Intentional abuse of participant hosts by projects
BOINC does nothing to prevent this (e.g. there is no 'sandboxing' of applications). Participants must understand that when they join a BOINC project, they are entrusting the security of their systems to that project.
Accidental abuse of participant hosts by projects
BOINC does nothing to prevent this.
That's a recipe for an exploit.
If they had some scheme where downloaded programs ran in a jail, and could do nothing but talk back to the appropriate server, that would be better. FreeBSD or NSA Secure Linux could support that.
Actually, this would be a good application for the NSA Secure Linux code, now that it's in the mainstream kernel.
The Bloomberg article offers some insight into the business strategy. The plan here is to make units that require a "fuel cartridge". "Fuel cartridges" contain just methanol and water, but
will have markups previously seen only for printer ink. Toshiba expects to make ten times as much on the "fuel cartridges" as they do on the fuel cells.
We have a Roomba at the Overbot shop. It's really dumb, doesn't clean well, gets tangled in wires, and
gets stuck under tables with chairs. It's basically useless.
One of the best is the Kensington Science Museum in London. Their collection is excellent, with the originals of many famous machines. Yet few people are interested in those items; all the crowds are around the dumbed-down "interpretive exhibits".
The Kensington Science Museum has early computers,
all the way back to Babbage. The first locomotive, the first lathe, Watt's first steam engine - they have it all. And that stuff you can at least figure out by looking at it.
Electronics is much worse to display. The Henry Ford Museum used to have display cases full of early electronics ("Capacitor, Cornell-Dublier, circa 1932"), ignored by almost everybody.
This guy apparently used an AOL-issued laptop to access AOL's data warehouse. Not only did he put the data on his laptop, his e-mails about how he was going to steal the data are on there. Some of the e-mails are in the court filing.
It's clear from reading them that this guy was not one of the brighter people at AOL.
Get the original article to your press contacts! This is a real story. It should be picked up by the Washington Post, New York Times, and a few news magazines. Get busy!
Hey, I know! Let's create a new operating system! But we're not actually going to do the work. We'll just put up a wiki, a blog, a web site, a mailing list handler, and let it all happen! It's open source, right?
The community does all the work!
The most recent e-mail SpamAssassin botched has this header:
X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on
main6.ezpublishing.com
X-Spam-Status: No, hits=1.1 required=4.0 tests=HTML_40_50,HTML_MESSAGE,
HTML_TAG_EXISTS_TBODY,MIME_HTML_ONLY,NO_REAL_NAME, RCVD_IN_SORBS
autolearn=no version=2.60
The mail content is "Major income on eBay", sent via a free account on Netster. If it can't recognize that as spam, it's not doing much.
I turned the threshold down from 5 to 4; at 5, it was useless. I tried "autolearn" at one time, but the "learning" algorithm choked on today's volumes of spam and used up 100MB on the server.
Yeah, maybe a later version is better. But if the product sucks, maybe it just sucks in all versions.
My hosting service, EZ Publishing, uses SpamAssassin. Their hosting service is fine, but incoming mail filtering is terrible. SpamAssassin is only filtering out about 25% of the incoming spam. I'm getting about 2000 spams per day after SpamAssassin filtering.
I use Netscape's Bayesian filter as a second tier, and that removes about 60% of the remaining spam.
SpamCop was better, until IronPort bought them and they went black-hat, with Bonded Spammer and the Spam Engine.
Exactly. That's what California enacted as law, and what the Direct Marketing Association successfully blocked by pushing the CAN-SPAM act through.
The California law made the "beneficiary" of the spam responsible for it. And anybody could sue. That would have made hiring a spammer very risky.
Broadly defining the "beneficiary" could go even further. The credit card service provider, and the bank behind them, could be held responsible
for spam if they processed a transaction resulting from spam. They profit from it, after all. A good lawyer could make the case now that they bear some responsibility, especially if they assist in any way in concealing the identity of the spammer.
We really need to go after the payment end of spam, not the sending end.
Putting executable code, even in an interpretive language like TCL, into DNS records is a terrible idea. That offers a whole new channel for attacks. A good one, too; the code would be executed without any user intervention, and sometimes it would be executed on servers.
When you read articles about fuel cells, note that some of these outfits don't have refueling worked out. Some are talking about "disposable" fuel cells.
"Disposable" fuel cells have to be compared against primary batteries, not rechargeable ones. Rechargable batteries typically have about half the energy density of primary batteries. So claiming a 2.5x improvement in battery life for a nonrechargeable system is not a win.
Ballard is further along than anybody else in larger fuel cells. Even they don't have much more than prototypes.
Their attempt to market a fuel cell under the Coleman brand was a failure.
The Coleman Powermate was launched with great fanfare in 2002, and never shipped. It's not clear what's wrong at Ballard. Their 1KW units should be providing backup power for cell phone sites and such, but it isn't happening.
Ballard uses hydrogen in their fuel cells. Despite all the hype about the "hydrogen economy", Praxair, which sells hydrogen for fuel cells, has this to say:
Clean burning and not considered an atmospheric pollutant, hydrogen is fast becoming the energy source of the future. Questions regarding cost, safety and infrastructure, however, need answers before hydrogen-fueled engines go into wide use.
Fuel cell grade hydrogen is specifically designed to be used as a fuel in fuel cell applications. It contains extremely low levels of impurities (e.g. ammonia (NH3), carbon monoxide (CO) and sulfur compounds) that can harm the catalyst-coated membranes inside the fuel cell.
It is supplied in high-pressure cylinders and can only be used by industrial customers, like factories, laboratories, universities, and military and government installations. Typically, industrial customers already use compressed gases as part of their daily activities. Its use requires adequate ventilation and/or monitoring systems appropriate to the size of the location, helping ensure the safety of personnel when non-air gases are present.
Lest we forget, back in 1963, the USAF's X-15 made a similar flight. Lifted under the wing of a B-52, the X-15 reached an altitude of 107960 meters.
The X-15 could do everything required to win the X-prize except carry three people. It reached 100km, and it was flown repeatedly, for a total of 199 X-15 flights of three aircraft.
Even if you want to use Microsoft, there are development tools for minimal kiosk systems. You just don't use a desktop XP system for that. Huge total cost of ownership if you do.
Kiosk systems should be running on something like QNX, not a desktop OS. People who insist on running kiosk systems on Microsoft software should use the Windows XP Embedded toolkit to build a minimal system.
They're lucky that Blaster was removable by remote control. A more effective virus would lock out any attempt to change system files.
Google is your friend. You can look up what reporters have written.
My general position is that I'll always talk to the working press, but I blow off "lifestyle" reporters. Running a DARPA Grand Challenge team, I get a fair amount of press interest. Some of it is wierd. Playboy and Men's Life contacted me for interviews. There were documentary producers, including one guy with an Alcatraz fixation. (He'd done five TV documentaries on Alcatraz.)
Most of this is just good surface chemistry, not "nanotechnology". Lately, we're seeing the term "nanotechnology" applied to fine powders, coatings, catalyst surfaces, and such. That's not about building large structures out of individual atoms; it's just surface treatments for ordinary bulk materials.
It gave in the Land: Microsoft clarifies asked for of explanation the director of the ITI
Editoria: Governments
18/Jun/2004 - 09:44
The Microsoft emitted a note today where he clarifies the episode of the explanation order that the company made Sergio Amadeu of the Silveira, president of the National Institute of Technology of Information (ITI). In interview to the magazine Capital Letter in the March month, Amadeu said that the company used "tactics of the gratuitous dealers" when supplying softwares programs of digital inclusion, what it would be a way to accustom the users.
The explanation order generated rumors of that the Microsoft would be processing managing of the ITI. It reads to follow the complete one of the note of the Microsoft, signed for Rinaldo Zangirolami, General Director of Legal and Corporative Subjects of the Microsoft Brazil:
Note of clarification
"we are not processing nobody, and the order of explanations is not related to a personal question.
The Microsoft continues engaged with a respectful and opened dialogue with the government, customers and the industry to address the necessities of the Brazilian economy and the community.
The Microsoft is present in the country has 14 years more than. Our commitment with the country is of long stated period. By means of ours 10,000 partners, 45,000 jobs are generated in Brazil and more than R$ 1 billion is collected in taxes annually.
Rinaldo General Managing Zangirolami of Legal and Corporative Subjects Microsoft Brazil"
Yeah. It's a 2700 pixel line scanner on a mechanical scanner, and a rather slow one. It takes 20 seconds to do a scan. Sports photography? No way.
I had something like this on my desk twenty years ago. It was the first commercial high-resolution scanning camera, made by Datacopy. Several thousand pixel line scanner, mirror driven by a stepping motor, a really good lens, and a big copy stand. B/W, no greyscale.
Slashdot Mirror
BOINC does nothing to prevent this (e.g. there is no 'sandboxing' of applications). Participants must understand that when they join a BOINC project, they are entrusting the security of their systems to that project.
Accidental abuse of participant hosts by projects
BOINC does nothing to prevent this.
That's a recipe for an exploit.
If they had some scheme where downloaded programs ran in a jail, and could do nothing but talk back to the appropriate server, that would be better. FreeBSD or NSA Secure Linux could support that. Actually, this would be a good application for the NSA Secure Linux code, now that it's in the mainstream kernel.
The Bloomberg article offers some insight into the business strategy. The plan here is to make units that require a "fuel cartridge". "Fuel cartridges" contain just methanol and water, but will have markups previously seen only for printer ink. Toshiba expects to make ten times as much on the "fuel cartridges" as they do on the fuel cells.
Look for strategies to prevent "refilling".
What's wrong with this picture?
Definitely do not run on any machine with important data.
We have a Roomba at the Overbot shop. It's really dumb, doesn't clean well, gets tangled in wires, and gets stuck under tables with chairs. It's basically useless.
The Kensington Science Museum has early computers, all the way back to Babbage. The first locomotive, the first lathe, Watt's first steam engine - they have it all. And that stuff you can at least figure out by looking at it.
Electronics is much worse to display. The Henry Ford Museum used to have display cases full of early electronics ("Capacitor, Cornell-Dublier, circa 1932"), ignored by almost everybody.
It's clear from reading them that this guy was not one of the brighter people at AOL.
It's the same bogus promises the telcos have been making for years. If only they were given unregulated monopoly power, they'd provide more bandwidth.
Here's SBC's announcement of fibre to the home in 2002. Where is that now?
Get the original article to your press contacts! This is a real story. It should be picked up by the Washington Post, New York Times, and a few news magazines. Get busy!
Next!
-
X-Spam-Checker-Version: SpamAssassin 2.60 (1.212-2003-09-23-exp) on
main6.ezpublishing.com
, RCVD_IN_SORBS
autolearn=no version=2.60
The mail content is "Major income on eBay", sent via a free account on Netster. If it can't recognize that as spam, it's not doing much.X-Spam-Status: No, hits=1.1 required=4.0 tests=HTML_40_50,HTML_MESSAGE,
HTML_TAG_EXISTS_TBODY,MIME_HTML_ONLY,NO_REAL_NAME
I turned the threshold down from 5 to 4; at 5, it was useless. I tried "autolearn" at one time, but the "learning" algorithm choked on today's volumes of spam and used up 100MB on the server.
Yeah, maybe a later version is better. But if the product sucks, maybe it just sucks in all versions.
I use Netscape's Bayesian filter as a second tier, and that removes about 60% of the remaining spam.
SpamCop was better, until IronPort bought them and they went black-hat, with Bonded Spammer and the Spam Engine.
The California law made the "beneficiary" of the spam responsible for it. And anybody could sue. That would have made hiring a spammer very risky.
Broadly defining the "beneficiary" could go even further. The credit card service provider, and the bank behind them, could be held responsible for spam if they processed a transaction resulting from spam. They profit from it, after all. A good lawyer could make the case now that they bear some responsibility, especially if they assist in any way in concealing the identity of the spammer.
We really need to go after the payment end of spam, not the sending end.
Putting executable code, even in an interpretive language like TCL, into DNS records is a terrible idea. That offers a whole new channel for attacks. A good one, too; the code would be executed without any user intervention, and sometimes it would be executed on servers.
"Disposable" fuel cells have to be compared against primary batteries, not rechargeable ones. Rechargable batteries typically have about half the energy density of primary batteries. So claiming a 2.5x improvement in battery life for a nonrechargeable system is not a win.
Ballard is further along than anybody else in larger fuel cells. Even they don't have much more than prototypes. Their attempt to market a fuel cell under the Coleman brand was a failure. The Coleman Powermate was launched with great fanfare in 2002, and never shipped. It's not clear what's wrong at Ballard. Their 1KW units should be providing backup power for cell phone sites and such, but it isn't happening.
Ballard uses hydrogen in their fuel cells. Despite all the hype about the "hydrogen economy", Praxair, which sells hydrogen for fuel cells, has this to say:
Fuel cell grade hydrogen is specifically designed to be used as a fuel in fuel cell applications. It contains extremely low levels of impurities (e.g. ammonia (NH3), carbon monoxide (CO) and sulfur compounds) that can harm the catalyst-coated membranes inside the fuel cell.
It is supplied in high-pressure cylinders and can only be used by industrial customers, like factories, laboratories, universities, and military and government installations. Typically, industrial customers already use compressed gases as part of their daily activities. Its use requires adequate ventilation and/or monitoring systems appropriate to the size of the location, helping ensure the safety of personnel when non-air gases are present.
The X-15 could do everything required to win the X-prize except carry three people. It reached 100km, and it was flown repeatedly, for a total of 199 X-15 flights of three aircraft.
Even if you want to use Microsoft, there are development tools for minimal kiosk systems. You just don't use a desktop XP system for that. Huge total cost of ownership if you do.
Although the model that was 4' wide and actually worked was kind of cool.
They tried this on a spokesman for the National Rifle Association. It backfired. "I came *this* close to pulling a knife on his dumbass. Had my hand all the way in my pocket. Paused there, thought better of it."
They're lucky that Blaster was removable by remote control. A more effective virus would lock out any attempt to change system files.
It is a direction descendant of SysIII with some bits of SysV unixware code added in.
Actually, BSD started from V7 Unix. System III came later. This is one of the reasons that signal handling semantics are such a mess.
My general position is that I'll always talk to the working press, but I blow off "lifestyle" reporters. Running a DARPA Grand Challenge team, I get a fair amount of press interest. Some of it is wierd. Playboy and Men's Life contacted me for interviews. There were documentary producers, including one guy with an Alcatraz fixation. (He'd done five TV documentaries on Alcatraz.)
Good technology, just too much hype.
Taxpayers to NASA: before blithering about terraforming Mars, build a new launch vehicle that works.
- It gave in the Land: Microsoft clarifies asked for of explanation the director of the ITI
Now we need to hear what Amadeu has to say.Editoria: Governments
18/Jun/2004 - 09:44
The Microsoft emitted a note today where he clarifies the episode of the explanation order that the company made Sergio Amadeu of the Silveira, president of the National Institute of Technology of Information (ITI). In interview to the magazine Capital Letter in the March month, Amadeu said that the company used "tactics of the gratuitous dealers" when supplying softwares programs of digital inclusion, what it would be a way to accustom the users.
The explanation order generated rumors of that the Microsoft would be processing managing of the ITI. It reads to follow the complete one of the note of the Microsoft, signed for Rinaldo Zangirolami, General Director of Legal and Corporative Subjects of the Microsoft Brazil:
Note of clarification
"we are not processing nobody, and the order of explanations is not related to a personal question.
The Microsoft continues engaged with a respectful and opened dialogue with the government, customers and the industry to address the necessities of the Brazilian economy and the community.
The Microsoft is present in the country has 14 years more than. Our commitment with the country is of long stated period. By means of ours 10,000 partners, 45,000 jobs are generated in Brazil and more than R$ 1 billion is collected in taxes annually.
Rinaldo General Managing Zangirolami of Legal and Corporative Subjects Microsoft Brazil"
Source: Land Computer science
I had something like this on my desk twenty years ago. It was the first commercial high-resolution scanning camera, made by Datacopy. Several thousand pixel line scanner, mirror driven by a stepping motor, a really good lens, and a big copy stand. B/W, no greyscale.