When Microsoft took over Hotmail, it took them years and many failed efforts to switch it over from unix to Windows. I'm not actually convinced they ever fully pulled it off.
20 years of Windows, and the more expert we are in either/both Windows and unix (or Linux), the less likely we are to use Windows technology for our most important development. Especially stuff that's less than 10 years in the field.
NASA does collect money from technology it produces and distributes, but its real ROI is in the general US economy (and beyond). Its US ROI is well documented, ever since the big investments of the 1960s. One sample reports only NASA Life Sciences R&D ROI on $64M 1972-1997 is $1.5B. That's a better than 20-fold ROI, unheard of in other industries, especially at that scale. And that's not even including military returns from rocketry/telemetry R&D and actual launches/deployment. Plus the diplomacy and science recruitment growth, which is hard to measure, but has a dollar cost by other means.
NASA is cheap, and hugely beneficial. Not to mention the various strategic benefits from staying ahead of our foreign propaganda competitors like Russia, China, India and Japan.
Finally the government operating a natural role as the infrastructure monopoly investing in the startup of private American industry. NASA overall has produced probably the best ROI on any US government investment in the 20th Century. And the US space industry is at the crossroads for going live, both positioned to deliver services and facing foreign competition.
Let's spend hundreds of billions of the dollars that we currently mostly waste on Pentagon corporate welfare that makes the US feared around the world instead spent on NASA investment in infrastructure to support private corporations. Let's get the US aerospace industry to compete by raising private investment to fund competitions for achieving goals like Lunar power stations and manned Martian research bases. Let's get NASA to become solely a policy, design, testing and certification agency, and subsidize American corporations to pass our highest criteria ahead of foreign ones.
The OS should send an alert when the CD drive is opened, and assume the CD has been violated. A more secure kernel might allow the CD drive to open only in single user mode, for OS upgrades. A more secure system might not allow the CD drive to open at all, requiring upgrades to open the CD on a separate machine, and coldswapping the CD/drive only by those with access to powercycle the machine and upgrade its hardware.
There's still not a lot of good protections from physical access to the machine. My suggestions are useful in protecting from remote attacks, like most defense strategies.
I don't want to give up my funny (polyhedral) dice. I started playing D&D with the "Basic Set", which included cutout "chits" bound into the rulebooks. We had to pick numbers from cups until the game was popular enough that a hobby store within 50 miles could mailorder some dice. I never went back. Throwing dice is the perfect physical connection between verbal roleplaying and throwing spells/punches/ropes. "Saving throw" is so much more real with a real throw. Lots of chance should be automated to smooth flow of the game and keep DM decisions secret. But I want the option of throwing some dice in the game, even if just to burn some entropy to show it who's playing.
Maybe a "Real Funny Interface Dice" controller, where a couple of RFID sensors detect the 3D position of a couple of RFID chips in each die? Or maybe a "dicepad" that images the bottoms of the dice after they roll. Just as long as I can throw some nuggets to pick a number, I'll be able to keep the moves I learned as a kid.
One of the best features of a secure Live CD is that the read-only media prevents attacks from writing to the stored OS (on CD). I'd love to see a virtualization system that reloads the OS from the CD every so often (hours, minutes, seconds) and switches all processes to the new, more trustworthy instance.
Maybe a safer system will just reload a single watchdog instance from the CD, which checks itself against the other running instances.
Any difference would send an alarm out of the system.
Of course, the virtualization layer itself needs authenticity checks. But that might be possible against a CD image, and in any case would be no less secure than without this system I'm describing.
Those computers offer journalists more than just faster . They can encrypt all their data. Which would require a judge order they produce the key in a formal decision, rather than leave it up to the cops, who always snoop in whatever they touch, and usually blab.
Newspapers saved hundreds of millions of dollars in paper process, and often make millions by publishing digitally. They should reinvest some of that gain in protecting their product from destruction by the competition.
If the industry joined open source efforts to encrypt messaging and research workflow in popular tools, we'd all have a lot more protection, and a lot cheaper, too.
In premed, I tried to produce a 4D graphic animation of a virus being reproduced from an RNA, protein by protein, in a host cell. My advisor told me it would be too hard - I should use the newfangled computer to instead process huge (1980s megabyte) datasets of genetic drift data. I thought he was too shortsighted based on his pessimism for the animation, so I just made PR animations of cameraviews flying around a double helix, which I reused as a lightshow for the psychedelic rock band I lived with.
Turns out I'd have spent at least 20 years on that animation. And that genetic drift project would have trapped me in COBOL programmer Mesozoic. The lightshow, however, had immediate effect on my chromasomes.
I don't want to have to heft the extra computing and batteries on my wrist. Why does the wristputer have to do anything but display and expose buttons/knobs/touchscreen? WiFi to a belt CPU and belt WWAN, wireless earphones. Maybe the entire belt volume dedicated to the non-UI parts, as well as battery, distributed around the waist.
More research will make GPGPU ever more "general". As will transforming computational patterns into dataflow processing, especially linear equations (vector processing).
Meanwhile, I just want to stuff a PC with cheap videocards and make an MP3 compression supercomputer running Linux.
Remember that those "graphics cards" are high-performance processors that can perform more "general purpose" tasks: GPGPU. I'd love to see a Linux kernel that is basically just a task scheduler for distributing computing among a network of GPGPU cards on these multiple PCI buses. Scalable desktop supercomputers running Linux apps.
The CIA is not supposed to be a "military agency". It was originally supposed to be an assassination agency supporting US military overthrow of enemy governments during WWII, when it was the "Office of Special Services". When it was converted to the CIA, its postwar role was supposed to be foreign intelligence gathering, with domestic operations confined to centrally processing government intelligence information. None of its operations are supposed to be military, as in tactical violence against strategic targets to support government policies.
The CIA is not supposed to trade guns for cocaine or peddle them in the US, or work with the mafia to fund operations secret from Congress, either.
I find that the more info I have (metainfo) about some info, the more of the mere info I can remember. I also generally find I can remember more of the metainfo as well. The more "complete picture" I have, the more I can remember it. Especially if multiply-interelated info offers more "paths" from one memory to another, I can recollect along the paths. Maybe it's just a matter of collecting more info overall to apply my "loss percentage" to, so I can remember more items from the larger population, while forgetting even more that I don't notice. The meta/info distinction lets me prioritize the info at the expense of the metainfo.
All those clarifications that Ekiga is a client to Asterisk's server are helpful corrections to my post's wrong assumptions. But it's not really so clear - lots of telephony/messaging/collaboration features can be delivered either by the client or the server, or in combination. So it would still be helpful to see which, if any, features they do have in common, and how successful was their placement on their respective sides of the divide.
It also would be helpful to turn my question around, and see how well these two packages work together where their features are complementary.
If I loan them my download inside a system that won't play it for me until they've can't play it again, like a streaming server that only plays one stream from a single copy at a time, then it's just as fair as loaning him my CD.
I'm sure an international legal ban on nanoweapons and "nanomalware" will keep them stopped. Kim Jong Il, Iran's theocrats, America's theocrats and their fellow capitalists, warmongers and nutjobs all respect international safety/security laws so well, now that we're all joined in the harmony of global peace and prosperity.
Wouldn't it be great if a judge finally just says "this is bullshit", and denounces the entire kangaroo patent system? Just declares that the current system is only tangentially related to the Constitutional provision to "promote progress in the useful arts and sciences", that only temporary monopolies are granted to recoup inventors' costs, that the whole rotten edifice built on those simple, sound principles is null and void?
And wouldn't it be nice if I could let a friend listen to my downloaded song the same way they can borrow my CD?
Slashdot Mirror
When Microsoft took over Hotmail, it took them years and many failed efforts to switch it over from unix to Windows. I'm not actually convinced they ever fully pulled it off.
20 years of Windows, and the more expert we are in either/both Windows and unix (or Linux), the less likely we are to use Windows technology for our most important development. Especially stuff that's less than 10 years in the field.
NASA does collect money from technology it produces and distributes, but its real ROI is in the general US economy (and beyond). Its US ROI is well documented, ever since the big investments of the 1960s. One sample reports only NASA Life Sciences R&D ROI on $64M 1972-1997 is $1.5B. That's a better than 20-fold ROI, unheard of in other industries, especially at that scale. And that's not even including military returns from rocketry/telemetry R&D and actual launches/deployment. Plus the diplomacy and science recruitment growth, which is hard to measure, but has a dollar cost by other means.
NASA is cheap, and hugely beneficial. Not to mention the various strategic benefits from staying ahead of our foreign propaganda competitors like Russia, China, India and Japan.
Senator Proxmire, is that you?
Finally the government operating a natural role as the infrastructure monopoly investing in the startup of private American industry. NASA overall has produced probably the best ROI on any US government investment in the 20th Century. And the US space industry is at the crossroads for going live, both positioned to deliver services and facing foreign competition.
Let's spend hundreds of billions of the dollars that we currently mostly waste on Pentagon corporate welfare that makes the US feared around the world instead spent on NASA investment in infrastructure to support private corporations. Let's get the US aerospace industry to compete by raising private investment to fund competitions for achieving goals like Lunar power stations and manned Martian research bases. Let's get NASA to become solely a policy, design, testing and certification agency, and subsidize American corporations to pass our highest criteria ahead of foreign ones.
Let's take it to the stars!
Moderation -1
100% Offtopic
Suggesting how to interface real dice with the online D&D game is "Offtopic" to "Dungeons and Dragons Online Impressions"?
GIVE SWORD TO TROLLMOD at high velocity, edge first.
The OS should send an alert when the CD drive is opened, and assume the CD has been violated. A more secure kernel might allow the CD drive to open only in single user mode, for OS upgrades. A more secure system might not allow the CD drive to open at all, requiring upgrades to open the CD on a separate machine, and coldswapping the CD/drive only by those with access to powercycle the machine and upgrade its hardware.
There's still not a lot of good protections from physical access to the machine. My suggestions are useful in protecting from remote attacks, like most defense strategies.
I don't want to give up my funny (polyhedral) dice. I started playing D&D with the "Basic Set", which included cutout "chits" bound into the rulebooks. We had to pick numbers from cups until the game was popular enough that a hobby store within 50 miles could mailorder some dice. I never went back. Throwing dice is the perfect physical connection between verbal roleplaying and throwing spells/punches/ropes. "Saving throw" is so much more real with a real throw. Lots of chance should be automated to smooth flow of the game and keep DM decisions secret. But I want the option of throwing some dice in the game, even if just to burn some entropy to show it who's playing.
Maybe a "Real Funny Interface Dice" controller, where a couple of RFID sensors detect the 3D position of a couple of RFID chips in each die? Or maybe a "dicepad" that images the bottoms of the dice after they roll. Just as long as I can throw some nuggets to pick a number, I'll be able to keep the moves I learned as a kid.
One of the best features of a secure Live CD is that the read-only media prevents attacks from writing to the stored OS (on CD). I'd love to see a virtualization system that reloads the OS from the CD every so often (hours, minutes, seconds) and switches all processes to the new, more trustworthy instance.
Maybe a safer system will just reload a single watchdog instance from the CD, which checks itself against the other running instances.
Any difference would send an alarm out of the system.
Of course, the virtualization layer itself needs authenticity checks. But that might be possible against a CD image, and in any case would be no less secure than without this system I'm describing.
Those computers offer journalists more than just faster . They can encrypt all their data. Which would require a judge order they produce the key in a formal decision, rather than leave it up to the cops, who always snoop in whatever they touch, and usually blab.
Newspapers saved hundreds of millions of dollars in paper process, and often make millions by publishing digitally. They should reinvest some of that gain in protecting their product from destruction by the competition.
If the industry joined open source efforts to encrypt messaging and research workflow in popular tools, we'd all have a lot more protection, and a lot cheaper, too.
Basic software production rule: first get it to work at all. Then optimize it.
In premed, I tried to produce a 4D graphic animation of a virus being reproduced from an RNA, protein by protein, in a host cell. My advisor told me it would be too hard - I should use the newfangled computer to instead process huge (1980s megabyte) datasets of genetic drift data. I thought he was too shortsighted based on his pessimism for the animation, so I just made PR animations of cameraviews flying around a double helix, which I reused as a lightshow for the psychedelic rock band I lived with.
Turns out I'd have spent at least 20 years on that animation. And that genetic drift project would have trapped me in COBOL programmer Mesozoic. The lightshow, however, had immediate effect on my chromasomes.
I don't want to have to heft the extra computing and batteries on my wrist. Why does the wristputer have to do anything but display and expose buttons/knobs/touchscreen? WiFi to a belt CPU and belt WWAN, wireless earphones. Maybe the entire belt volume dedicated to the non-UI parts, as well as battery, distributed around the waist.
Moderation +3
60% Interesting
20% Troll
20% Informative
OK, TrollMod, try googling for (cia "iran/contra" cocaine).
I'm not surprised when an Anonymous rightwing Coward denies the truth about their conspiracy.
More research will make GPGPU ever more "general". As will transforming computational patterns into dataflow processing, especially linear equations (vector processing).
Meanwhile, I just want to stuff a PC with cheap videocards and make an MP3 compression supercomputer running Linux.
Thanks for the reminder that "the NSA" in various capacities is reading these threads ;).
Or Ecstasy during the Terror War.
Remember that those "graphics cards" are high-performance processors that can perform more "general purpose" tasks: GPGPU. I'd love to see a Linux kernel that is basically just a task scheduler for distributing computing among a network of GPGPU cards on these multiple PCI buses. Scalable desktop supercomputers running Linux apps.
The CIA is not supposed to be a "military agency". It was originally supposed to be an assassination agency supporting US military overthrow of enemy governments during WWII, when it was the "Office of Special Services". When it was converted to the CIA, its postwar role was supposed to be foreign intelligence gathering, with domestic operations confined to centrally processing government intelligence information. None of its operations are supposed to be military, as in tactical violence against strategic targets to support government policies.
The CIA is not supposed to trade guns for cocaine or peddle them in the US, or work with the mafia to fund operations secret from Congress, either.
I would demand less users.
I find that the more info I have (metainfo) about some info, the more of the mere info I can remember. I also generally find I can remember more of the metainfo as well. The more "complete picture" I have, the more I can remember it. Especially if multiply-interelated info offers more "paths" from one memory to another, I can recollect along the paths. Maybe it's just a matter of collecting more info overall to apply my "loss percentage" to, so I can remember more items from the larger population, while forgetting even more that I don't notice. The meta/info distinction lets me prioritize the info at the expense of the metainfo.
All those clarifications that Ekiga is a client to Asterisk's server are helpful corrections to my post's wrong assumptions. But it's not really so clear - lots of telephony/messaging/collaboration features can be delivered either by the client or the server, or in combination. So it would still be helpful to see which, if any, features they do have in common, and how successful was their placement on their respective sides of the divide.
It also would be helpful to turn my question around, and see how well these two packages work together where their features are complementary.
If I loan them my download inside a system that won't play it for me until they've can't play it again, like a streaming server that only plays one stream from a single copy at a time, then it's just as fair as loaning him my CD.
Anyone have a side-by-side comparison of Ekiga/GnomeMeeting with Asterisk, at least on their common telephony/conferencing features?
I'm sure an international legal ban on nanoweapons and "nanomalware" will keep them stopped. Kim Jong Il, Iran's theocrats, America's theocrats and their fellow capitalists, warmongers and nutjobs all respect international safety/security laws so well, now that we're all joined in the harmony of global peace and prosperity.
Wouldn't it be great if a judge finally just says "this is bullshit", and denounces the entire kangaroo patent system? Just declares that the current system is only tangentially related to the Constitutional provision to "promote progress in the useful arts and sciences", that only temporary monopolies are granted to recoup inventors' costs, that the whole rotten edifice built on those simple, sound principles is null and void?
And wouldn't it be nice if I could let a friend listen to my downloaded song the same way they can borrow my CD?