If the UCSB admins were smart they would have conveniently posted information about how to make Windows 2000 Profesional reasonably secure.
Things like installing Service Pack 3, setting accounts correctly, banning the use of personal web servers on a client machine, and mandatory installation of a good antivirus and/or firewall program would have saved the UCSB sysadmins a lot of headaches.
Windows 2000 Professional with its Service Pack 3 and McAfee AntiVirus 7.0 (antivirus + firewall) is actually a reasonably secure OS setup.
Re:P4 did flop, for quite awhile
on
Itanium Problems
·
· Score: 2
Hold it right there.
Didn't Intel introduce the Northwood-core Pentium 4's starting with the 2.0 and 2.2 GHz versions first? I think increasing the L2 cache on the CPU die to a generous 512 KB did wonders for the CPU. It'll be very interesting to see what the Prescott-core Pentium 4's with its 1024 KB L2 cache does in terms of performance.
I don't think you realize what the Chinese are trying to do.
By developing their own CPU and operating system through official government sanction, it gives the government a way to effectively spy on Internet users because the government knows how everything works and will very likely use this knowledge to attempt such control. You are forgetting that mainland China is still in many ways an authoritarian state and the government is more than willing to spy on its own people to stamp out enemies of the state such as the Fulan Gong movement.
Does the book 1984 have any meaning to you? Mainland China is headed in that direction if government control of hardware and software technology has its way.
South Korea has a very high percentage of young people and thus more people in the work force.
And more importantly, a work force that is highly literate and have lots of technical skills, too. I saw a show on The Learning Channel about a month ago on Hyundai's amazing shipbuilding plant and it shows just how good their skills at shipbuilding have gotten.
I think Lester Thurow was wrong on Japan because their economic collapse showed that Japan's cultural norms could not accommodate the changes necessary to improve their economic systems.
Look at South Korea--after the horrid experience of the Asian financial crisis of 1997-1999 this country was willing to take drastic steps to improve its economic system; as a result the country is doing quite well indeed.
Here in the USA, the fact we're more than willing to make changes in our economic system to correct problems show why the USA will do well economically.
However, ATI was saved by the fact they had a good number of OEM contracts, which pretty much kept them going until the more modern Radeon models started to make serious inroads among upgrade customers. Yes, the Radeon 8500 was a bit late and the initial drivers for the board was a bit disappointing, but in the end ATI did finally release decent drivers and the Radeon 8500 was much-liked.
But frankly, I do think that 3dfx's inability to take on nVidia and ATI in the OEM market by 1999 and their ill-advised purchase of STB did them in.:-(
I still contend that the one-board solution was what really did in 3dfx.
Once nVidia developed the RIVA 128 and TNT/TNT2 chipsets, you could get decent 3-D performance without having to hog a valuable expansion slot(s) like what you had to do with the earlier Voodoo cards. This is something a lot of end users and OEM's really liked in terms of simpler installation.
Alas, by the time 3dfx released Voodoo5, both nVidia and ATI with their one-board solutions pretty much sewed up the market, and it was essentially all over for 3dfx.
By the way, the original ATI Rage Pro chipset wasn't really that great for 3-D; it wasn't until the Rage 128 that ATI started to really make strides for 3-D performance, and the Radeon R100/R200 chipsets finally had pretty good 3-D performance.
In short, 3dfx sat on its laurels too long and could not come back against nVidia and ATI.
3dfx started to fail for this reason
on
The Last Days at 3dfx
·
· Score: 4, Insightful
I think one thing that really started to kill 3dfx was the fact until Voodoo5, 3dfx acceleration required you buy a separate board in addition to the main graphics card, something many users and OEM's intensely dislike.
When both nVidia and ATI started offering better 3-D graphics cards that didn't need a second card for good 3-D performance, that seriously hurt 3dfx very quickly. It also didn't help that 3dfx's offerings when the Voodoo5 did finally get released didn't compare well with the nVidia and ATI competition, either.
What finally killed 3dfx was the release of nVidia's GeForce 256 chipset, which offered a quantum leap forward in 3-D acceleration. ATI's rapid development of the Radeon R100 and R200 chipsets didn't help things for 3dfx, either.
Oh, I remember back in 1991 when Intel said they would develop a 1 GHz clock cycle CPU by 2000. People scoffed at the idea but in reality Intel did reach 1 GHz by 2000 on the Pentium III CPU.
Anything that requires serious computing power: processing multimedia files (video and audio) and image processing. A large cluster of 4.7 GHz Pentium 4 CPU's could dramatically speed up computer animation creation, for starters.
I'm not surprised that Hewlett-Packard is working with Microsoft on the.NET initiative.
Don't forget that Microsoft is a company sitting on US$40 billion in liquid assets and HP knows Microsoft will be around for a long time, which gives MS time to develop and improve.NET software.
Besides, I'm sure HP is well-aware of Ximian's Mono project, which essentially is an Open Source version of.NET. That way, HP can sell servers running Linux with Mono which means HP will come out a winner in the end in the long run.
(Mind you, I think Microsoft has neferious reasons for assisting Ximian in developing Mono--it will essentially do an end-around attack on Sun's Liberty Alliance initiative. Sun might not realize what hit them when they find out why most of the world is supporting something akin to.NET for web services.)
I think electric/fossil-fuel hybrids are the way to go for now until the arrival of low-cost, practical fuel-cell engines.
The Toyota Prius, Honda Insight and Honda Civic Hybrid have demonstrated you can get a very practical car with good driveability, very low emissions, and most importantly long range. Why get a car with 70-100 mile range (at best) when you can get something that can be used as a daily driver and also take a small family on long trips?
It's small wonder why both Toyota and Honda have begun to expand hybrid technology to their other model lines. Don't be surprised within a few years we'll be seeing small vans and station wagons with hybrid drivetrains from both Toyota and Honda; Toyota has stated they may produce a hybrid version of the Corolla and Honda is looking at putting hybrids into the Latitude (neé Stream) minivan and Jazz supermini hatchback.
What's interesting is that Honda may have done a better job creating a real car using a hybrid drivetrain than what Toyota did with the Prius, according to Motor Trend magazine in a recent review.
The problem with the Prius is that while is quite roomy for its size and has good pep for the hybrid drivetrain, the road handling and the quirky ergonomics of the car may not be to everyone's tastes. Because the Honda Civic Hybrid is based upon the current model Civic four-door sedan, you get exactly the same interior arrangements as the current Civic (which is quite good), though the instrument panel is a little different (to reflect operation of the Integrated Motor Assist (IMA) hybrid drive system) and you do lose a little trunk space in order to accommodate the electric motor batteries.
I've driven the Civic Hybrid sedan with the CVT automatic and it's actually quite good, with decent acceleration thanks to the CVT transmission. If I had the money I'd trade in my Civic HX CVT coupe for the Civic Hybrid sedan with CVT automatic.:-)
Actually, Windows 2000 Professional works great even on a Celeron A 500 MHz machine provided you give it enough RAM (256 MB minimum, but you really want more like 384 MB) and install the OS on a modern ATA-33 or faster interface hard drive with 20 GB of disk space available.
I think people don't realize that there are ways to substantially increase the speed of your computer without having to buy a new very fast computer altogether.
The first upgrade everyone should consider is getting as much RAM as you can afford. People are surprised that many computers with DIMM slots built during the late 1990's only sported around a measly 64 MB of system RAM; with the price of RAM so cheap nowadays there's no excuse to run at minimum 256 MB of system RAM. With that much RAM available, the need from almost every operating system to use the hard drive for virtual memory is very low, which substantially speeds up overall system performance (not to mention being able to run more apps simultaneously safely).
The other major improvement that really helps is to get the fastest hard drive you can afford; today's ATA-100/133 hard drives run very fast even on ATA-33 connections because of the fact most of them run at 7200 RPM, which really speeds up general hard drive access; you can get low-cost controllers and at full ATA-100/133 access speeds also.
Finally, if the first two upgrades I suggested still doesn't improve things you can get CPU upgrades to substantially up your CPU speed. If your computer has software CPU configuration and uses Socket 370, you can get upgrades that could increase the CPU speed by a factor of 2 to 3 times what you have now; Powerleap makes the S370 adapter to install later-production Coppermine-core CPU's on older Socket 370 motherboards, and Powerleap also makes the PL-370/T CPU upgrade that allows you to run a Tualatin-core Celeron as fast as 1.2 GHz depending on your motherboard design.
I think one factor that not that many people talk about is the fact that the latest generation of graphics cards can do hardware acceleration at a pretty high level (at least both hardware motion compensation and Inverse Discrete Cosine Tranform) for decoding MPEG-2 video streams from DVD movies.
ATI has been doing this since their Rage 128 chipset arrived around 1999; nVidia's GeForce4 MX and GeForce4 Ti also do this now, and Matrox's Parhelia board do it now, too.
In short, the market for seperate DVD decoding boards has pretty much vanished on the desktop and laptop computer market.
Right now, the USA is probably the largest market for legal DVD players and discs.
There are two reasons for this: 1) the price of players have dropped dramatically--you can even now get 480p progressive scan DVD players for under US$200 nowadays; and 2) the price of DVD discs are still pretty reasonable (US$20 to US$30 for new releases, US$10 to US$15 for older releases).
Because the price of DVD discs are still reasonable, the incentive to pirate movies is still very low. DVD's very sharp picture quality is one reason why VCD's have never really taken off in the USA. Say what you want about the MPAA, but the fact that movie companies have pretty much followed Disney's sell-through policy for pre-recorded movies in the DVD market has actually discouraged movie piracy for the vast majority of computer users out there.
While AMD works out the bugs of their Hammer line of CPU's, don't forget that AMD still has a card to play in terms of CPU competition with Intel: the Barton-core Athlon CPU due later this fall.
Unlike the Athlon CPU core designs since the original Thunderbird-core Athlon's, the Barton-core Athlon sports a larger 512 KB L2 cache on the CPU die, which will offer dramatic performance increases, especially with memory-intensive programs. Remember, the current Thoroughbred-core Athlon CPU rated at 2600+ already has reached parity with the Intel Pentium 4 2.53 GHz part, and that's with only 256 KB of L2 cache on the CPU die and using DDR266 DDR-SDRAM! What will the Barton-core Athlon do?
2. The ballot is inserted into what looks like a larger version of the Votematic machine.
3. When you mark off the ballot, instead of punching out holes in the ballot you mark off your selections with a small permanent ink stamp.
4. The ballot is turned into the voting station worker at the voting site, where the ballot is read electronically (but without telling the worker what selections were made) to make sure all the ink marks are in the right locations; this will detect the possibility of overvotes, undervotes and improper marking of the ballot.
5. Once the voter verifies that the selections are what they want, the ballot is turned in and the voter gets a receipt of voting at the voting site.
The advantage of a ink-marked ballot is that not only are they machine-readable, but they can be easily read by hand counts as a backup. It's not completely perfect but it's way better than the punch card ballot and electronic balloting, both of which can be tampered with.
If you're referring to 1/3 height 3.5" form factor hard drives, I can understand your statements. In fact, right now the smallest new production hard drives for the IDE interface is about 40 GB or so.
However, by 2010 the 20-30 GB hard drive in my proposed TiVo unit with the 400-800 GB removeable optical drive storage with by a very tiny unit about the size of an IBM Microdrive, a unit that uses so little power that it would have insignificant draw on the power supply of the TiVo unit itself.
1) Increase the size of the ballot sheet to 8.5" x 11".
2) Insert that sheet into something that looks like a oversized Votematic machine.
3) It will work like a Votematic punch card machine, but instead of punching holes in the ballot sheet there is enough area exposed on the ballot where you mark off your choice with a small ink stamp.
4) Once the voting is finished, you give the completed ballot to the people at the voting station and they will do a preliminary optical read (without revealing what was voted on) to make sure the voter has marked off all the right spots; this will prevent double-voting, not marking in the right area, etc.
5) Once that is verified and the voter says they are satisfied with what they voted for, the voter gets a receipt proving they have voted.
The advantage of using a marker to make the selection is that the ballot can be both machine read AND hand-counted easily. That way, the accuracy will be very high indeed.
In regards to all of Europe having more Internet users than the USA and Canada combined, we have to consider the following factors:
1. Europe's total population is 350 million plus, and given that Europe is a highly developed continent technologically I'm not surprised that Internet usage has gone so high.
2. Does the Nua.com survey also include users of cellphones? If it does that skews the usage equation since cellphone Internet connections are usually done in short bursts, not the long, continuous connections you get with a desktop computer. Subtract the cellphone users and let's see how many Europeans are accessing the Internet using desktop/laptop computers via dial-up or broadband connections.
However, I am happy that the European telcos are finally getting the message that one fee per month unmetered Internet access is the way to go. The fact I've read on this message thread that DSL access is rapidly growing in Europe tells me we'll see some very rapid growth in Internet usage in Europe over the next decade.
Personally, I think TiVo may not go the direction of the huge hard drive for recorded program storage.
With the pace of rapid advancements in re-writeable optical storage in the last four years, it'll be far more likely that by 2010 TiVo units will sport a 20 to 30 GB hard drive for the Linux-based OS, TiVo program code itself, and recorded program index pointers, then you'll see a 400-800 GB removable optical drive for the actual recorded program storage connected using a faster version of the Serial ATA interface. Such a device will finally spell the end of VHS.
Slashdot Mirror
If the UCSB admins were smart they would have conveniently posted information about how to make Windows 2000 Profesional reasonably secure.
Things like installing Service Pack 3, setting accounts correctly, banning the use of personal web servers on a client machine, and mandatory installation of a good antivirus and/or firewall program would have saved the UCSB sysadmins a lot of headaches.
You're right about that!
Windows 2000 Professional with its Service Pack 3 and McAfee AntiVirus 7.0 (antivirus + firewall) is actually a reasonably secure OS setup.
Hold it right there.
Didn't Intel introduce the Northwood-core Pentium 4's starting with the 2.0 and 2.2 GHz versions first? I think increasing the L2 cache on the CPU die to a generous 512 KB did wonders for the CPU. It'll be very interesting to see what the Prescott-core Pentium 4's with its 1024 KB L2 cache does in terms of performance.
I can understand your issues with our government, but at least you have our courts to try to redress this.
This isn't China, where the government is NOT going to address such concerns, I'm sorry to say.
I don't think you realize what the Chinese are trying to do.
By developing their own CPU and operating system through official government sanction, it gives the government a way to effectively spy on Internet users because the government knows how everything works and will very likely use this knowledge to attempt such control. You are forgetting that mainland China is still in many ways an authoritarian state and the government is more than willing to spy on its own people to stamp out enemies of the state such as the Fulan Gong movement.
Does the book 1984 have any meaning to you? Mainland China is headed in that direction if government control of hardware and software technology has its way.
South Korea has a very high percentage of young people and thus more people in the work force.
And more importantly, a work force that is highly literate and have lots of technical skills, too. I saw a show on The Learning Channel about a month ago on Hyundai's amazing shipbuilding plant and it shows just how good their skills at shipbuilding have gotten.
I think Lester Thurow was wrong on Japan because their economic collapse showed that Japan's cultural norms could not accommodate the changes necessary to improve their economic systems.
Look at South Korea--after the horrid experience of the Asian financial crisis of 1997-1999 this country was willing to take drastic steps to improve its economic system; as a result the country is doing quite well indeed.
Here in the USA, the fact we're more than willing to make changes in our economic system to correct problems show why the USA will do well economically.
However, ATI was saved by the fact they had a good number of OEM contracts, which pretty much kept them going until the more modern Radeon models started to make serious inroads among upgrade customers. Yes, the Radeon 8500 was a bit late and the initial drivers for the board was a bit disappointing, but in the end ATI did finally release decent drivers and the Radeon 8500 was much-liked.
:-(
But frankly, I do think that 3dfx's inability to take on nVidia and ATI in the OEM market by 1999 and their ill-advised purchase of STB did them in.
I still contend that the one-board solution was what really did in 3dfx.
Once nVidia developed the RIVA 128 and TNT/TNT2 chipsets, you could get decent 3-D performance without having to hog a valuable expansion slot(s) like what you had to do with the earlier Voodoo cards. This is something a lot of end users and OEM's really liked in terms of simpler installation.
Alas, by the time 3dfx released Voodoo5, both nVidia and ATI with their one-board solutions pretty much sewed up the market, and it was essentially all over for 3dfx.
By the way, the original ATI Rage Pro chipset wasn't really that great for 3-D; it wasn't until the Rage 128 that ATI started to really make strides for 3-D performance, and the Radeon R100/R200 chipsets finally had pretty good 3-D performance.
In short, 3dfx sat on its laurels too long and could not come back against nVidia and ATI.
I think one thing that really started to kill 3dfx was the fact until Voodoo5, 3dfx acceleration required you buy a separate board in addition to the main graphics card, something many users and OEM's intensely dislike.
When both nVidia and ATI started offering better 3-D graphics cards that didn't need a second card for good 3-D performance, that seriously hurt 3dfx very quickly. It also didn't help that 3dfx's offerings when the Voodoo5 did finally get released didn't compare well with the nVidia and ATI competition, either.
What finally killed 3dfx was the release of nVidia's GeForce 256 chipset, which offered a quantum leap forward in 3-D acceleration. ATI's rapid development of the Radeon R100 and R200 chipsets didn't help things for 3dfx, either.
Oh, I remember back in 1991 when Intel said they would develop a 1 GHz clock cycle CPU by 2000. People scoffed at the idea but in reality Intel did reach 1 GHz by 2000 on the Pentium III CPU.
Anything that requires serious computing power: processing multimedia files (video and audio) and image processing. A large cluster of 4.7 GHz Pentium 4 CPU's could dramatically speed up computer animation creation, for starters.
I'm not surprised that Hewlett-Packard is working with Microsoft on the .NET initiative.
.NET software.
.NET. That way, HP can sell servers running Linux with Mono which means HP will come out a winner in the end in the long run.
.NET for web services.)
Don't forget that Microsoft is a company sitting on US$40 billion in liquid assets and HP knows Microsoft will be around for a long time, which gives MS time to develop and improve
Besides, I'm sure HP is well-aware of Ximian's Mono project, which essentially is an Open Source version of
(Mind you, I think Microsoft has neferious reasons for assisting Ximian in developing Mono--it will essentially do an end-around attack on Sun's Liberty Alliance initiative. Sun might not realize what hit them when they find out why most of the world is supporting something akin to
I think electric/fossil-fuel hybrids are the way to go for now until the arrival of low-cost, practical fuel-cell engines.
The Toyota Prius, Honda Insight and Honda Civic Hybrid have demonstrated you can get a very practical car with good driveability, very low emissions, and most importantly long range. Why get a car with 70-100 mile range (at best) when you can get something that can be used as a daily driver and also take a small family on long trips?
It's small wonder why both Toyota and Honda have begun to expand hybrid technology to their other model lines. Don't be surprised within a few years we'll be seeing small vans and station wagons with hybrid drivetrains from both Toyota and Honda; Toyota has stated they may produce a hybrid version of the Corolla and Honda is looking at putting hybrids into the Latitude (neé Stream) minivan and Jazz supermini hatchback.
What's interesting is that Honda may have done a better job creating a real car using a hybrid drivetrain than what Toyota did with the Prius, according to Motor Trend magazine in a recent review.
:-)
The problem with the Prius is that while is quite roomy for its size and has good pep for the hybrid drivetrain, the road handling and the quirky ergonomics of the car may not be to everyone's tastes. Because the Honda Civic Hybrid is based upon the current model Civic four-door sedan, you get exactly the same interior arrangements as the current Civic (which is quite good), though the instrument panel is a little different (to reflect operation of the Integrated Motor Assist (IMA) hybrid drive system) and you do lose a little trunk space in order to accommodate the electric motor batteries.
I've driven the Civic Hybrid sedan with the CVT automatic and it's actually quite good, with decent acceleration thanks to the CVT transmission. If I had the money I'd trade in my Civic HX CVT coupe for the Civic Hybrid sedan with CVT automatic.
Actually, Windows 2000 Professional works great even on a Celeron A 500 MHz machine provided you give it enough RAM (256 MB minimum, but you really want more like 384 MB) and install the OS on a modern ATA-33 or faster interface hard drive with 20 GB of disk space available.
I think people don't realize that there are ways to substantially increase the speed of your computer without having to buy a new very fast computer altogether.
The first upgrade everyone should consider is getting as much RAM as you can afford. People are surprised that many computers with DIMM slots built during the late 1990's only sported around a measly 64 MB of system RAM; with the price of RAM so cheap nowadays there's no excuse to run at minimum 256 MB of system RAM. With that much RAM available, the need from almost every operating system to use the hard drive for virtual memory is very low, which substantially speeds up overall system performance (not to mention being able to run more apps simultaneously safely).
The other major improvement that really helps is to get the fastest hard drive you can afford; today's ATA-100/133 hard drives run very fast even on ATA-33 connections because of the fact most of them run at 7200 RPM, which really speeds up general hard drive access; you can get low-cost controllers and at full ATA-100/133 access speeds also.
Finally, if the first two upgrades I suggested still doesn't improve things you can get CPU upgrades to substantially up your CPU speed. If your computer has software CPU configuration and uses Socket 370, you can get upgrades that could increase the CPU speed by a factor of 2 to 3 times what you have now; Powerleap makes the S370 adapter to install later-production Coppermine-core CPU's on older Socket 370 motherboards, and Powerleap also makes the PL-370/T CPU upgrade that allows you to run a Tualatin-core Celeron as fast as 1.2 GHz depending on your motherboard design.
I think one factor that not that many people talk about is the fact that the latest generation of graphics cards can do hardware acceleration at a pretty high level (at least both hardware motion compensation and Inverse Discrete Cosine Tranform) for decoding MPEG-2 video streams from DVD movies.
ATI has been doing this since their Rage 128 chipset arrived around 1999; nVidia's GeForce4 MX and GeForce4 Ti also do this now, and Matrox's Parhelia board do it now, too.
In short, the market for seperate DVD decoding boards has pretty much vanished on the desktop and laptop computer market.
Right now, the USA is probably the largest market for legal DVD players and discs.
There are two reasons for this: 1) the price of players have dropped dramatically--you can even now get 480p progressive scan DVD players for under US$200 nowadays; and 2) the price of DVD discs are still pretty reasonable (US$20 to US$30 for new releases, US$10 to US$15 for older releases).
Because the price of DVD discs are still reasonable, the incentive to pirate movies is still very low. DVD's very sharp picture quality is one reason why VCD's have never really taken off in the USA. Say what you want about the MPAA, but the fact that movie companies have pretty much followed Disney's sell-through policy for pre-recorded movies in the DVD market has actually discouraged movie piracy for the vast majority of computer users out there.
Folks,
While AMD works out the bugs of their Hammer line of CPU's, don't forget that AMD still has a card to play in terms of CPU competition with Intel: the Barton-core Athlon CPU due later this fall.
Unlike the Athlon CPU core designs since the original Thunderbird-core Athlon's, the Barton-core Athlon sports a larger 512 KB L2 cache on the CPU die, which will offer dramatic performance increases, especially with memory-intensive programs. Remember, the current Thoroughbred-core Athlon CPU rated at 2600+ already has reached parity with the Intel Pentium 4 2.53 GHz part, and that's with only 256 KB of L2 cache on the CPU die and using DDR266 DDR-SDRAM! What will the Barton-core Athlon do?
Like I said yesterday, here's my solution:
1. Increase the size of the ballot to 8.5" x 11".
2. The ballot is inserted into what looks like a larger version of the Votematic machine.
3. When you mark off the ballot, instead of punching out holes in the ballot you mark off your selections with a small permanent ink stamp.
4. The ballot is turned into the voting station worker at the voting site, where the ballot is read electronically (but without telling the worker what selections were made) to make sure all the ink marks are in the right locations; this will detect the possibility of overvotes, undervotes and improper marking of the ballot.
5. Once the voter verifies that the selections are what they want, the ballot is turned in and the voter gets a receipt of voting at the voting site.
The advantage of a ink-marked ballot is that not only are they machine-readable, but they can be easily read by hand counts as a backup. It's not completely perfect but it's way better than the punch card ballot and electronic balloting, both of which can be tampered with.
If you're referring to 1/3 height 3.5" form factor hard drives, I can understand your statements. In fact, right now the smallest new production hard drives for the IDE interface is about 40 GB or so.
However, by 2010 the 20-30 GB hard drive in my proposed TiVo unit with the 400-800 GB removeable optical drive storage with by a very tiny unit about the size of an IBM Microdrive, a unit that uses so little power that it would have insignificant draw on the power supply of the TiVo unit itself.
I think the ultimate solution is this:
1) Increase the size of the ballot sheet to 8.5" x 11".
2) Insert that sheet into something that looks like a oversized Votematic machine.
3) It will work like a Votematic punch card machine, but instead of punching holes in the ballot sheet there is enough area exposed on the ballot where you mark off your choice with a small ink stamp.
4) Once the voting is finished, you give the completed ballot to the people at the voting station and they will do a preliminary optical read (without revealing what was voted on) to make sure the voter has marked off all the right spots; this will prevent double-voting, not marking in the right area, etc.
5) Once that is verified and the voter says they are satisfied with what they voted for, the voter gets a receipt proving they have voted.
The advantage of using a marker to make the selection is that the ballot can be both machine read AND hand-counted easily. That way, the accuracy will be very high indeed.
In regards to all of Europe having more Internet users than the USA and Canada combined, we have to consider the following factors:
1. Europe's total population is 350 million plus, and given that Europe is a highly developed continent technologically I'm not surprised that Internet usage has gone so high.
2. Does the Nua.com survey also include users of cellphones? If it does that skews the usage equation since cellphone Internet connections are usually done in short bursts, not the long, continuous connections you get with a desktop computer. Subtract the cellphone users and let's see how many Europeans are accessing the Internet using desktop/laptop computers via dial-up or broadband connections.
However, I am happy that the European telcos are finally getting the message that one fee per month unmetered Internet access is the way to go. The fact I've read on this message thread that DSL access is rapidly growing in Europe tells me we'll see some very rapid growth in Internet usage in Europe over the next decade.
Personally, I think TiVo may not go the direction of the huge hard drive for recorded program storage.
With the pace of rapid advancements in re-writeable optical storage in the last four years, it'll be far more likely that by 2010 TiVo units will sport a 20 to 30 GB hard drive for the Linux-based OS, TiVo program code itself, and recorded program index pointers, then you'll see a 400-800 GB removable optical drive for the actual recorded program storage connected using a faster version of the Serial ATA interface. Such a device will finally spell the end of VHS.