors that are automatically mapped in when a regular one fails?

[google]

DRAMs typically improve yield by using spare rows and columns to replace those occupied by defective cells. The repairs are performed using current-blown fuses, laser-blown fuses, or laser-annealed resistor connections. [Coc94] references one case in which memory repair increased the yield from 1% to 51%.

from http://www.cs.berkeley.edu/~rfromm/Courses/SP96/cs 294-4/project1/dram-test.html

That, and go to SETI@Home's website.

Their Top 100 list suggests that maybe Sun, Silicon Graphics, Compaq, Intel, IBM, Apple and HP all use SETI at Home for burn-in...

If the CPU screws up, it's unlikely to create a data unit that gets passed back to SETI and screws up their project; it's more likely that the computer will just spit out a kernel panic or stop responding or something. If the system is otherwise stable, that should be all the error message you need.

BTW, burn-in is used by builders to prevent shipping D.O.A. systems to customers. As far as I've ever known in all my years with electronics, an extended run won't make any physical changes to the ICs that would make it more reliable. If anything, the heat of being overclocked and run 24/7 is likely to upset the metallurgy of solder joints and the doping of semiconductors more than anything else...

Bottom line is that you're *proving* that it's reliable, not *making* it reliable.

The only thing this discovery will do for the scientific community is allow us to make sure that our toy space probes don't wrap themselves around it.
Don't get me wrong, I'm not saying that it isn't neet, but really, it dosn't have much to offer. Couldn't we be spending our space research resources in a better way, like say researching dark matter?
just my $.02
---

Anthony Sale, founder of the Bletchley Park Museums, will talk about the British effort during WWII to break the German Enigma code and the development of the Colossus computer used in that effort. The talk is on Thursday, October 19 at 8:00pm - Check out this link for details.

A couple of problems with #2

2. Unlimited reverse engineering aka "Perfect Reverse Engineering". All products may be disassembled and duplicated without hinderence of patent, trade secrecy, or any other form of intellectual property. Knowledge flow is instantaneous from creator to user.

That's exactly the way it works now. Even the disassembly and duplication of patented inventions is legal, so long as it is "for the mere
purpose of philosophical experimentation, or to ascertain the verity and exactness of the specification"

Trade secrecy laws do not protect against reverse engineering. Trade secrecy laws only protect against "insider jobs" -- where the trade secret is disclosed by someone who is contractually obligated to keep it secret.

The other applicable form of IP is copyright, and it is well established that you have the right to read copyrighted works, and thus, to understand them.

Outcome: The tragedy of the commons. Companies will play "wait-n-see" to see who comes up with difficult to engineer solutions to problems. If they are making a profit, they will not bother to spend money on R&D.

This is not what "tragedy of the commons" means. Tragedy of the commons only applies to depletable resources, like a silo full of corn. If everyone takes corn out of the silo, and no one refills it (or pays money which is then used to refill it), the silo will quickly empty out, and no one will have corn. IP is not a depletable resource, and the "tragedy of the commons" does not apply.

Copyrights and patents create incentives to publish, which is good, but also turn unlimited resources into limited, scarce resources at the same time, which is bad.

Here's hoping that we can remain civil, and arrive at solutions that provide a fair balance for each individual case.

Good patent and copyright laws maximize the amount of disclosure of inventions and publication of works, while at the same time minimize the tremendous accumulation of power that can result from granting a corporate monopoly over an unlimited resource. The fact that our media corporations, which are basically holding companies for copyrights on nearly all of the intellectual work of the 20th century, are quickly becoming the most powerful entities on the planet -- more powerful then even national governments, is a sign that the system is not fairly balanced.

it started with a bang, but will it end a crunch?


redsaso

For Those who haven't had to chance to discover it, and for those who may get nostalgic about it, here is a link to The Story of Mel from the Hacker Folklore section of the Jargon Dictionary.

that aside, Berklee College of Music in Massachusetts (named for founder Lee Berk) doesn't have a psych department. Berkeley in California might, although their Music Dept, it seems, does not even list such a class

Of course, we all know about the importance of the japanese to pearl

;-)

- - - - - - - -
"Never apply a Star Trek solution to a Babylon 5 problem."

For Those who haven't had to chance to discover it, and for those who may get nostalgic about it, here is a link to The Story of Mel from the Hacker Folklore section of the Jargon Dictionary.

that aside, Berklee College of Music in Massachusetts (named for founder Lee Berk) doesn't have a psych department. Berkeley in California might, although their Music Dept, it seems, does not even list such a class

Of course, we all know about the importance of the japanese to pearl

;-)

- - - - - - - -
"Never apply a Star Trek solution to a Babylon 5 problem."

For Those who haven't had to chance to discover it, and for those who may get nostalgic about it, here is a link to The Story of Mel from the Hacker Folklore section of the Jargon Dictionary.

that aside, Berklee College of Music in Massachusetts (named for founder Lee Berk) doesn't have a psych department. Berkeley in California might, although their Music Dept, it seems, does not even list such a class

Of course, we all know about the importance of the japanese to pearl

;-)

- - - - - - - -
"Never apply a Star Trek solution to a Babylon 5 problem."

As a former doctoral candidate at Berkeley, I hasten to point out that:
- Mr. Lockard shares Berkeley's radical dream of politics and community. The real substance of this dream is being able to sit around writing armchair articles about community and politics (see resume above, and Howard Rheingold's much-more-real take on things, HR comment.).
- Academia's 'community' is a sham several orders of magnitude greater than "Virtual Community." Lockard is part of a priviledged (if deluded) few who get to think about "democracy," etc., while enjoying access to libraries, publications and resources that are not available to the common public -- even the Web public.
- The real issue here is access to knowledge. Abhay Bhushan said, that when he wrote ftp, he thought of making any piece of human knowledge available to anyone on Earth. The Internet has always been committed to that. The humanities, like the medieval Catholic church, have always been far from it.
- Most (humanities) academics are glorified gatekeepers to knowledge, whose jobs are fundamentally threatened by the InterNet. Mr. Lockard and others get to lead a genteel life of sitting around in cafes and chatting with students, etc., because they know where information is in hard-to-search publications, where it is buried in libraries, etc. The essence of their work is digging through those publications, taking notes, and remembering how to get to things -- in essence, knowledge management. On top of this, of course, the usual rigamarole of initiation to knowledge, interpretation to students, etc. -- just like the Catholic church had when the Bible was in Latin and 'interpreted' to the people.
- The Internet threatens the jobs of academics in the humanities. It is sad but unsurprising that the humanities have not embraced sharing knowledge online like the sciences. If everything in Berkeley's libraries were online, humanities students would start reading like programmers -- I mean they would have thosands of pages of documentation always with them, and easily find what they needed, instead of spending hours in libraries, looking for and pouring over MeatSpace texts -- and quickly outpace their professors, who have spent years learning that "X is discussed on page yy of T," and other jobs that can be aided greatly by computer automation. But if the public would suddenly have instant access to this information, professors would have more to think about than where to buy their next latte.
- Academia is much harder for the poor to access, than the InterNet Access problems to the InterNet??? Huh? Compared to what? Lockard is missing free email kiosks for the homeless in Budapest and net access in 80% of public library in America. He clearly has no comprehension of what the people who do real work creating the InterNet think about, or the dedication to the sharing of knowledge that permeates everyone who has been involved. No doubt he has a publish-or-perish deadline for his next article, but his thourough ignorance of the history of internet technology is insulting. (I suspect this is because the limited line of knowledge his professors feed his doesn't include much writing by Berners-Lee, even if it is easily available on the InterNet). So much for "access problems."
- The Internet is several orders of magnitude more accessable than academic knowledge, especially in the humanities, who call themselves 'guardians of democracy.' Or was that "guardians of knowledge?" Simply, the InterNet is the most democratically availabel communications medium ever, and Lockard doesn't even bother to touch the wordwide communities it enables -- from programming collectives to Yugoslavian dissidents -- and the very real effects it has on their lives. See Rheingold. Enough said.
- And if the InterNet isn't a democratic enough community, it's academia's fault. Exactly why isn't Lockard committed enough to sharing knowledge, to put his article online (scientists regularly flout copyright restrictions). Academics are supposed to create community, share knowledge, foster understanding. But they -- especially they in Berkeley -- don't do this. Why? It threatens their jobs, their ability to get published in their old boys' networks, their career advancement, their sense of superiority in handing out knowledge, and their very way of life. To wit,
THE REAL SHAM HERE IS ACADEMICS' COMPLAINING. The InterNet represents a real opportunity to democratically distribute knowledge and access to community, but instead of getting on with that real work, Lockard and his acadmic masters complain about the fact that it is elite, that it is corporate, that it ... whatever they can find. That technology is ruining our real relations. That online publishing ruins the ability to learn from a written text. Etc. And why? Because they're scared, they don't want to understand, they don't want newness, they just want another latte and the New World to go away.

As a former doctoral candidate at Berkeley, I hasten to point out that:
- Mr. Lockard shares Berkeley's radical dream of politics and community. The real substance of this dream is being able to sit around writing armchair articles about community and politics (see resume above, and Howard Rheingold's much-more-real take on things, HR comment.).
- Academia's 'community' is a sham several orders of magnitude greater than "Virtual Community." Lockard is part of a priviledged (if deluded) few who get to think about "democracy," etc., while enjoying access to libraries, publications and resources that are not available to the common public -- even the Web public.
- The real issue here is access to knowledge. Abhay Bhushan said, that when he wrote ftp, he thought of making any piece of human knowledge available to anyone on Earth. The Internet has always been committed to that. The humanities, like the medieval Catholic church, have always been far from it.
- Most (humanities) academics are glorified gatekeepers to knowledge, whose jobs are fundamentally threatened by the InterNet. Mr. Lockard and others get to lead a genteel life of sitting around in cafes and chatting with students, etc., because they know where information is in hard-to-search publications, where it is buried in libraries, etc. The essence of their work is digging through those publications, taking notes, and remembering how to get to things -- in essence, knowledge management. On top of this, of course, the usual rigamarole of initiation to knowledge, interpretation to students, etc. -- just like the Catholic church had when the Bible was in Latin and 'interpreted' to the people.
- The Internet threatens the jobs of academics in the humanities. It is sad but unsurprising that the humanities have not embraced sharing knowledge online like the sciences. If everything in Berkeley's libraries were online, humanities students would start reading like programmers -- I mean they would have thosands of pages of documentation always with them, and easily find what they needed, instead of spending hours in libraries, looking for and pouring over MeatSpace texts -- and quickly outpace their professors, who have spent years learning that "X is discussed on page yy of T," and other jobs that can be aided greatly by computer automation. But if the public would suddenly have instant access to this information, professors would have more to think about than where to buy their next latte.
- Academia is much harder for the poor to access, than the InterNet Access problems to the InterNet??? Huh? Compared to what? Lockard is missing free email kiosks for the homeless in Budapest and net access in 80% of public library in America. He clearly has no comprehension of what the people who do real work creating the InterNet think about, or the dedication to the sharing of knowledge that permeates everyone who has been involved. No doubt he has a publish-or-perish deadline for his next article, but his thourough ignorance of the history of internet technology is insulting. (I suspect this is because the limited line of knowledge his professors feed his doesn't include much writing by Berners-Lee, even if it is easily available on the InterNet). So much for "access problems."
- The Internet is several orders of magnitude more accessable than academic knowledge, especially in the humanities, who call themselves 'guardians of democracy.' Or was that "guardians of knowledge?" Simply, the InterNet is the most democratically availabel communications medium ever, and Lockard doesn't even bother to touch the wordwide communities it enables -- from programming collectives to Yugoslavian dissidents -- and the very real effects it has on their lives. See Rheingold. Enough said.
- And if the InterNet isn't a democratic enough community, it's academia's fault. Exactly why isn't Lockard committed enough to sharing knowledge, to put his article online (scientists regularly flout copyright restrictions). Academics are supposed to create community, share knowledge, foster understanding. But they -- especially they in Berkeley -- don't do this. Why? It threatens their jobs, their ability to get published in their old boys' networks, their career advancement, their sense of superiority in handing out knowledge, and their very way of life. To wit,
THE REAL SHAM HERE IS ACADEMICS' COMPLAINING. The InterNet represents a real opportunity to democratically distribute knowledge and access to community, but instead of getting on with that real work, Lockard and his acadmic masters complain about the fact that it is elite, that it is corporate, that it ... whatever they can find. That technology is ruining our real relations. That online publishing ruins the ability to learn from a written text. Etc. And why? Because they're scared, they don't want to understand, they don't want newness, they just want another latte and the New World to go away.

And more like SETI@Home -- using the net to do distributed computing, but the article seems to indicate that storage will be distributed as well. And I think that all the computers that are part of the network will be dedicated to this task; you won't be able to hook up your home machine to GriPhyN. If you're going to spread out your storage, you will need to be able to rely on it being there (i.e., Joe Homeuser shut off his PC and now we can't get access to the secrets of the atom),

-------------

Most of the algorhythms I play with only use 8 bits, so they would have to be some pretty dumb terrorists. :)

So you play around with encryption that can be trivially broken? How does that enhance people's privacy? It seems like it would just give naive users a false sense of security. I hope you don't plan on releasing your code.

My point is that one of these things (encryption) is really designed to enhance people's privacy while the other (the codec, at least at this point) is designed to take away intellectual property.

This kind of attitude makes me very sad. This attitude is why the MPAA is winning.

If this were a battle between the artists who create great movies and the pirates who would deprive them of their source of payment, I would side with the artists. But it is not about payment; it is about control. And it's not even about control by the artists; it's about control by the large movie studios. If they succeed in getting the U.S. government to grant them that control, then the studios' shareholders will benefit, and everyone else will suffer -- artists most of all.

If we do not have a free video codec, then we will have only a proprietary video codec, legally playable only using commercial players made by the big studios, or by companies who have signed a strict licensing agreement with them.

I don't know if "Divx Deux" will amount to anything; personally I'm more inclined to place my bets on the Ogg project. But someone needs to do it, and it needs to happen as quickly as possible, before MS MPEG-4 becomes too entrenched.

For more on this topic please read my essay on digital media and the DMCA.

Whilst projects like distributed.net and Seti@Home have clocked up shocking amounts of processor time (410497.11 years on Seti@Home), they're still running on the 'cool factor' of having your machine break codes or search for aliens.

Sites such as ProcessTree, and others, have been talking of paying for your computer time with micropayments, but so far nothing seems to have got off the ground.

Presumably with the added incentive of cash, the number of computers taking part will rocket. Does anyone have any firm information on the progress of these schemes?

I found this interesting article on the internet recently:

http://www.cs.berkeley.edu/~nweaver/x box.html

I'm wondering about those legacy PC bottlenecks on the xbox. It's on a x86 based processor going against a 100% 128-bit PS2 Emotion Engine that's got a main 300 mhz CPU and two vector coprocessors that blow Pentium III floating point ops out of the water.

For example, SquareSoft is trying to make their beat-em-up, The Bouncer, have Dolby Digital 5.1 support in gameplay. They used to think it was impossible, but it might not be so with the VU units. It seems MS just thought "Hey, we can beat 300 mhz easy!" without thinking about the hidden trump card in the PS2. Also, I noticed the 6.4 GB/sec memory bandwidth on the Xbox (shared, so it indludes video memory speed). And the PS2 has dedicated VRAM that flies at 48 GB/sec. So what do you guys think?
- Amon CMB

That is incorrect. The Integrated Circuit was most definitely patented. And I quote:

"Robert Noyce took the helm of the new enterprise and it was his invention of the integrated circuit that same year (along with Jack Kilby of TI who shared the patents) that would make Fairchild's fortune"

Taken from this article.

Here are a few other references:

http://inv entors.about.com/science/inventors/library/weekly/ aa080498.htm

http://www.usatoday.com/life/cyb er/tech/ctb218.htm

Poor SOBs. Their patents ran out in 81. But it looks like they got to have a good run of it with the screwed up Japanese patent system!

I was 12 years old. I had just been introduced to computers. The first language I had been taught was BASIC at the Lawrence Hall of Science in Berkeley, CA. Eventually a microcomputer store opened in my home town called The Byte Shop, where I started annoying the sales people by fiddling with their SOL-20s and Apples, writing little ditties in BASIC. I got to know BASIC real well. I got to know several dialects of BASIC, and could intelligently discuss the relative merits of each.

With all that hands-on experience, I can say without fear of contradiction: Microsoft BASIC was one of the worst BASIC interpreters available. The only one I can think of off-hand that was even worse was Northstar BASIC.

I settled in to a happy relationship with a variant called Extended Cassette BASIC, published by Processor Technology for the SOL-20. This BASIC (back in 1978, mind), had:

• Multiple-line user-defined functions,
• Matrix math operations
• Auto-indent of program LISTings,
• Ran in 16K,
• Cost less than $100 (less than 1/5 of Microsoft's inferior offering).

Microsoft, in typical form, took another ten years to get as far, and consumed ten times as much memory doing it.

I really should drag out my old SOL-20 and do some side-by-side comparisons of Microsoft's old stuff.

Schwab

There were even a few before Vannevar Bush (not to take anything away from Van).

H.G. Wells conceived a "world Brain" or "world Mind" back in the late thirties. The basic concepts are very similar to the Web.

And even earlier (like 1907/8) A Belgian named Paul Otlet had similar ideas and actually executed some of them in a paper fashion. See this entry at Michael Buckland's site, or a brief mention in Redesigning Library Services: A Manifesto

There were even a few before Vannevar Bush (not to take anything away from Van).

H.G. Wells conceived a "world Brain" or "world Mind" back in the late thirties. The basic concepts are very similar to the Web.

And even earlier (like 1907/8) A Belgian named Paul Otlet had similar ideas and actually executed some of them in a paper fashion. See this entry at Michael Buckland's site, or a brief mention in Redesigning Library Services: A Manifesto

There were even a few before Vannevar Bush (not to take anything away from Van).

H.G. Wells conceived a "world Brain" or "world Mind" back in the late thirties. The basic concepts are very similar to the Web.

And even earlier (like 1907/8) A Belgian named Paul Otlet had similar ideas and actually executed some of them in a paper fashion. See this entry at Michael Buckland's site, or a brief mention in Redesigning Library Services: A Manifesto

Have a look at this site for some information about anonymous publishing. I found the method they used quite interesting, not too easy to think of it... But I don't know if there are actually servers available that do stuff like this or if all of this is only purely theoretical... Greetings, Alex

more info can be found here.

Total Group stats at SETI@HOME

This is on both the d.net and SETI FAQ lists. Such clients transfer small (<300 KB) chunks of data during (say) your wife's e-mail check.
<O
( \
XGNOME vs. KDE: the game!

At least according to some of our TA's. http://www-inst.eecs.berkeley.edu/~cs61c/ A first? :)

DigiCash was founded in 1990 by cryptologist David Chaum (site apparently not updated in recent years...) who owned the two (three?) major patents covering completely anonymized digital cash. To my knowledge it still isn't certain whether it's possible to create a truly anonymous digital money scheme without violating these patents.

David Wagner has done some work on that. Sample code is available in lucre.

It appears to avoid Chaum's patents but I don't know if any patent lawyer has had a good look at it. Your "not certain whether it's possible" comment may stand even after reading this (if you haven't read it already).

Use any spare cycles to read Slashdot

Read Slashdot? Why not something useful like Seti@home or RC5?

I should add that 78% of SETI@Home users, that filled out the poll, are in favor for sending out signals to ETs. OTOH, that means that 22% have certain doubts...

They're really the same car.

I don't know if anybody is still reading the comments here; I'm going to make a similar post on a usenet group or two.

Anyway, I'd like to form a team and do it! Yeah, I know that I should have put a team together before the contest, but I thought I might be able to do the whole thing myself. Fortunately, the task divides neatly into two or three (mostly) independent parts, so that teammates won't need to be in close contact with each other.

This paragraph won't make a lot of sense unless you've read the task description, so do that now if you haven't already. My plan is to write (in Perl, because it's the rapid-prototyping language I'm most familiar with) a program to translate GML into C. This code will be compiled and linked with the actual rendering code, which will be written by someone else on the team. I'm assuming that the procedural texturing precludes the use of canned open-source code. A third teammate should write GML code to thoroughly test both the translator and the renderer. Of course, there may be more than one person on each task. Please contact me if you're interested in joining my team, or if you have any useful suggestions.

Note: I will probably not be able to answer my mail until after midnight tonight. I probably won't have time to post this to usenet, so feel free to post a copy of this (together with an explanation of what the contest is all about) to all appropriate newsgroups.

--Eric Kuritzky
"Insert shiny happy .sig here"

Speaking of software challenge and tremendous amounts of data...

Will SETI be allowed access to the telescope? I hope so...

I think a ratio type of thing would be a great idea, but how in the world can this be done? Obviously this is not too practicle in an anonymous situation

Perhaps the solution is a pseudonymous system, like the pseudonymous mail sytem. It could also enforce the use of GnuPG and PGP.

I maintain a MPEG 4 Structured Audio decoder, sfront, that supports real-time low latency work -- MIDI and audio input and audio output, suitable for performance work (and people use it on stage today, under Linux).

Using the techniques described on linux-audio-dev, and using a machine pruned of some badly-behaving daemons, can make Linux work well for low-latency audio today, for some apps. Note that my app doesn't write or read to disk -- a lot of the remaining problem areas are for apps like hard disk recorders which need disk-I/O.

Personally, I found the AI aspect of the article the most interesting. I am a fan of Douglas Hofstadter. His books Fluid Concepts and Creative Analogies and Le Ton Beau de Marot are incredibly beautiful and insightful. I strongly recommend them to anyone who is interested in AI, philosophy of knowledge, psychology etc. etc.

Now, in an attempt to stay on-topic, I will come back to the article. Hofstadter's basic message seems to be that intelligence is a matter of degree more than anything else. Things such as medium (biological, mechanical, electrical etc.) are irrelevant to the basic question. If he is right, then I can see intelligent beings being much more likely.

As a strongly religious person, I believe that there is some "soul" which imparts to us an aspect that makes us (the human species) unique on the planet (please moderators - that's just an opinion :). But the teachings of my faith seem to be quite clear that every planet has its creatures (not necessarily intelligent) (Unfortunately I don't have a good reference to this specific aspect of my faith).

But in another way, I have to admit that I feel very threatened by the possibility of other intelligence. As an individual I like the fact that I am a member of a species which has done absolutely incredible things. Humanity has created intended beauty in a way that so far seems completely unique in the world. I believe strongly that we can overcome our multitude of problems, and when we do, we will do more that is completely unimaginable at this time. If AI or ET's were to usurp that position of beauty-creators, it would be very very damaging to humanities self-concept. That said, I am fascinated by AI, and I would love to do work in the field (after I make my startup millions :).

As a final comment in this decidedly rambling post: get SETI@home !

Hello
Theres an article in TheStandard detailing Dr. David Anderson's (of seti@home fame) new start up: United Devices. Its would seem from the spiel on their web page that they plan on using a set@home model to provide computer muscle specifically for research.
Not a great deal on their web page at the moment, but it looks interesting.

Hello
Theres an article in TheStandard detailing Dr. David Anderson's (of seti@home fame) new start up: United Devices. Its would seem from the spiel on their web page that they plan on using a set@home model to provide computer muscle specifically for research.
Not a great deal on their web page at the moment, but it looks interesting.

Some pieces that free software could do much better:

• dynamic optimization rather than a simple JIT (see also Self),
• layers of intermediate representation carrying more semantic information, giving a big boost to projects with code available,
• a more powerful security model,
• using the same deduction framework to assist with optimizations and parsing,
• going even further and integrating the language into the programming environment,
• completely kick-ass garbage collection,
• and so on...

Re-inventing OLE / DCOM is silly. There's so much else to do with much greater pay-offs...

Looks like it that at these levels of size you find normal mechanical properties and semi-quantum properties such as the van der waals force combine and actually improve the function of these tubes, a van der waals bearing instead of a ball bearing.

• MMM Group at University of Nijmegen [publications]
• Machine Listening @ MIT Media Lab
• Affective Computing @ MIT Media Lab
• Musclefish

• Music, Cognition, and Computerized Sound, Perry R. Cook
• Music, Mind and Machine, Peter Desain and Henkjan Honing
• The Scientist and Engineer's Guide to Digital Signal Processing, Steven W. Smith
• Neural Networks for Pattern Recognition, Christopher M. Bishop

• Tracking Musical Beats in Real Time, Paul E. Allen and Roger B. Dannenberg
• A Model for Musical Rhythm, Jeff A. Bilmes
• Autocorrelation and the Study of Musical Expression, Peter Desain, Siebe de Vos
• A Beat Tracking System for Audio Signals, Simon Dixon
• Prediction-Driven Computational Auditory Scene Analysis for Dense Sound Mixtures, Daniel P. W. Ellis
• A Similarity Measure for Automatic Audio Classification, Jonathan Foote
• Representing Rhythmic Patterns in a Network of Oscillators, Michael Gasser and Douglas Eck
• Adaptive Signal Models: Theory, Algorithms, and Audio Applications, Michael Mark Goodwin
• Recognition of Music Types, Hagen Soltau, Tanja Schultz, Martin Westphal, Alex Waibel
• Irrelevant Features and the Subset Selection Problem, George H. John, Ron Kohavi, Karl Pfleger
• Beat tracking with a nonlinear oscilator, Edward W. Large
• Modeling beat perception with a nonlinear oscilator, Edward W. Large
• Automatic Transcription of Simple Polyphonic Music: Robust Front End Processing, Keith D. Martin
• Musical instrument identification: A pattern-recognition approach, Keith D. Martin and Youngmoo E. Kim
• Music Content Analysis through Models of Audition, Keith D. Martin, Eric D. Scheirer, Barry L. Vercoe
• Musical Sound Information: Musical gestures and embedding synthesis, Eric Metois
• A Machine Learning Approach to Musical Style Recognition, Roger B. Dannenberg, Belinda Thom, and David Watson
• Resonanc e and the perception of musical meter, Large, E. W., & Kolen, J. F.
• Music-Listening Systems, Eric D. Scheirer
• Tempo and beat analysis of acoustic musical signals, Eric D. Scheirer
• Content-Based Classification, Search, and Retrieval of Audio, Erling Wold, Thom Blum, Douglas Keislar, James Wheaton
• Classification, Search, and Retrieval of Audio, Erling Wold, Thom Blum, Douglas Keislar, James Wheaton

• MMM Group at University of Nijmegen [publications]
• Machine Listening @ MIT Media Lab
• Affective Computing @ MIT Media Lab
• Musclefish

• Music, Cognition, and Computerized Sound, Perry R. Cook
• Music, Mind and Machine, Peter Desain and Henkjan Honing
• The Scientist and Engineer's Guide to Digital Signal Processing, Steven W. Smith
• Neural Networks for Pattern Recognition, Christopher M. Bishop

• Tracking Musical Beats in Real Time, Paul E. Allen and Roger B. Dannenberg
• A Model for Musical Rhythm, Jeff A. Bilmes
• Autocorrelation and the Study of Musical Expression, Peter Desain, Siebe de Vos
• A Beat Tracking System for Audio Signals, Simon Dixon
• Prediction-Driven Computational Auditory Scene Analysis for Dense Sound Mixtures, Daniel P. W. Ellis
• A Similarity Measure for Automatic Audio Classification, Jonathan Foote
• Representing Rhythmic Patterns in a Network of Oscillators, Michael Gasser and Douglas Eck
• Adaptive Signal Models: Theory, Algorithms, and Audio Applications, Michael Mark Goodwin
• Recognition of Music Types, Hagen Soltau, Tanja Schultz, Martin Westphal, Alex Waibel
• Irrelevant Features and the Subset Selection Problem, George H. John, Ron Kohavi, Karl Pfleger
• Beat tracking with a nonlinear oscilator, Edward W. Large
• Modeling beat perception with a nonlinear oscilator, Edward W. Large
• Automatic Transcription of Simple Polyphonic Music: Robust Front End Processing, Keith D. Martin
• Musical instrument identification: A pattern-recognition approach, Keith D. Martin and Youngmoo E. Kim
• Music Content Analysis through Models of Audition, Keith D. Martin, Eric D. Scheirer, Barry L. Vercoe
• Musical Sound Information: Musical gestures and embedding synthesis, Eric Metois
• A Machine Learning Approach to Musical Style Recognition, Roger B. Dannenberg, Belinda Thom, and David Watson
• Resonanc e and the perception of musical meter, Large, E. W., & Kolen, J. F.
• Music-Listening Systems, Eric D. Scheirer
• Tempo and beat analysis of acoustic musical signals, Eric D. Scheirer
• Content-Based Classification, Search, and Retrieval of Audio, Erling Wold, Thom Blum, Douglas Keislar, James Wheaton
• Classification, Search, and Retrieval of Audio, Erling Wold, Thom Blum, Douglas Keislar, James Wheaton

A good source for concise, challenging programming problems is the archives of past ACM programming contests. Take a look at The Annual Berkeley Programming Contest. While during the contests, the only metric that matters is whether your program generates the correct output for the given (unknown) input, many of the problems raise issues that can be debated in more depth: what is the easiest solution to code? what is the most elegant approach? what is the most efficient? What are the strengths and weaknesses of using C, C++, or Java as the implementation language? etc.

When I was learning to program, my favorite source sof programming inspiration were popular math books, eg The Mathematical Tourist by Ivars Peterson, and the Amateur Scientist and Mathematical Recreations columns of Scientific American. I'd implement simulations of percolation; Conway's game of life in one, two, three dimensions; plot the Mandelbrot and Julia sets; iterated function systems fractals; simulating preditor-prey models; simulating the spread of a disease or forest-fire. Of course, all of that was on DOS using Turbo Pascal. Now that access to networked unix machines is so widely available, there are many new opportunities (eg, for network-related stuff) available.

Good reading material is a very important source of challenges, problems, and insight into computer science as a science. I'm thinking particularly of the famous Structure and Interpretation of Computer Programs, the textbook used in freshman computer science cources at MIT and University of California, Berkeley. The book demonstrates clearly that there are multiple approaches to problem solving, each with its advantages and disadvantages; SICP also convinces the student of the beauty and power of high-level abstraction.

Load balancing? Easy to write, hard to make work well. You need to compare the cost of migration to the benefits of balancing, and you need to make decisions based on partial and outdated information. Many early systems thrashed because everybody would migrate to the idle processor, which then became overloaded, so everybody migrated somewhere else, etc.

Speaking of migration, it's a mess. The only system I know of that implemented migration fully was Locus, out of UCLA. The trouble is that whenever a process has a dependency on or a hook into its environment, that connection must be migrated too. Open files, working directory, sockets, controlling tty, signals, process parent/child relationships, and many more details must be handled. Not fun, and the benefits turned out to be mostly minor (though I do recall writing a cool version of "find" that migrated itself to the machine that stored the current subtree as it ran).

The issue of supporting distributed applications is generally considered to be separate from writing a truly distributed OS. Most of what a distributed application needs can be provided by a good communications library. To some extent, we're still learning exactly what such a library should have. What about SETI@home is specialized to it, and what's universal? I don't think we've completely figured it out.

The following is a non-exhaustive list of major concerns and design issues that must be addressed in a distributed OS. We have fairly good solutions to some, but most have not yet been solved:

• Process control. How much process migration is a Good Thing? How do you decide what machine to use to start a process, and when do you decide to migrate it to another?
• Communication and synchronization. What facilities does a distributed application need? How do we make those easy to use?
• Reliability. How do we deal with the inevitable machine failures?
• Replication. What processes and data should be duplicated on different systems? Are you doing the replication for performance, for reliability, or both? How do you manage updates to replicated data? How do you keep replicated process synchronized?
• Lack of global knowledge. How do you make decisions based on partial information?
• Naming. What names to things have. Do you have a shared global namespace, or a private one? How do you resolve names? What do you do when people and objects move?
• Scalability. How does the system behave when the number of computers/users/programs jumps by a factor of 10 or 100? (This is a place where Napster doesn't do real well.)
• Compatibility. How do you support existing software? Do you run on only one kind of hardware, or many?
• Security. Who gets to run on what machine?

Finally, I should note that the list of projects at U of Arizona might appear to be complete, but it omits a lot of important projects. Four that jump to my mind are Locus and Ficus from UCLA (though the latter is more of a distributed filesystem than an OS), Coda from CMU (again a DFS, rather well-known to Linux folks), and of course the extremely important Network of Workstations work out of UC Berkeley, which led to Inktomi and Hotbot.