Slashdot Mirror

Stack machines are not necessarily inefficient- it's a common myth that they "obviously" can't be parrallelised. In fact, the BOOST project (read the "final paper" PDF, not the html) showed that superscalar stack machines are easy to design, and have comparable performance to register machines, while retaining stack machine's superior interrupt handling and downright simpler and more sensible programming model!

It's mainly just an accident of history that the register-machine geeks got their superscalar on before the stack-machine geeks. Once you go superscalar, register files are really just optimisation hints to the chip anyway - might as well use the implicit encoding of a stack machine asm, and use up less mem bandwidth...

This site is for searching CS publications based on citations and other criteria. There are numerous papers on quantum computing many of which provide an excellent introduction. Ofcourse the webpage of faculty and their lecture notes ( here and here ) provide an excellent introduction. I would recomment going through the lecture slides before attacking a few of the more readable and fundamental papers.

This site is for searching CS publications based on citations and other criteria. There are numerous papers on quantum computing many of which provide an excellent introduction. Ofcourse the webpage of faculty and their lecture notes ( here and here ) provide an excellent introduction. I would recomment going through the lecture slides before attacking a few of the more readable and fundamental papers.

Why fork?

Good question. The initial fork occured a few years ago, when a team of developers started to work on Gimp16 for adding 16-bit color channels into the GIMP. This was needed for editing films, but it was not appropriate at that time to integrate the new code into the core so this became a fork. However, the GIMP developers expected that the main GIMP code and Gimp16 (which was later called Hollywood Gimp, Gimp's Film Version and then FilmGimp) would converge later and that the core of GIMP 2.0 would support most of the features that were required for film editing (mainly 16-bit and floating point color channels). The user interface may have kept some differences due to the specific needs of film editors.

Unfortunately, for various reasons (political as well as technical), the gap between GIMP and FilmGimp widened a few months ago, soon after Robin Rowe took over the maintenance of FilmGimp and resurected the project that had not been very active in the last two years. There was some discussion about the fork on the GIMP developers mailing list in November and December last year (you can check the list archives here). The conclusion was that the FilmGimp developers were not interested in bringing their code closer to the current GIMP, and there are too few people working on GEGL (the library that should bring 16-bit and float channels into the GIMP) so it will still take a while before the main GIMP code is suitable for film editing. I am still sad about the way this whole thing happened. I tried to bring the two projects closer to each other, but obviously I failed.

I don't know how the future will look like. I wish the CinePaint developers good luck (honestly) and I hope that they will be successful. This fork of the GIMP suits the specific needs of the film industry and I hope that many studios will be able to use it and do great stuff with it. However, I expect that most people interested in photo editing, web design and general graphic editing will find that the GIMP is more suitable for them than CinePaint.

By the way, if you want to know some of the plans for the future of the GIMP, I suggest that you have a look at developer.gimp.org. In particular, read the plans for the future of the GIMP, posted in December 2000 but still valid. Besides this, the developers mailing list and the list of enhancements submitted to Bugzilla are good sources of information.

The Cylons are obviously related to the indians in Mark Twain's "Fenimore Cooper's Literary Offences." I encourage any fans of literature, or the cinema, to read this. Yes, way back then, the bad guys were notoriously bad aims.

Here's a link I found to some sort of essay about this: http://www.ocf.berkeley.edu/~jparsons/twain/cooper 1.html.
This work of Twain's is found in many Twain anthologies and other collections.

I hate to point this out, but physical modelling has been around since the mid-80's in music technology research labs like CCRMA and CNMAT, but only until the early 90's was the technology available to implement the algorithms cheaply.

Many other companies (specifically synthesizer companies) make products based on modelling - Access Music, Waldorf GmbH, Novation, etc. Don't forget the big boys like Yamaha, Roland, and Korg.

If you want more information on new technologies in music, I'd suggest looking at Hartmann's Neuron and related products - they're actually using neural nets and controlled feedback to add musical randomness into the sound.

Finally, there are other people who have been making unique music instruments for quite some time - but not necessarily for child development. Check out Buchla and Associates for some really unique instruments.

I have been looking at a Java-based hybrid modeling package called Ptolemy II, which does signal processing, graph theory, and linear algebra. The code looks really nice; I'm planning to use it for realtime systems analysis. Any experience with this?

I'd recommend Geoffry Moore's Crossing the Chasm book to read more about marketing hi-tech products to mainstream customers. Here is a rough estimate of the bell curve I was referring to.

Of course not all real numbers (or even rationals) can be represented as finite binary fractions -- let alone onex that fit into 53 bits! But floating point numbers are numbers, not fuzz. Add 0.5 and 0.5 in floating point, and you'll get 1.0 exactly -- not 1.0 plus or minus a unit in the last place.

Why does this matter if you can't represent your original data exactly? Because in intermediate computations, we rely on certain relationships holding. This is important in geometric computation, for example, since we'd like a consistent view of the world. Three ordered points on the plane should be listed clockwise or counterclockwise, or be colinear -- but not more than one of those. See the papers on robust geometric predicates by Jonathan Shewchuk, or papers on floating point computation by W. Kahan for details.

Of course not all real numbers (or even rationals) can be represented as finite binary fractions -- let alone onex that fit into 53 bits! But floating point numbers are numbers, not fuzz. Add 0.5 and 0.5 in floating point, and you'll get 1.0 exactly -- not 1.0 plus or minus a unit in the last place.

Why does this matter if you can't represent your original data exactly? Because in intermediate computations, we rely on certain relationships holding. This is important in geometric computation, for example, since we'd like a consistent view of the world. Three ordered points on the plane should be listed clockwise or counterclockwise, or be colinear -- but not more than one of those. See the papers on robust geometric predicates by Jonathan Shewchuk, or papers on floating point computation by W. Kahan for details.

Try http://www-inst.eecs.berkeley.edu/~pshyu/engnews/e ngnews.html

Look. It's a trade off. Go to Carnegie-Mellon, meet other geeks and commiserate. Go to a state university, meet some kindergarten teachers-in-training and get some shaggin' done. You can't have it both ways, or can you?

http://robotics.eecs.berkeley.edu/~pister/SmartDus t/
http://www-bsac.eecs.berkeley.edu/~warneke/SmartDu st/

http://robotics.eecs.berkeley.edu/~pister/SmartDus t/
http://www-bsac.eecs.berkeley.edu/~warneke/SmartDu st/

Hey, man: the Beaver was smart enough to graduate from Cal so he'd probably make a pretty good patent examiner.

CANNES, France -- 3GSM Congress -- There's a big problem with connecting public wireless LAN access points to GSM/GPRS cellular networks, according to SIM card vendor SchulmbergerSema. 802.11b hotspots provide hackers with an easy way to grab user information from the wide-area network itself, the company tells Unstrung.

The heart of the problem is that when the GSM standard was being defined back in the late 80s, no one imagined that a hacker could set up his own wireless network to gain access to an operator's network and the user data therein. Therefore, GSM networks only authenticate the details held on the SIM card in a user's device before starting a session on the network. The user's device doesn't check the credentials of the network it is attempting to access.

This was fine before the advent of wireless LAN. But now for a minimal outlay anyone can own a wireless network.

At the same time, vendors and operators are starting to use SIM card-based authentication front-end systems for public wireless LAN networks, which allow them to link the user back to the home location register (HLR) database on the GSM network and thus manage and bill a subscriber on the WLAN network in the same way as they would on the wide-area network.

This all adds up to networks that could be vulnerable to hacker attacks, according to Schlumberger.

Hackers can set up "rogue" hotspots that users will access in the belief they are on the genuine public wireless LAN network. Once users are on the fake network, it is easy for the hacker to access data held on the device via the 802.11 connection (see WLAN: The Four S's and this paper for more on the insecurity of wireless LAN). Hackers can then break into the SIM software on the user's device and get the codes held there. They can then use that information to fool the GSM authentication system and thus gain access to the network.

Schlumberger say that this won't be a problem once UMTS networks are available, because the 3G standard ensures what's known as "mutual authentication" -- the network authenticates a user device, and the device confirms that it is actually on a valid network before the session can proceed.

However, for public wireless LAN implementations that will connect to backend systems on GSM and GPRS networks, Schlumberger has developed a SIM card-based system (surprise!) that enables mutual authentication between the device and networks that are accessed via the gateway of public wireless LAN hotspots. The mutual authentication takes place via algorithms on the card itself rather than in SIM card software on the device.

Schlumberger is showing a system at the 3GSM congress that uses a separate smartcard and reader plugged into a WLAN-enabled laptop. However, the firm says that the smartcard and radio could be integrated into one PCMCIA card, much in the way that Nokia Corp. (NYSE: NOK - message board) has done.

Orange France is currently testing Schlumberger's security system. Schlumberger expects that operators will start to roll it out before the end of this year.

-- Dan Jones, Senior Editor, Unstrung
http://www.unstrung.com

If ya don't read the article, check out WLAN: The Four S's, and a WEP FAQ.

Actually, what we need is a Superscalar stack architecture like BOOST.

It's a common myth that stack machines can't go superscalar - the BOOST project definitively smashed that myth.

42.7% of all statistics are made up on the spot. - Steven Wright

was anybody else curious what the technical difference between a bug and insect was?

Umm, without asking my wife (which would probably earn me a beating), I don't actually recall what the defining characteristics are, but bugs are a strict subset of insects. I believe one difference is that all bugs live in the water.

Here, a bit of googling found this. Looks like I was wrong about the water bit, although I think a lot of them do live their entire lives in water and the hardened part of the half-wing is somehow useful to them there.

If finding things like this, or the fact that there are more suns in the universe than grains of sands on earth, doesn't convince religious fundamentalists of evolution, do you think that the discovery of a bunch of microorganisms will convince them?

I have been folowing an OSS project for some time now. BOINC from berkley. I wonder if the SETI gruop is aware of this and if the opening of the software is in co-op with BOINC.

BOINC is a framework on which to build distributed computing projects with ease. The overview is rather complete, in my humble opinion.

Anyone familar with BOINC or have any comments?

I have been folowing an OSS project for some time now. BOINC from berkley. I wonder if the SETI gruop is aware of this and if the opening of the software is in co-op with BOINC.

BOINC is a framework on which to build distributed computing projects with ease. The overview is rather complete, in my humble opinion.

Anyone familar with BOINC or have any comments?

"Linux has a harder time with some of the ultra-small notebooks; they use weird proprietary drivers which Linux does not support a lot of the time." Enh. I'm running an ultra-small (Portege 2000 and the only thing that won't work is the modem. Which sucks for just about every laptop out there, regardless of size.

SETI is quite different. It gets data from some misterious source, does some unknown analysis of it (in theory), draws pretty pictures, and sends processed data back. It could as well work on cracking encryption, and I bet nobody would notice.

Perhaps I am not nearly as concerned that Berkeley is actually a front organization for Ashcroft and the rest of the "right wing conspiracy".

But seriously, I consider the source. It is sponsored by a university known for protesting against every military action, located in a town that was one of the first to pass a resolution against the current war.

Also, SETI@home major users of the client include Sun, Ohio University, Intel and others, who have surely looked inside. If you go to SETI@home they explain in great detail what the program is doing.

Because Berkeley is so liberal and what I consider "anti-USA" I had considered NOT participating in SETI. In the end, I decided the science was more important than the politics.

Basically, I am saying that there is so much info out there, you would have to be paranoid to be worried that Berkeley is actually a front to have us decrypt stuff for the govt. I have done my homework (more than the moderator who modded this as a troll). I know what server it connects to, its not a mysterious source. (I have a firewall that only lets it connect on port 80 to one FQDN)

We may disagree on principal, which is fine. But don't be under any illusion that I have not done my homework. I trust seti because I feel I have good reason to trust them. More so than other distributed computing programs.

Oh, and Adobe Photoshop ALSO is for editing pictures. It tries to connect to mysterious servers all the time. Ask anyone who has it behind a firewall.

SETI is quite different. It gets data from some misterious source, does some unknown analysis of it (in theory), draws pretty pictures, and sends processed data back. It could as well work on cracking encryption, and I bet nobody would notice.

Perhaps I am not nearly as concerned that Berkeley is actually a front organization for Ashcroft and the rest of the "right wing conspiracy".

But seriously, I consider the source. It is sponsored by a university known for protesting against every military action, located in a town that was one of the first to pass a resolution against the current war.

Also, SETI@home major users of the client include Sun, Ohio University, Intel and others, who have surely looked inside. If you go to SETI@home they explain in great detail what the program is doing.

Because Berkeley is so liberal and what I consider "anti-USA" I had considered NOT participating in SETI. In the end, I decided the science was more important than the politics.

Basically, I am saying that there is so much info out there, you would have to be paranoid to be worried that Berkeley is actually a front to have us decrypt stuff for the govt. I have done my homework (more than the moderator who modded this as a troll). I know what server it connects to, its not a mysterious source. (I have a firewall that only lets it connect on port 80 to one FQDN)

We may disagree on principal, which is fine. But don't be under any illusion that I have not done my homework. I trust seti because I feel I have good reason to trust them. More so than other distributed computing programs.

Oh, and Adobe Photoshop ALSO is for editing pictures. It tries to connect to mysterious servers all the time. Ask anyone who has it behind a firewall.

SETI is quite different. It gets data from some misterious source, does some unknown analysis of it (in theory), draws pretty pictures, and sends processed data back. It could as well work on cracking encryption, and I bet nobody would notice.

Perhaps I am not nearly as concerned that Berkeley is actually a front organization for Ashcroft and the rest of the "right wing conspiracy".

But seriously, I consider the source. It is sponsored by a university known for protesting against every military action, located in a town that was one of the first to pass a resolution against the current war.

Also, SETI@home major users of the client include Sun, Ohio University, Intel and others, who have surely looked inside. If you go to SETI@home they explain in great detail what the program is doing.

Because Berkeley is so liberal and what I consider "anti-USA" I had considered NOT participating in SETI. In the end, I decided the science was more important than the politics.

Basically, I am saying that there is so much info out there, you would have to be paranoid to be worried that Berkeley is actually a front to have us decrypt stuff for the govt. I have done my homework (more than the moderator who modded this as a troll). I know what server it connects to, its not a mysterious source. (I have a firewall that only lets it connect on port 80 to one FQDN)

We may disagree on principal, which is fine. But don't be under any illusion that I have not done my homework. I trust seti because I feel I have good reason to trust them. More so than other distributed computing programs.

Oh, and Adobe Photoshop ALSO is for editing pictures. It tries to connect to mysterious servers all the time. Ask anyone who has it behind a firewall.

Now, maybe someone will be able to make this go. But this company doesn't look like it. If you manage to get to their web site and look at the programming language "Viva" they have designed, it looks like you are drawing circuit diagrams. Imagine programming a complex algorithm with that.

There are already better approaches to programming FPGAs (here, here, here). Look for "reconfigurable computing" on Google and browse around.

Slammer hit so hard and fast (doubling every 8 seconds, peak scanning rate in 3 minutes, analysis.

An "hour" before is a preposterous claim. They might have gotten in 10 seconds before, or even a minute if the first couple of copies were on bad links, but an hour is total, complete, and UTTERLY ridiculous claims to make.

The only way they could make the claim is if they found an extra-buggy, prerelease version. IF so, we need to know about it as it aids in understanding the author.

My bet is they saw some unrelated script-kiddie scanning (we saw some of this in our OWN data sets) and someone in marketing is trying to say that they saw the worm 2 hours ahead of time.

Justin.

I highly recommend using the Berkeley:

SafeTP

for secure ftp transfers. The SafeTP client & server code runs well on Unix / Linux as well as those MS-based boxen.

Our ISP has been successfully using SafeTP for several years. Their windoz users transparently use SafeTP with any Windoz FTP application.

Unix folks get that good ol' command line tool. Other Unix interfaces exist for HP-UX, Solaris, Linux, DEC OSF, FreeBSD, OpenBSD, Irix, AIX, etc. SafeTP works across firewalls as well.

With cryptographic protection and authentication, SafeTP is a nice ftp solution.

I highly recommend using the Berkeley:

SafeTP

for secure ftp transfers. The SafeTP client & server code runs well on Unix / Linux as well as those MS-based boxen.

Our ISP has been successfully using SafeTP for several years. Their windoz users transparently use SafeTP with any Windoz FTP application.

Unix folks get that good ol' command line tool. Other Unix interfaces exist for HP-UX, Solaris, Linux, DEC OSF, FreeBSD, OpenBSD, Irix, AIX, etc. SafeTP works across firewalls as well.

With cryptographic protection and authentication, SafeTP is a nice ftp solution.

I think it will be a real challenge for programmers to beat a profession go player. Go is more complex than chess. You not only have to kill ur opponents stones. You have to gain terrority as well (like chessmate in chess but it isn't quite clear whether you have gain those terrorities until the game end.) Another thing is go board is bigger (19x19 whereas chess is 8x8) and has more moves. so computer has to calculate more if u are gonna use simple tree search algorithms like deepblue does. . Go game start with 361 empty spaces(~200 moves). With current go algorithms, the performance requirements for a chess-like approach to Go can be estimated as 10^27 times greater than that for computer chess. To put it simply u cant brute force a go game yet(not with current computation power and algorithms). Check out this and this

• Paper Prototyping. The basic idea here is to do a quick mockup of the user interface with paper, and then to stick real users in front of it and test it. Why? Because it's fast, cheap, and effective. Rather than spending weeks on features that users might not need or understand, you can do it in a single day.
  
• Contextual Design, by Holtzblatt and Beyer. This book looks at simple techniques for observing how end-users do their work, and then using those observations in the development of high quality user interfaces
  
• Design of Everyday Things, by Norman. The classic book on why design is important, and some guidelines for design.
  
• The Design of Sites, by van Duyne, Landay, and Hong. Ok, a shameless plug, this is my book on principles, processes, and patterns for web site design. We cover things like rapid prototyping, field studies, human capabilities, as well as 90 different user interface design patterns.
  
• And lastly, here is the website for a 3-day Human-Computer Interaction Course that I co-taught last summer. We have our syllabus and all of our slides online, for free.
  

• Take some time to understand the end-users, their tasks, their tools, and their social organization. Try to see things from their point of view.
  
• Do several iterations of rapid prototyping and testing before building the real thing. It will help tremendously.
  

>The problem is that to date, nobody in the OSS world has done so.

Well, in that case, I suggest we wrap this with quicktime.

Then we'll have the OSS QT codec everyone wants. Not to mention lossless video that's still compressed better than sound has ever been. Sweet.

...did we as a whole ever evolve out of the swamps???

I really do need to step up my efforts vis a vis setathome so I can finally get off this goshforsakenrock!

So does this have any advantages (political or technical) over the Integral Fast Reactor?

Breeders produce a lot.

The "waste" which is U-235 depleted but plutonium enriched must be further processed to produce weapons-grade material.

This is all-important for making a bomb. U-235 has a half-life of around 700 million years, and making a bomb with it is easy: squeeze together a prompt-supercritical mass, and wait a few milliseconds. Pu-239 is tricky, because its half-life is only about 25000 years and you have very little time to get it into a prompt-supercritical configuration before a spontaneous fission starts the reaction going. If the reaction starts too soon, the bomb blows itself apart into a sub-critical configuration before releasing much energy and all you have is a fizzle. Now imagine dealing with a substantial fraction of Pu-240 (half-life 6564 years or Pu-241 (half-life 14 years).

Bomb-grade material is made in special reactors which allow the fuel to be irradiated relatively briefly at a low level, and then removed and processed to remove the plutonium. This is specifically to avoid the production of enough higher isotopes of plutonium to be a problem. The stuff coming out of a power reactor after a full fuel cycle is dirty as hell, but amateur proliferators are not going to be able to make a serious bomb (as opposed to dirty weapon) out of it. This is why we had few objections to building pressurized-water reactors for North Korea; they are essentially proliferation-proof.

For 25 years we have banned reprocessing even to the level needed for use as fuel because of the concern is could be stolen and further enriched.

I think we should have gone with the Integral Fast Reactor, but it seems to have succumbed to the fundamentalist anti-nukes (who probably couldn't figure out that there are medical and explosive grades of nitroglycerine either...).

I figured that he was too chicken to show up on a regular day when classes were in session and the staff was on site.

From: "Nicholas Weaver"
Date: Fri, 31 Jan 2003 6:09 PM
To: bugtraq@securityfocus.com
Subject: The Spread of the Sapphire/Slammer SQL Worm
We have completed our preliminary analysis of the spread of the Sapphire/Slammer SQL worm. This worm required roughly 10 minutes to spread worldwide making it by far the fastest worm to date. In the early stages the worm was doubling in size every 8.5 seconds. At its peak, achieved approximately 3 minutes after it was released, Sapphire scanned the net at over 55 million IP addresses per second. It infected at least 75,000 victims and probably considerably more.

This remarkable speed, nearly two orders of magnitude faster than Code Red, was the result of a bandwidth-limited scanner. Since Sapphire didn't need to wait for responses, each copy could scan at the maximum rate that the processor and network bandwidth could support.

There were also two noteworthy bugs in the pseudo-random number generator which complicated our analysis and limited our ability to estimate the total infection but did not slow the spread of the worm.

The full analysis is available at

• http://www.caida.org/analysis/security/sapphire/
• http://www.silicondefense.com/sapphire/
• http://www.cs.berkeley.edu/~nweaver/sapphire/

David Moore, CAIDA & UCSD CSE
Vern Paxson, ICIR & LBNL
Stefan Savage, UCSD CSE
Colleen Shannon, CAIDA
Stuart Staniford, Silicon Defense
Nicholas Weaver, Silicon Defense and UC
Berkeley EECS

Hasn't this been thought of before? Small-scale parallel processing sounds a lot like the queues in SEDA:

SEDA Homepage

By the same logic, the search for the new world should have taken place with ships populated by Moors

Completely bizarre! No comparison like that makes sense at all. (Ignoring, of course, the archaic idea that some races' lives are less valued, because that's not what you meant)

A more correct analogy should be this: "the search for the new world should have taken place with ships populated by tough, sailing men, because obviously it was too dangerous for civilized thinkers and women to roam the seas". And they were absolutely correct. That's why Columbus went, and not Queen Isabella. The best man for the job. Or best robot, as the case may be.

They knew they were pioneers, and that not all pioneers return home safe.
The first shuttle crew were pioneers. Maybe even the 12th crew. But there have been hundreds of flights since then. Those guys? Truckdrivers.

They are heroes not because they died
Heroes are defined subjectively. You're a hero if the public thinks you are. Only Ilan Ramon was popular enough to be called "heroic" until yesterday. The rest of them were anonymous to the public, until death cast a limelight on them.

sitting pretty on Earth waiting for robots to solve our problems* is not the solution
Not just sitting around! We've got to build and maintain those robots. Tinkering in a lab or pondering at a computer isn't as glamorous as blasting into orbit, but it's where the real results will come from. The benefit:cost ratio for advancements in robots and computation simply dwarfs anything astronauts can give us.

read Penrose's "the emperor's new mind"
That book includes much interesting discussion, but by no means proves its thesis without granting some unsupportable leaps. (At least you didn't invoke his like-minded colleague Searle, whose argument is simply laughable)

Here's a few links to the next level of annoyances:

• http://www.cs.berkeley.edu/~nweaver/cdc.web/
• http://blanu.net/curious_yellow.html
• http://www.cs.berkeley.edu/~nweaver/warhol.html

Here's a few links to the next level of annoyances:

• http://www.cs.berkeley.edu/~nweaver/cdc.web/
• http://blanu.net/curious_yellow.html
• http://www.cs.berkeley.edu/~nweaver/warhol.html

Well I think the two extremes can be eliminated. That is, the ISS won't be trashed but the launch timeline and construction plan will be substantially affected.

This thing is so expensive and time-consuming that it would be impractical to just stop construction. I do believe a moratorium on STS launches while an investigation is undertaken is appropriate.

Yes, the risks are huge in manned spaceflight. Just the reason why we shouldn't be spending such risk on a politically motivated project like the ISS.

Well I think the two extremes can be eliminated. That is, the ISS won't be trashed but the launch timeline and construction plan will be substantially affected.

This thing is so expensive and time-consuming that it would be impractical to just stop construction. I do believe a moratorium on STS launches while an investigation is undertaken is appropriate.

Yes, the risks are huge in manned spaceflight. Just the reason why we shouldn't be spending such risk on a politically motivated project like the ISS.

Steady on - C++ is a high level language, at least by the standards people always used to use, and it does have a good range of bounds-checked containers (safe arrays and the like). It does allow you to do low-level stuff if you need to and you know what you're doing; the trouble is that too many people who don't really need the low-level memory access still don't use safe containers for spurious performance reasons. Yes, moving everyone to Lisp or ML or Perl or even Java would solve a lot of memory scribbling problems, but so would making sure everyone uses safe string libraries rather than char*s and safe container libraries rather than raw arrays.

Also you can use tools like Splint and CCured to check at compile time and at run time for unsafe memory accesses. As with so much else in C and C++, you can write safe and correct memory access, and you can shoot yourself in the foot. You just have to take care and do the right thing, or if you don't want to worry about being careful then switch to a different language. But switching is not the only way to solve the problem.

OceanStore is the UC Berkely project to do something like this, except a little more generalized. I run a freenet node and it isn't THAT slow. After the index built (had to leave it up for three days straight), the access are much quicker (prolly most of the data is local, now...ha). The slowdowns with FreeNet is in the Onion Routing and the encryption. Also, GNU has a project called GNUNet that has aims similar to FreeNets.

DIBS encrypts all data transmissions so that the peers you trade files with can not access your data.

Security
DIBS uses Gnu Privacy Guard (GPG) to encrypt and digitally sign all transactions. Thus you can be confident that even though you are sending your files to others for backup, your data will remain private. Furthermore, by using digital signatures, DIBS prevents others from impersonating you to store files with your peers.