>Regardless of how much you disagree with a license, >doesn't make it any more right to turn around and >do the same thing that you hate so much when you're >on the other side.
Why not? A licence is NOT a licence is NOT a licence. There are differences between licences; many of the ones widely opposed in the/. community are quite draconian, when they're even enforcable. OTOH, the GNU licences typically provide benefits for all parties involved.
Why does it have to be a tradeoff?
on
Copyright Rumblings
·
· Score: 4, Insightful
Call me inflamatory, but we screwed up copyright the first time (Gosh... if forever was good, forever and a day is better!). Why do we have to give the content industry a lollipop (copy-protection enforcement) to get them to accept something that's sensible and in the public good?
The only copy-protection I'll settle for will consist of a Supreme Court justice in every box that tells me that what I'm doing is or isn't fair use. Unfortunately, unless the content industry and the Raelians team up, that tends to limit sales.
If the goal is protecting the integrity of processes, wasn't this what we were supposed to expect from any halfway-decent hardware and OS combination? It sure didn't seem too hard when the operating-systems books describe the principle.
Or is this more of a thing where "protecting" means "attempting to cram protection onto a system that's fundamentally broken so some dork can still use his copy of Excel for Windows/286?"
But the CD burner you're buying is really, really fast, so doesn't it cost the equal of 3 or 4 times 57.99? Remember, the strength of math (even the RIAA flavour) is its consistency!
I've used LCDs (in the context of laptops) and CRTs. I'd rather have a CRT where the option exists.
-There's no chance of a dead pixel wiggling across the virtual screen when I scroll my 1600x1200 virtual desktop on a 1152x864 actual screen.
-No viewing angle problems. Period.
-All the colours are attractive. Have they finally made LCDs that do adequate red and brown?
-No scaling problems. I need 640x480. I want 1152x864. If my screen is tied to 1280x1024, I'd have either a viewing window the size of a postage stamp or terrible scale-up.
Fortunately, I already have a 19" tube. Only way I'm getting a bigger one is if someone chains one of those dirt-cheap 10-year-old 20" Sun or HP fixed-frequency monitors to the back of a Voodoo III.
> For example, McDonalds is bigger than Burger King. >If they wanted to (and were allowed to), McD's >could put BK out of business in 6 months by selling >everything on their menu for $0.05. That is what >the government considers "unfair" and will step in >to stop, which is why BK and McDs coexist, as they >do.
Actually, not necessarily true. If you're losing 50 cents on every burger, you'd better have enough cash or credit to draw on to absorb the monstrous losses. Most companies can't. Generally, the 'loss leader' still results in a profit per transaction (get a cheap burger and overpriced soda)
To avoid fixing the problem once and for all (new, longer numbers and a more rational distribution of them), which would inconvinence everyone ONCE, it seems that they'd rather constantly divide and redivide area codes and tack on bizarre variations (dial 1, don't dial 1, smash the phone against your head, do the hokey pokey, and shake it all about), inconvinencing many, many people several times over.
For example, probably 15 years ago, all of Arizona was 602. Then everywhere outside of the Phoenix area became 520. Then they changed everyone outside the Tucson area (928?). That means at least two sets of business cards rendered obsolete and two reprogramming of the switches.
How about a "routing-based" numbering system. Just keep enterring numbers until you've given the complete routing to a phone, rather than another switchbox... I could see it working like follows:
"If I'm calling out of town, I dial 562 to get to the main trunk, then the number for the city, the number for the street, the number for the block, the number for the building, then finally the number for the person I'm calling."
Dialing local calls could probably be shorter (one figure to get to the local box, then route it to your neighbour with three or four more), long-distance wouldn't be much more difficult than the current system (particularly if you're already dialing 10-10-foo), and it's infinitely extensible.
>The Humvee may be bad compared to a Metro, but I >bet it still gets better gas mileage and lower >emissions out of the Chevy 350 it has under the hood >than any of the millions of 60's,70's and early 80's >vehicles on the
Not necessarily. Used to get ~18 MPG city out of a '76 Toyota Corona. Fairly big car, seated six. It leaked oil, but had the car been worth more than $38, it would have justified replacing the leaking oil pan. Your Hummer example gets between 8 and 13. (reference http://www.madison.com/captimes/news/stories/37942.php)
The environmental problems with older cars are not necessarily inherent. Rather, much of their leaks, half-combustion, and smoke is likely due to old, broken, or worn-out parts that can't be replaced with anything even close to factory spec affordably (it's real fun to hunt through the junkyard for spares!).
If they built the same Corona today, it would probably get 20-25MPG, but it wouldn't drip oil.
>Today, however, nearly everyone with a broadband >internet connection is in a location to receive the >major networks.
It depends, and it probably will depend more in the future.
Situation 1. I have a TV tuner card. It's great, but channels below 8 or 10 simply don't come in in a legible form for me. In other cases, other channels don't come in. This is probably due to interference with the computer components.
Situation 2: Low-power broadcasters. I can't get most of the low-power stations even above 10. They're visible across town, but not here. Even on a "real" TV, they come in poorly. I'd rather stream a quality view.
Situation 3: Digital TV. Isn't digital blessed with much worse degradation qualities? If this is true, the acceptable-viewing radius shrinks up.
>Satellite radio would be something I'd be >interested in if it was more feature-rich and >Internet enabled. Imagine a service with a >Tivo-like reciever that is capable of storing >songs, seeking new songs you might be interested in >based on your past preferences and allowing you to >build your own playlist.
I would like such a system too. I wonder if a system like the following could work without a pay service:
1. Research phase. Encode the first few seconds of many recorded songs. 2. Server design. Checksum the recordings in a manner where similar sounds deliver similar checksums. Store the checksums on a server. 3. Client design. After each pause of n milliseconds, assume a new song begins. Record the first five seconds to a buffer and checksum the song. Keep recording while you ask the server "What song is checksum xxxxxxx?". If the response is on the playlist, begin transferring a recording to permanent storage. The server can supply data about the song length so it cuts off before the "You're listening to 171.5 KXALQF!" line.
The big issues would be when different recordings sound similar in the first 5 seconds (false positives or negatives, wrong cutoff time values), live shows (you'd have to start and stop recording manually), and attempts, perhaps by the radio station, to affect the sound of the recordings so they aren't accurately guessed and mangle the system.
Why don't states gain revenue through the ownership and profitable operation of businesses; perhaps purchasing successful ones or starting competitors to them? This would supply revenue without ANY percieved cost to consumers. I would think the ideal situations for this would be businesses that already enjoy limited need to add new services, like utilities and banks. A slight variation on this has been seen in some "Idiot Pays" laws... if you drive yourself into a flooded river, you pay the costs of the rescue, so why not tack on a 20% markup and call it profit?
In fact, the primary actions I've seen are completely to the contrary: state constitutions forbidding the state from competing with private businesses.
I'm not sure if I should feel comfortable with the idea of code "called before main" combined with OOP.
What stops someone from writing a malicious routine called before main is run, and slipping it into a tainted version of a popular (how do you say library in OOPese?) A truly paranoid programme could run MD5SUMs on each library it requested functions from and say "okay, this is the libc.so.6 that came with Slackware 8.1, so it's good"... but it couldn't do that until it got to main, and the damage was done.
I can also see a problem with race conditions. Library A sets things up how it needs them before execution, and so does Library B. Is the order in which the pre-main code is run standardised? Even if it is, it seems to open the door for a lot of confusion, especially if you add a library in and suddenly things break because of non-immediately-obvious changes to the pre-main sequence.
OTOH, I appreciate that you aren't forced to do OO. I'll believe OO is a panacea when it buries me knee-deep in gold sovereigns, and smites my enemies with heavy objects. I don't think those functions are in the Java spec yet.
There are reasons to want to buy individual shows at 50 cents instead of a disc at an average of $3/show.
Issue 1: Flexibility. I don't necessarily want every episode of a series. Personally, I'd buy most of the 'Gargoyles' series, but I don't want the last 10 or so episodes. If you throw it all on one disc, I'm paying for stuff I don't want. Quite reasonable not to want to do that.
Issue 2: Cost. $3 is still 6x as much as 50 cents. The $500 difference is, despite your opinion, a lot of money for millions of Americans.
Issue 3: Convinence. People are willing to pay a little for a convinent service. The trick is to figure out what people are willing to pay not to fight with getting the files down a 56k connection to Gnutella and ensuring they're correct, and price the product at that level.
>Think about it: instead of having to put together a >complete breakfast for your troops, you can just >feed them a generic mass of proteins and sugars, >and slap a patch on them to take care of the >vitamin and mineral requirements. Much faster, >cheaper, and field-portable.
It also produces an "all your eggs in one basket" problem. If an enemy wanted to mess with a food supply, all they have to do is destroy or steal a few kilogrammes of patches, and you no longer have the supplies to guarantee sufficient nutrients for the troops.
With conventional food, to take down the micronutrients, you also have to take down the macronutrients, which are too bulky to easily steal and too big to burn without someone noticing the fire before it's too late.
This would seem to be a legitimate strategic concern. I suppose they have to guard food shipments now, but if you break food down into "bulk" and "essential vitamins", it creates a more interesting target.
Isn't a large part of the problem still the basic trojan horse? Signing systems can't PRACTICALLY (operative word) defend against those. They can:
1. Offer the user a chance to approve unsigned code, which he will likely get in the habit of doing for the legitimate but unsigned code out there (will there be a signature for my copy of Railroad Tycoon on 360k floppies?), in which case a user may be easily tricked into running the trojan, and we're no better off.
or
2. Refuse all unsigned code. Application base evaporates. Many, many users refuse to upgrade because custom apps or even older apps (or competing apps that mysteriously aren't certified) break.
The real problem with progressive taxes, as done now, is that the function used to calculate the rate isn't smooth-- you have the distinct rate brackets.
This results in regions where you're actually taking home less, despite earning more.
Example: Consider two brackets, 10% below $20,000 and 12% above. (these numbers are hypothetical, intended to show the point).
The person earning $19,800 takes home $17,820, but someone earning $20,100 actually only takes home $17,688!
Why not some sort of clever exponential function instead?
I too have the 133MHz version, in the form of one of those aftermarket upgrades which sits on a small voltage-converter card. I bought mine recently, because I always wanted one but could never justify them ($60-100) when I had a 486 desktop.
Mine sits in an old laptop (Dual DC-4000) which I gave up on because the suspend feature seemed non-functional and the screen hinges were wrecked. Ironically, mine ran OS/2 as well... performance wasn't as good as I hoped because there was no L2 cache on the board:( Still, it's a fun little part, and impressively takes a much smaller heatsink than my old DX2/66 took!
I seem to recall only hearing about the 160MHz part-- possibly "P90" rated-- in terms of "They were going to release these as 160s, but instead just did them as 133s, so they're the best overclockers"
Novelty is worth big points, so I'd expect interest in:
1. Prototypes (reference Commodore 65) 2. Unusual design or implementation decisions (Pen Computing devices) 3. Firsts (Why not drop $10 on an original Palm Pilot in case it's worth something someday) 4. Lasts, conversely (the last entries in the Amiga and ST lines, for example) 5. Things with an undesirable reputation-- bet you wish you still had that P60 with the bad FPU! 6. Items that were rarities due to supply or marketing decisions (I bet that 1.2GHz Hammers they sent around for demonstration will be worth something, and did anyone ever get a 160MHz Am5x86?)
You can easily do better than $5. Check if your local university has a surplus equipment sales yard. Mine does. Huge box of keyboards... everything from real Model Ms to Sun something-or-others to Apple Extended IIs to modern rubbish... as-is, 50 cents a throw. Take home a bucketload of Ms, and swap parts (I took one with sound switches, but several keys missing, and another with all keys there but a sticky tab) until you have a good one.
It was several years ago, and it was a choice between using a Packard Bell P100 (no prize pig itself, often had the "Sub 10fps tortoise" icon showing), and the 486, built with love.
Also, a 40MHz 486 (apparently usually a DX2, not a DX) is a common speed for laptops in the price category that counts as "disposable" (sub-USD 50). Such machines, even today, are great for low-horsepower apps where damage to the machine is considered too likely to risk more money on a better machine. You can have my Toshiba T1950CT (DX2/40, very nice 8.5" TFT screen) when you even-swap me a nice P200MMX for it.
Slashdot Mirror
They built an anti-radiation shield out of all those unspent Susan B. Anthony dollars.
>Regardless of how much you disagree with a license, >doesn't make it any more right to turn around and >do the same thing that you hate so much when you're >on the other side.
/. community are quite draconian, when they're even enforcable. OTOH, the GNU licences typically provide benefits for all parties involved.
Why not? A licence is NOT a licence is NOT a licence. There are differences between licences; many of the ones widely opposed in the
Call me inflamatory, but we screwed up copyright the first time (Gosh... if forever was good, forever and a day is better!). Why do we have to give the content industry a lollipop (copy-protection enforcement) to get them to accept something that's sensible and in the public good?
The only copy-protection I'll settle for will consist of a Supreme Court justice in every box that tells me that what I'm doing is or isn't fair use. Unfortunately, unless the content industry and the Raelians team up, that tends to limit sales.
If the goal is protecting the integrity of processes, wasn't this what we were supposed to expect from any halfway-decent hardware and OS combination? It sure didn't seem too hard when the operating-systems books describe the principle.
Or is this more of a thing where "protecting" means "attempting to cram protection onto a system that's fundamentally broken so some dork can still use his copy of Excel for Windows/286?"
But the CD burner you're buying is really, really fast, so doesn't it cost the equal of 3 or 4 times 57.99? Remember, the strength of math (even the RIAA flavour) is its consistency!
I've used LCDs (in the context of laptops) and CRTs. I'd rather have a CRT where the option exists.
-There's no chance of a dead pixel wiggling across the virtual screen when I scroll my 1600x1200 virtual desktop on a 1152x864 actual screen.
-No viewing angle problems. Period.
-All the colours are attractive. Have they finally made LCDs that do adequate red and brown?
-No scaling problems. I need 640x480. I want 1152x864. If my screen is tied to 1280x1024, I'd have either a viewing window the size of a postage stamp or terrible scale-up.
Fortunately, I already have a 19" tube. Only way I'm getting a bigger one is if someone chains one of those dirt-cheap 10-year-old 20" Sun or HP fixed-frequency monitors to the back of a Voodoo III.
I believe you could get just as terrifying effects much more easily.
Four large stripes, one in each colour, and at the bottom, a simple line of black text. "This page consumed $3.62 worth of ink!"
> For example, McDonalds is bigger than Burger King. >If they wanted to (and were allowed to), McD's >could put BK out of business in 6 months by selling >everything on their menu for $0.05. That is what >the government considers "unfair" and will step in >to stop, which is why BK and McDs coexist, as they >do.
Actually, not necessarily true. If you're losing 50 cents on every burger, you'd better have enough cash or credit to draw on to absorb the monstrous losses. Most companies can't. Generally, the 'loss leader' still results in a profit per transaction (get a cheap burger and overpriced soda)
d) It's the graphs. They were too mathy for the average American.
I would at least assume that someone like Jobs would be fairly pro-technology. That matters to me.
To avoid fixing the problem once and for all (new, longer numbers and a more rational distribution of them), which would inconvinence everyone ONCE, it seems that they'd rather constantly divide and redivide area codes and tack on bizarre variations (dial 1, don't dial 1, smash the phone against your head, do the hokey pokey, and shake it all about), inconvinencing many, many people several times over.
For example, probably 15 years ago, all of Arizona was 602. Then everywhere outside of the Phoenix area became 520. Then they changed everyone outside the Tucson area (928?). That means at least two sets of business cards rendered obsolete and two reprogramming of the switches.
How about a "routing-based" numbering system. Just keep enterring numbers until you've given the complete routing to a phone, rather than another switchbox... I could see it working like follows:
"If I'm calling out of town, I dial 562 to get to the main trunk, then the number for the city, the number for the street, the number for the block, the number for the building, then finally the number for the person I'm calling."
Dialing local calls could probably be shorter (one figure to get to the local box, then route it to your neighbour with three or four more), long-distance wouldn't be much more difficult than the current system (particularly if you're already dialing 10-10-foo), and it's infinitely extensible.
>The Humvee may be bad compared to a Metro, but I >bet it still gets better gas mileage and lower >emissions out of the Chevy 350 it has under the hood >than any of the millions of 60's,70's and early 80's >vehicles on the
2 .php)
Not necessarily. Used to get ~18 MPG city out of a '76 Toyota Corona. Fairly big car, seated six. It leaked oil, but had the car been worth more than $38, it would have justified replacing the leaking oil pan. Your Hummer example gets between 8 and 13. (reference http://www.madison.com/captimes/news/stories/3794
The environmental problems with older cars are not necessarily inherent. Rather, much of their leaks, half-combustion, and smoke is likely due to old, broken, or worn-out parts that can't be replaced with anything even close to factory spec affordably (it's real fun to hunt through the junkyard for spares!).
If they built the same Corona today, it would probably get 20-25MPG, but it wouldn't drip oil.
>Today, however, nearly everyone with a broadband >internet connection is in a location to receive the >major networks.
It depends, and it probably will depend more in the future.
Situation 1. I have a TV tuner card. It's great, but channels below 8 or 10 simply don't come in in a legible form for me. In other cases, other channels don't come in. This is probably due to interference with the computer components.
Situation 2: Low-power broadcasters. I can't get most of the low-power stations even above 10. They're visible across town, but not here. Even on a "real" TV, they come in poorly. I'd rather stream a quality view.
Situation 3: Digital TV. Isn't digital blessed with much worse degradation qualities? If this is true, the acceptable-viewing radius shrinks up.
I'm sorry, but I don't think this was the first Internet-based vote. I thought that's what the 2000 Arizona Democratic Primary was.
>Satellite radio would be something I'd be >interested in if it was more feature-rich and >Internet enabled. Imagine a service with a >Tivo-like reciever that is capable of storing >songs, seeking new songs you might be interested in >based on your past preferences and allowing you to >build your own playlist.
I would like such a system too. I wonder if a system like the following could work without a pay service:
1. Research phase. Encode the first few seconds of many recorded songs.
2. Server design. Checksum the recordings in a manner where similar sounds deliver similar checksums. Store the checksums on a server.
3. Client design. After each pause of n milliseconds, assume a new song begins. Record the first five seconds to a buffer and checksum the song. Keep recording while you ask the server "What song is checksum xxxxxxx?". If the response is on the playlist, begin transferring a recording to permanent storage. The server can supply data about the song length so it cuts off before the "You're listening to 171.5 KXALQF!" line.
The big issues would be when different recordings sound similar in the first 5 seconds (false positives or negatives, wrong cutoff time values), live shows (you'd have to start and stop recording manually), and attempts, perhaps by the radio station, to affect the sound of the recordings so they aren't accurately guessed and mangle the system.
Real bicycle: 40-50C in the summer locally
Virtual bike: 25C air-conditioned room
Real bicycle: To the mailbox and past the annoying dog
Virtual bike: Across the scenic bridges of the Draconic Homeworld
Real bicycle: Flat tyre every few weeks
Virtual bike: No tyre to flatten
Why don't states gain revenue through the ownership and profitable operation of businesses; perhaps purchasing successful ones or starting competitors to them? This would supply revenue without ANY percieved cost to consumers. I would think the ideal situations for this would be businesses that already enjoy limited need to add new services, like utilities and banks. A slight variation on this has been seen in some "Idiot Pays" laws... if you drive yourself into a flooded river, you pay the costs of the rescue, so why not tack on a 20% markup and call it profit?
In fact, the primary actions I've seen are completely to the contrary: state constitutions forbidding the state from competing with private businesses.
And BTW, since nobody said it...
1. Tax Internet Sales
2. ???
3. Profit!
I'm not sure if I should feel comfortable with the idea of code "called before main" combined with OOP.
What stops someone from writing a malicious routine called before main is run, and slipping it into a tainted version of a popular (how do you say library in OOPese?) A truly paranoid programme could run MD5SUMs on each library it requested functions from and say "okay, this is the libc.so.6 that came with Slackware 8.1, so it's good"... but it couldn't do that until it got to main, and the damage was done.
I can also see a problem with race conditions. Library A sets things up how it needs them before execution, and so does Library B. Is the order in which the pre-main code is run standardised? Even if it is, it seems to open the door for a lot of confusion, especially if you add a library in and suddenly things break because of non-immediately-obvious changes to the pre-main sequence.
OTOH, I appreciate that you aren't forced to do OO. I'll believe OO is a panacea when it buries me knee-deep in gold sovereigns, and smites my enemies with heavy objects. I don't think those functions are in the Java spec yet.
There are reasons to want to buy individual shows at 50 cents instead of a disc at an average of $3/show.
Issue 1: Flexibility. I don't necessarily want every episode of a series. Personally, I'd buy most of the 'Gargoyles' series, but I don't want the last 10 or so episodes. If you throw it all on one disc, I'm paying for stuff I don't want. Quite reasonable not to want to do that.
Issue 2: Cost. $3 is still 6x as much as 50 cents. The $500 difference is, despite your opinion, a lot of money for millions of Americans.
Issue 3: Convinence. People are willing to pay a little for a convinent service. The trick is to figure out what people are willing to pay not to fight with getting the files down a 56k connection to Gnutella and ensuring they're correct, and price the product at that level.
>Think about it: instead of having to put together a >complete breakfast for your troops, you can just >feed them a generic mass of proteins and sugars, >and slap a patch on them to take care of the >vitamin and mineral requirements. Much faster, >cheaper, and field-portable.
It also produces an "all your eggs in one basket" problem. If an enemy wanted to mess with a food supply, all they have to do is destroy or steal a few kilogrammes of patches, and you no longer have the supplies to guarantee sufficient nutrients for the troops.
With conventional food, to take down the micronutrients, you also have to take down the macronutrients, which are too bulky to easily steal and too big to burn without someone noticing the fire before it's too late.
This would seem to be a legitimate strategic concern. I suppose they have to guard food shipments now, but if you break food down into "bulk" and "essential vitamins", it creates a more interesting target.
Isn't a large part of the problem still the basic trojan horse? Signing systems can't PRACTICALLY (operative word) defend against those. They can:
1. Offer the user a chance to approve unsigned code, which he will likely get in the habit of doing for the legitimate but unsigned code out there (will there be a signature for my copy of Railroad Tycoon on 360k floppies?), in which case a user may be easily tricked into running the trojan, and we're no better off.
or
2. Refuse all unsigned code. Application base evaporates. Many, many users refuse to upgrade because custom apps or even older apps (or competing apps that mysteriously aren't certified) break.
The real problem with progressive taxes, as done now, is that the function used to calculate the rate isn't smooth-- you have the distinct rate brackets.
This results in regions where you're actually taking home less, despite earning more.
Example: Consider two brackets, 10% below $20,000 and 12% above. (these numbers are hypothetical, intended to show the point).
The person earning $19,800 takes home $17,820, but someone earning $20,100 actually only takes home $17,688!
Why not some sort of clever exponential function instead?
I too have the 133MHz version, in the form of one of those aftermarket upgrades which sits on a small voltage-converter card. I bought mine recently, because I always wanted one but could never justify them ($60-100) when I had a 486 desktop.
:( Still, it's a fun little part, and impressively takes a much smaller heatsink than my old DX2/66 took!
Mine sits in an old laptop (Dual DC-4000) which I gave up on because the suspend feature seemed non-functional and the screen hinges were wrecked. Ironically, mine ran OS/2 as well... performance wasn't as good as I hoped because there was no L2 cache on the board
I seem to recall only hearing about the 160MHz part-- possibly "P90" rated-- in terms of "They were going to release these as 160s, but instead just did them as 133s, so they're the best overclockers"
Novelty is worth big points, so I'd expect interest in:
1. Prototypes (reference Commodore 65)
2. Unusual design or implementation decisions (Pen Computing devices)
3. Firsts (Why not drop $10 on an original Palm Pilot in case it's worth something someday)
4. Lasts, conversely (the last entries in the Amiga and ST lines, for example)
5. Things with an undesirable reputation-- bet you wish you still had that P60 with the bad FPU!
6. Items that were rarities due to supply or marketing decisions (I bet that 1.2GHz Hammers they sent around for demonstration will be worth something, and did anyone ever get a 160MHz Am5x86?)
You can easily do better than $5. Check if your local university has a surplus equipment sales yard. Mine does. Huge box of keyboards... everything from real Model Ms to Sun something-or-others to Apple Extended IIs to modern rubbish... as-is, 50 cents a throw. Take home a bucketload of Ms, and swap parts (I took one with sound switches, but several keys missing, and another with all keys there but a sticky tab) until you have a good one.
It was several years ago, and it was a choice between using a Packard Bell P100 (no prize pig itself, often had the "Sub 10fps tortoise" icon showing), and the 486, built with love.
Also, a 40MHz 486 (apparently usually a DX2, not a DX) is a common speed for laptops in the price category that counts as "disposable" (sub-USD 50). Such machines, even today, are great for low-horsepower apps where damage to the machine is considered too likely to risk more money on a better machine. You can have my Toshiba T1950CT (DX2/40, very nice 8.5" TFT screen) when you even-swap me a nice P200MMX for it.