I think the bigger problem is, what happens when we reach the long-tail of process development, and demand tapers off to the point they can't fund further R&D?
IE: Systems are "good enough" and people go from buying one every 3 years to "only when they break". That could be 10+ years.
I suppose Intel would just follow the carrot to the next profitable market like they are pushing Atom CPUs lately?
Design them to be replaced every 3 years.
Servers are designed for a 5 year replacement cycle. Desktops are designed for a 3-5 year replacement cycle. Laptops are designed for a 3 year replacement cycle. Tablets are designed for a 2 year replacement cycle (and they're going the way of phones). Phones are designed for a 1 year replacement cycle (down from 2 years only recently).
This extends to nearly everything tech-related. They're trying to push TVs to a 3 year replacement cycle, they've got printers down to 3 years or less, they're trying to get fridges and other major appliances down to 5 or less (they're currently at 10, down from the 20-30 they used to be). Cars have been on a 3 year cycle for idiots for ages (36 month lease rolling into a new 36 month lease on a new vehicle).
Whether that design involves failure of the device, lack of support/updates, pushed updates to make the device run worse, etc. doesn't matter. The industry is built on planned obsolescence. It's not rare to see someone using a device past its intended replacement cycle, but shit is designed to get people onto a purchase cycle.
This was a lot of years ago. Things weren't as tightly controlled back then. '386 days...
The 386 debuted in 1985 (the beginning of the "'386 days"). The 486 debuted in 1989 (the end of the "'386 days").
You claimed that you were looking at hardware that was up to 10-15 years ahead in terms of performance and capability. That means you saw the equivalent of 1995-2000 level hardware in 1985, 1999-2004 level hardware in 1989, or any corresponding range in the years between. The Pentium 4 was released in 2000.
They are probably thinking something along the lines of "I make decisions for a multi billion dollar company. I probably know better than some dude on the Internet."
My father told me growing up that my stutter was just something I was doing for attention, and I could stop as soon as I wanted (and I was consequently grounded and sent to my room when I "didn't want to stop"). Surely he'd never have done that if there was actually a REASON for my stuttering.
And here you are today, with absolutely no stuttering in that post. More people should listen to their fathers.
I don't know what fantasy land you've shaped in mom's basement, but outside a tiny fraction of the US you need a car to make a living.
I figure the best and most ecological way to do this is to make the car I have last, and one of the way to make it last is to take care of it. Road salt is highly corrosive, the sand they put down turns to dust which in turn can etch the paint. Once rust starts, you can't really stop it and then you need a new car. And salt is corrosive to more than just the finish, it's corrosive to the undercarriage and mechanical systems, too.
But I suppose you think it's more ecological to just make more cars.
Fuck off, shitwick. You don't need to wash your fucking car 3 fucking times a week to prevent it from rusting out unless you live in a fucking salt mine. You're one of those aging failures who see their cars as a replacement for their underused, undersized penises.
Is there a way for sites to detect and block this?
No. The host is compromised.
Even if the bank mailed you a copy of their real cert, the compromised host could just update the malware to fetch the real cert and display that when the user tries to view the cert's details.
Even if the bank handed you a copy of a UNIQUE cert they use for ONLY for you, IN PERSON, and you handed them your own UNIQUE client cert, the compromised host could just watch all the legit shit happen when you log in the first time, then fuck you in the ass with that legit information.
Even "2-factor" authentication with a RSA clock won't help - these codes are good for a window of time (to allow people time to enter them and to allow for latency, clock skew, etc.). A compromised host can just use the same valid code rapidly within that window. Some systems require you to enter two distinct codes for a transaction, but this doesn't solve anything either as a compromised host can just trick the user into thinking they're moving $100 into their account when the real transaction is moving $10000 into the attackers account.
It seems that javascript is a common factor in most malware infection mechanisms. Is there a way to make javascript safe?
0: Don't write and host your own malicious javascript. 1: Don't host third-party javascript. 2: Don't host third-party content without sanitizing it to ensure it doesn't cause users to load other third-party shit.
1 & 2 can be combined into a simpler, more secure rule: Don't host third-party content.
It doesn't need to be anything more than HTML and images.
What if I want to convert to/from metric units? With Javascript I can click a button and have the page change all the values. Without Javascript I'd have to click a link instead!
What if I want to adjust the number of servings? With Javascript I can type in a value for servings and have the page tell me I need 2 and a half eggs. Without Javascript I'd have to type in a value and then hit the Enter key or click a Submit button!
What if I want to read user comments about the recipe? With Javascript, I can click a "Show Comments" "link" and just wait for the page to load comments. Without Javascript I'd have to click a "Show Comments" link and wait for the page to load comments!
What if I want to rate the recipe? With Javascript, I can give it a thumbs up or thumbs down, or give it 3.5/5 stars or little chef hats or whatever cutesy icon they want. Without Javascript, I'd have to click a button or use a drop down!
That's all avoidable by GPs "reasonable protection from the owner".
I wouldn't deal with the police at all. I'd deal with my card issuer. (They could design the scam this way, but I for one would be going to my card issuer's website and contacting them via whatever shitty webform they have.) I wouldn't believe the card issuer would hand deliver a new card to me. I don't think I've ever had to tell a rep my pin. I would never hand my old card in to some courier, I'd destroy it. I would not be reusing my pin.
The thing I don't like about it, is on the signature block on the back of the card I just write check id
Massive FAIL there, Psyko. If your card is ever stolen, instead of the CC company being responsible for losses you are!
Wrooooooooooooooooooooooooooooooooooooooooooooooooooooooooooong. An unauthorized charge is an unauthorized charge and you are not liable for unauthorized charges.
They measure the transfer rate of a single file in 1 direction. USB 3 gets you 5 Gbps. 500 MBps after overhead. 450 is definitely achievable in real-world use with a decent USB controller.
SCSI over USB only really adds queuing, improving speed when many small reads/writes are performed, and you'd need an SSD supporting SCSI and an enclosure/adapter supporting SCSI over USB. Further, for large transfers plain USB 3 is just as fast, while having the benefit of being cheaper, and more readily available and compatible than SCSI over USB. Of course, straight SATA III (via eSATA if you want) is still faster.
USB 3 gets you 5 Gbps and has to be handled by the CPU. SATA III gets you 6 Gbps without going through the CPU.
USB 3.1 promises to get you 10 Gbps (and lower overhead), but still has to go through the CPU. And Thunderbolt is just a convoluted and expensive way of piping a limited number of PCIe lanes to a random physical port and requiring the user to buy an expensive cable. 10 Gbps or 20 Gbps. 40 Gbps in the next revision.
SATA Express / M.2 can get you 32 Gbps using 4 PCIe 3.0 lanes or 2 PCIe 4.0 lanes wrapped up in NVMe. And you can always just throw more PCIe lanes at some controller (on-board or an via a PCIe slot) or some device directly if you want more bandwidth.
USB 3 will be the standard for external shit for a long time. The C connector and USB 3.1 are going to have a hell of a time gaining traction. For people who want performance, SATA Express / M.2 using NVMe or other direct PCIe solutions win.
When blackboxed, even "return 5" is indistinguishable from a true random number generator. People want noisy numbers, not random numbers. Which is a good thing, because a true random number generator will never exist.
So now we're assuming dark matter exists and we know how much of it is in any given galaxy? And we're correlating it to the "size" (singularity? event horizon? mass?) of a black hole?
I might as well correlate the amount of Leprechauns in my ass to the "price" of Bitcoin.
He knew that his Job was illegal, and he was making money for that. So dont be so naive. If you do a Job that something is fishy is up to you to continue
I'd like to see you trot that out the next time a cop or TLA employee is caught doing illegal shit as part of their illegal job duties and no one suffers any consequences.
Slashdot Mirror
I think the bigger problem is, what happens when we reach the long-tail of process development, and demand tapers off to the point they can't fund further R&D?
IE: Systems are "good enough" and people go from buying one every 3 years to "only when they break". That could be 10+ years.
I suppose Intel would just follow the carrot to the next profitable market like they are pushing Atom CPUs lately?
Design them to be replaced every 3 years.
Servers are designed for a 5 year replacement cycle.
Desktops are designed for a 3-5 year replacement cycle.
Laptops are designed for a 3 year replacement cycle.
Tablets are designed for a 2 year replacement cycle (and they're going the way of phones).
Phones are designed for a 1 year replacement cycle (down from 2 years only recently).
This extends to nearly everything tech-related. They're trying to push TVs to a 3 year replacement cycle, they've got printers down to 3 years or less, they're trying to get fridges and other major appliances down to 5 or less (they're currently at 10, down from the 20-30 they used to be). Cars have been on a 3 year cycle for idiots for ages (36 month lease rolling into a new 36 month lease on a new vehicle).
Whether that design involves failure of the device, lack of support/updates, pushed updates to make the device run worse, etc. doesn't matter. The industry is built on planned obsolescence. It's not rare to see someone using a device past its intended replacement cycle, but shit is designed to get people onto a purchase cycle.
And you are just too stupid to form complete sentences.
This was a lot of years ago. Things weren't as tightly controlled back then. '386 days...
The 386 debuted in 1985 (the beginning of the "'386 days").
The 486 debuted in 1989 (the end of the "'386 days").
You claimed that you were looking at hardware that was up to 10-15 years ahead in terms of performance and capability.
That means you saw the equivalent of 1995-2000 level hardware in 1985, 1999-2004 level hardware in 1989, or any corresponding range in the years between.
The Pentium 4 was released in 2000.
Care to revise your bullshit claim?
They've decided to hit 7nm and then call it a day.
I asked Gordon Moore about this and he said it would be illegal.
I'm really fucking tired of people referencing Moore's Law incorrectly.
Moore's Law is only about the number of transistors doubling every 2 years.
I'm also really fucking tired of people saying "The goggles! They do nothing!" when the quote is "My eyes! The goggles do nothing!".
WHAT ARE THEY THINKING?!
They are probably thinking something along the lines of "I make decisions for a multi billion dollar company. I probably know better than some dude on the Internet."
And they are probably wrong.
My father told me growing up that my stutter was just something I was doing for attention, and I could stop as soon as I wanted (and I was consequently grounded and sent to my room when I "didn't want to stop"). Surely he'd never have done that if there was actually a REASON for my stuttering.
And here you are today, with absolutely no stuttering in that post. More people should listen to their fathers.
Answer is "no".
SEGA is not "the next Atari". They've been a fucking dead husk for over a decade.
No, fuck you.
I don't know what fantasy land you've shaped in mom's basement, but outside a tiny fraction of the US you need a car to make a living.
I figure the best and most ecological way to do this is to make the car I have last, and one of the way to make it last is to take care of it. Road salt is highly corrosive, the sand they put down turns to dust which in turn can etch the paint. Once rust starts, you can't really stop it and then you need a new car. And salt is corrosive to more than just the finish, it's corrosive to the undercarriage and mechanical systems, too.
But I suppose you think it's more ecological to just make more cars.
Fuck off, shitwick.
You don't need to wash your fucking car 3 fucking times a week to prevent it from rusting out unless you live in a fucking salt mine. You're one of those aging failures who see their cars as a replacement for their underused, undersized penises.
Neither have electrons.
I think you misspelled "erection" and forgot that not everyone has a dick as tiny and impotent as your own.
They're real - ask your mommy.
I wash mine 2-3 times per week
Fuck you, you selfish, vain, shallow eco terrorist.
If your dealer is that big of a douchbag
It's a car dealer. Douchebag is redundant.
It's a BMW dealer. Big douchebag is an understatement.
Security isn't hard
LOOOOOOOOOOOOOOOOOOOOL
It's time to play "Slashdot Feud"!
We asked 100 Slashdotters who they thought OP was after reading his post.
Survey says:
1) An Ignorant Piece of Shit Talking Out of His Ass - 57
2) Your Typical Troll - 36
3) Frosty Piss - 6
4) APK - 1
Is there a way for sites to detect and block this?
No. The host is compromised.
Even if the bank mailed you a copy of their real cert, the compromised host could just update the malware to fetch the real cert and display that when the user tries to view the cert's details.
Even if the bank handed you a copy of a UNIQUE cert they use for ONLY for you, IN PERSON, and you handed them your own UNIQUE client cert, the compromised host could just watch all the legit shit happen when you log in the first time, then fuck you in the ass with that legit information.
Even "2-factor" authentication with a RSA clock won't help - these codes are good for a window of time (to allow people time to enter them and to allow for latency, clock skew, etc.). A compromised host can just use the same valid code rapidly within that window. Some systems require you to enter two distinct codes for a transaction, but this doesn't solve anything either as a compromised host can just trick the user into thinking they're moving $100 into their account when the real transaction is moving $10000 into the attackers account.
True one-time use keys don't fix this either.
Whoosh.
It seems that javascript is a common factor in most malware infection mechanisms. Is there a way to make javascript safe?
0: Don't write and host your own malicious javascript.
1: Don't host third-party javascript.
2: Don't host third-party content without sanitizing it to ensure it doesn't cause users to load other third-party shit.
1 & 2 can be combined into a simpler, more secure rule: Don't host third-party content.
It's a fucking recipe for a pork roast.
It doesn't need to be anything more than HTML and images.
What if I want to convert to/from metric units? With Javascript I can click a button and have the page change all the values. Without Javascript I'd have to click a link instead!
What if I want to adjust the number of servings? With Javascript I can type in a value for servings and have the page tell me I need 2 and a half eggs. Without Javascript I'd have to type in a value and then hit the Enter key or click a Submit button!
What if I want to read user comments about the recipe? With Javascript, I can click a "Show Comments" "link" and just wait for the page to load comments. Without Javascript I'd have to click a "Show Comments" link and wait for the page to load comments!
What if I want to rate the recipe? With Javascript, I can give it a thumbs up or thumbs down, or give it 3.5/5 stars or little chef hats or whatever cutesy icon they want. Without Javascript, I'd have to click a button or use a drop down!
That's all avoidable by GPs "reasonable protection from the owner".
I wouldn't deal with the police at all. I'd deal with my card issuer. (They could design the scam this way, but I for one would be going to my card issuer's website and contacting them via whatever shitty webform they have.)
I wouldn't believe the card issuer would hand deliver a new card to me.
I don't think I've ever had to tell a rep my pin.
I would never hand my old card in to some courier, I'd destroy it.
I would not be reusing my pin.
Massive FAIL there, Psyko. If your card is ever stolen, instead of the CC company being responsible for losses you are!
Wrooooooooooooooooooooooooooooooooooooooooooooooooooooooooooong.
An unauthorized charge is an unauthorized charge and you are not liable for unauthorized charges.
OffJacker or OffJack.
They measure the transfer rate of a single file in 1 direction.
USB 3 gets you 5 Gbps. 500 MBps after overhead. 450 is definitely achievable in real-world use with a decent USB controller.
SCSI over USB only really adds queuing, improving speed when many small reads/writes are performed, and you'd need an SSD supporting SCSI and an enclosure/adapter supporting SCSI over USB. Further, for large transfers plain USB 3 is just as fast, while having the benefit of being cheaper, and more readily available and compatible than SCSI over USB. Of course, straight SATA III (via eSATA if you want) is still faster.
USB 3 gets you 5 Gbps and has to be handled by the CPU.
SATA III gets you 6 Gbps without going through the CPU.
USB 3.1 promises to get you 10 Gbps (and lower overhead), but still has to go through the CPU.
And Thunderbolt is just a convoluted and expensive way of piping a limited number of PCIe lanes to a random physical port and requiring the user to buy an expensive cable. 10 Gbps or 20 Gbps. 40 Gbps in the next revision.
SATA Express / M.2 can get you 32 Gbps using 4 PCIe 3.0 lanes or 2 PCIe 4.0 lanes wrapped up in NVMe.
And you can always just throw more PCIe lanes at some controller (on-board or an via a PCIe slot) or some device directly if you want more bandwidth.
USB 3 will be the standard for external shit for a long time. The C connector and USB 3.1 are going to have a hell of a time gaining traction.
For people who want performance, SATA Express / M.2 using NVMe or other direct PCIe solutions win.
When blackboxed, even "return 5" is indistinguishable from a true random number generator.
People want noisy numbers, not random numbers. Which is a good thing, because a true random number generator will never exist.
Bazinga.
So now we're assuming dark matter exists and we know how much of it is in any given galaxy?
And we're correlating it to the "size" (singularity? event horizon? mass?) of a black hole?
I might as well correlate the amount of Leprechauns in my ass to the "price" of Bitcoin.
He knew that his Job was illegal, and he was making money for that. So dont be so naive. If you do a Job that something is fishy is up to you to continue
I'd like to see you trot that out the next time a cop or TLA employee is caught doing illegal shit as part of their illegal job duties and no one suffers any consequences.