The point of the article is that/dev/zero is not good enough, since it's possible to read what's "underneath" the zeros if you have the right equipment. Writing pseudorandom data is slightly better, and writing multiple passes of random data is even better. However, as someone else once said, the only truely secure deletion method is a smelter.:-)
According to the version 4.0 registration template, section B:
4) Payment: Payment is due to Network Solutions within thirty (30) days from the date of the invoice.
So, it seems entirely possible to 'squat' on a pile of names for 30 days, then pay for the ones you've re-sold and let the rest expire.
The article is very unclear about which domains were deleted, but if it was just ones that were more than 30 days overdue on payment, that's fine with me. I also wouldn't mind Internic moving to a prepayment-required system, or like most businesses where a corporate customer (like an ISP) could get a credit account if they submitted appropriate financial records and signed a "no squatting" agreement.
Unlike some people, I've never had a real problem with Internic (I've registered 4 domains in the last few years).
There was very little in Win95 that wasn't in the Amiga or some other OS at least 5 years before. Long filenames, 32-bit, preemptive multitasking, desktop-based file manager, auto-configuring hardware,... All old news. The only innovations of Win95 were in the areas of marketing and licensing.
Yes, M$ made a big leap from CP/M, er, PC-DOS, er, MS-DOS to Win95, but when you're starting from that far in the hole, even someone else's 5-year-old technology is going to look good.
None whatsoever; "E=mc^2" will continue to be a widely-parroted simplification of the more general definition of the total energy of a particle, (E^2) = (m0c^2)^2 + (pc)^2, where m0 is the particle's rest mass and p is its momentum, and c is the speed of light _in vacuum_ (which hasn't changed). This news story is about some unique optical properties of a new type of matter, but it doesn't affect electromagnetism or relativity theories.
E=mc^2 may have been a big deal back when the "wireless" and the "horseless carriage" were on the top of every geek's Christmas wish-list, but these days it's just a routine part of physics like Ohm's law (V=IR) is in electronics. It's not a big deal. The rest of Einstein's work is much more interesting (general relativity, stimulated emission (lasers), the photoelectric effect, etc).
"E=mc^2" may look good on a T-shirt, but apart from that there's little reason why this particular Thought McNugget deserves to be so widely quoted when so much more of the really interesting physics is completely ignored.
By the way, nuclear fission was observed in the lab by people like Otto Hahn, without the need for "E=mc^2" to explain where the energy came from. Nuclear bombs are just a matter of banging rocks together, once you have the right rocks - who cares if the total mass of the bomb fragments is slightly less after the blast? Your city still ends up as kitty litter thanks to the binding energy in the nucleus.
To a decent approximation, the speed of light (measured locally) is not affected by gravitational fields - the only thing that changes is the apparant frequency of the light as seen by different observers (known as gravitational redshift).
As others have noted, the speed of light in a non-vacuum medium is slower by a ratio known as the refractive index, which is less than 2.5 for most materials (I think diamond is 2.42 or something) and is exactly 1 in a vacuum. So, the real news story here is the creation of a material with a *huge* refractive index (which is really neat).
Depending on how you manipulate the equations, it *is* possible to have a refractive index less than 1 (and therefore a "speed of light" faster than in vacuum). These conditions can occur in an ionized gas, such as the layers of our atmosphere which bounce radio signals around the globe. However, it also turns out that you can't transmit any information through this medium faster than in vacuum, so it's really just a mathematical curiosity.
fbcon is *extremely* useful for anyone with a new video card (e.g. Voodoo Banshee) for which there is not yet an X server. Yes, it switches to a single resolution (1280x1024 for me) at boot-up and stays there, but I rarely need to change resolution anyway (and "text mode" at 1280x1024 is actually quite nice).
fbcon could also be used as the base for a prettier-looking installation/setup utility, by giving you nice graphics and windows before trying to probe for the details of your graphics card (sorry, but the RedHat install screens are rather DOS-ugly). Given my Amiga heritage, I guess I'm just not in love with "text mode" in the first place.
We were expecting to have Redhat and SuSE CDs as well as the Caldera ones, but these did not arrive in time for Comdex. So, we went with what we had. We even talked to a CD-duplicating place a couple of booths down, but no luck.
>Is there some reason to warrant the incredible amount of money they're charging for it?
It looks like with the GS you're paying for software and setup convenience (definitely targeted for corporate users). The DM (same hardware, DIY software) is much better priced, and does have some advantages over a bare x86 system:
- Integrated video and audio in/out - Low power (add a $30 gel-cell for a UPS) - Quieter (good for 24x7 in your bedroom), though the fan is still a bit too loud - Small real-estate requirements - IrDA and IR remote control interface - 2 Ethernet ports (great home server for ADSL/cable) - Coolness Factor - Help to encourage other companies to support Linux (this should be worth at least $50) - New architectures tend to expose hidden bugs in software and encourage greater portability, thus helping the rest of the free software community.
However, the price is still a bit on the high side, and has gone up since I got mine in the summer. Presumably Corel will bring the price down when they want to sell more units, but if the current price sells enough to keep their manufacturing near capacity, they will probably keep it there (I have no idea what their actual sales/manufacturing numbers are, however).
But then, I paid over CDN$500 for a 14.4 modem a few years back, and I remember when even a 4M SIMM was a few hundred. All computer hardware is obscenely cheap these days.
>The Qube has a backup system (backup to workstation disk), while the NetWinder doesn't (at least, not that they mention)
With the Netwinder, you'd just use any normal network backup method (rdump, tar to an NFS-mounted drive, etc). I don't know if Corel has any GUI frontend for this, but it wouldn't be hard to add (maybe I'll put it on my list of Netwinder projects).
Slashdot Mirror
The point of the article is that /dev/zero is not good enough, since it's possible to read what's "underneath" the zeros if you have the right equipment. Writing pseudorandom data is slightly better, and writing multiple passes of random data is even better. However, as someone else once said, the only truely secure deletion method is a smelter. :-)
According to the version 4.0 registration template, section B:
4) Payment: Payment is due to Network Solutions within thirty (30) days from the date of the invoice.
So, it seems entirely possible to 'squat' on a pile of names for 30 days, then pay for the ones you've re-sold and let the rest expire.
The article is very unclear about which domains were deleted, but if it was just ones that were more than 30 days overdue on payment, that's fine with me. I also wouldn't mind Internic moving to a prepayment-required system, or like most businesses where a corporate customer (like an ISP) could get a credit account if they submitted appropriate financial records and signed a "no squatting" agreement.
Unlike some people, I've never had a real problem with Internic (I've registered 4 domains in the last few years).
There was very little in Win95 that wasn't in the Amiga or some other OS at least 5 years before. Long filenames, 32-bit, preemptive multitasking, desktop-based file manager, auto-configuring hardware, ... All old news. The only innovations of Win95 were in the areas of marketing and licensing.
Yes, M$ made a big leap from CP/M, er, PC-DOS, er, MS-DOS to Win95, but when you're starting from that far in the hole, even someone else's 5-year-old technology is going to look good.
>What effect does this have on E=mc^2?
-----BEGIN RANT BLOCK-----
version 2.10a
None whatsoever; "E=mc^2" will continue to be a widely-parroted simplification of the more general definition of the total energy of a particle, (E^2) = (m0c^2)^2 + (pc)^2, where m0 is the particle's rest mass and p is its momentum, and c is the speed of light _in vacuum_ (which hasn't changed). This news story is about some unique optical properties of a new type of matter, but it doesn't affect electromagnetism or relativity theories.
E=mc^2 may have been a big deal back when the "wireless" and the "horseless carriage" were on the top of every geek's Christmas wish-list, but these days it's just a routine part of physics like Ohm's law (V=IR) is in electronics. It's not a big deal. The rest of Einstein's work is much more interesting (general relativity, stimulated emission (lasers), the photoelectric effect, etc).
"E=mc^2" may look good on a T-shirt, but apart from that there's little reason why this particular Thought McNugget deserves to be so widely quoted when so much more of the really interesting physics is completely ignored.
By the way, nuclear fission was observed in the lab by people like Otto Hahn, without the need for "E=mc^2" to explain where the energy came from. Nuclear bombs are just a matter of banging rocks together, once you have the right rocks - who cares if the total mass of the bomb fragments is slightly less after the blast? Your city still ends up as kitty litter thanks to the binding energy in the nucleus.
------END RANT BLOCK------
To a decent approximation, the speed of light (measured locally) is not affected by gravitational fields - the only thing that changes is the apparant frequency of the light as seen by different observers (known as gravitational redshift).
As others have noted, the speed of light in a non-vacuum medium is slower by a ratio known as the refractive index, which is less than 2.5 for most materials (I think diamond is 2.42 or something) and is exactly 1 in a vacuum. So, the real news story here is the creation of a material with a *huge* refractive index (which is really neat).
Depending on how you manipulate the equations, it *is* possible to have a refractive index less than 1 (and therefore a "speed of light" faster than in vacuum). These conditions can occur in an ionized gas, such as the layers of our atmosphere which bounce radio signals around the globe. However, it also turns out that you can't transmit any information through this medium faster than in vacuum, so it's really just a mathematical curiosity.
>if NT systems are commonly used in POS systems?
It probably depends how you decode the acronym "POS"...
fbcon is *extremely* useful for anyone with a new video card (e.g. Voodoo Banshee) for which there is not yet an X server. Yes, it switches to a single resolution (1280x1024 for me) at boot-up and stays there, but I rarely need to change resolution anyway (and "text mode" at 1280x1024 is actually quite nice).
fbcon could also be used as the base for a prettier-looking installation/setup utility, by giving you nice graphics and windows before trying to probe for the details of your graphics card (sorry, but the RedHat install screens are rather DOS-ugly). Given my Amiga heritage, I guess I'm just not in love with "text mode" in the first place.
We were expecting to have Redhat and SuSE CDs as well as the Caldera ones, but these did not arrive in time for Comdex. So, we went with what we had.
We even talked to a CD-duplicating place a couple of booths down, but no luck.
-Mike (VanLUG member)
>Is there some reason to warrant the incredible amount of money they're charging for it?
It looks like with the GS you're paying for software and setup convenience (definitely targeted for corporate users). The DM (same hardware, DIY software) is much better priced, and does have some advantages over a bare x86 system:
- Integrated video and audio in/out
- Low power (add a $30 gel-cell for a UPS)
- Quieter (good for 24x7 in your bedroom), though the fan is still a bit too loud
- Small real-estate requirements
- IrDA and IR remote control interface
- 2 Ethernet ports (great home server for ADSL/cable)
- Coolness Factor
- Help to encourage other companies to support Linux (this should be worth at least $50)
- New architectures tend to expose hidden bugs in software and encourage greater portability, thus helping the rest of the free software community.
However, the price is still a bit on the high side, and has gone up since I got mine in the summer. Presumably Corel will bring the price down when they want to sell more units, but if the current price sells enough to keep their manufacturing near capacity, they will probably keep it there (I have no idea what their actual sales/manufacturing numbers are, however).
But then, I paid over CDN$500 for a 14.4 modem a few years back, and I remember when even a 4M SIMM was a few hundred. All computer hardware is obscenely cheap these days.
>The Qube has a backup system (backup to workstation disk), while the NetWinder doesn't (at least, not that they mention)
With the Netwinder, you'd just use any normal network backup method (rdump, tar to an NFS-mounted drive, etc). I don't know if Corel has any GUI frontend for this, but it wouldn't be hard to add (maybe I'll put it on my list of Netwinder projects).