For similar reasons, I'm still not sold on the idea of cloud-based password managers. That seems like a problem just waiting to happen.
If you have a highly secure password/passphrase, which you really should, what's the risk? Unless the encryption is not as good as I've been led to believe, it's not hard to make a password that would take hundreds to thousands of years to crack with current technology. Change your master password every ten or twenty years and you should be OK even if someone gets hold of your encrypted password storage.
A hundred people can operate a multi gigawatt nuclear reactor. How many installers / maintainers do you need to operate 100000 individual solar installations on rooftops to do the same thing?
I couldn't tell you what the ramifications are but the higher measurement is about 3.5% higher than the lower one. So not something that clearly indicates some kind of massive error like one group was measuring the wrong universe or something, but sounds pretty significant.
Umm, have a basic map on the phone. It doesn't have to be super detailed though it could have a cache of your neighborhood.
I doubt many people would have much use for a nationwide map with nothing more than interstate highways, and preloading the device with a detailed map of the US, let alone the world, would be impossible or at least prohibitive. You can certainly instruct the app to cache an area for offline use, but it's not going to do that without you instructing it to (which is appropriate), and the first time it gets that map it has to come from somewhere.
There are map apps that work differently if you don't like the ones from Google and Apple. You can choose what maps to download ahead of time (oh no, FROM THE CLOUD!) and then you don't need an internet connection to use them. Google and Apple have designed their apps to match how they think most people want to use them, and I think they're probably right. But no solution will work for everyone.
The question is how to account for "for now"? How long does a person have to stop looking for a job before they should no longer be considered unemployed? I don't think there's any clearly correct answer, but maybe the government's accounting is too short.
I acknowledged that. It is possible internet service is by the byte, in which case there is some extremely small marginal cost.
And is the unit of measure a single download, or a thousand, of for hugely popular items literally millions?
When talking marginal cost, it's a single download.
If your perspective is that the pricing is to be challenged, I expect you to define costs as largely fixed and minimal.
I'm not looking to challenge anyone's pricing, just describing the nature of the product.
If you examine the 'production' effort, you can readily find that costs scale with sales. For digital goods, this is mostly hugely nonlinear. And so it's hard to accept.
If you can find a cost that goes up when one additional person downloads a book or song or app (other than that tiny bandwidth charge) then that would be evidence that the good is rivalrous. But I haven't seen that yet.
Ps- we are also blithely giving market forces and monopolistic market share a pass.
If by giving them a pass you mean not discussing them, yes. There's lots of stuff we're not discussing.
Microsoft Office is a virtual monopoly. Candy Crush is a phenomenal but brief success. Both have huge total distribution costs, but marginal costs seem low. One finances continued development and less profitable projects, the other is a windfall...
I'm not sure what your point is. That fixed costs matter? That monopolies affect markets? Something else?
If it costs more to serve a million downloads a month than it does one because you need a CDN or a bigger pipe or whatever, that isn't a marginal cost, it's still a fixed cost. A marginal cost is how much it costs to serve ONE additional unit. For a physical product, if you need a bigger truck because you're shipping more units, that truck is not a marginal cost. If you need to use another box to ship another unit, that box is a marginal cost, because it's a cost associated with a single additional sale.
Digital goods are not without costs. But they are very close to without marginal costs.
No, successive downloading of digital goods is not without cost. Servers must be maintained and connections as well. Security is an expense.
Those are fixed costs that are incurred whether there are zero downloads or 10,000. Rivalry is about marginal costs. In fact the definition I quoted even has the word "marginal" in it. If you don't know the difference between fixed and marginal costs, you should look that up because you are not going to understand rivalrous and non-rivalrous goods without understanding that first.
A good is considered non-rivalrous or non-rival if, for any level of production, the cost of providing it to a marginal (additional) individual is zero.
You could quibble about companies being charged for internet service by the megabyte or something like that, but in any meaningful sense the cost of providing another copy of a digital good is essentially zero. They're about as non-rivalrous as anything can get.
The only way those IP laws will help you is if you successfully sue your competitor (and the next competitor who tries it, and the next...). And if you think marketing is expensive, try lawsuits.
This is a great fallacy with IP, trying to categorize it as somehow different form all other products.
It's not different from all other products, but it is different from medicine and many other products in that it's non-excludable and non-rivalrous. That doesn't mean the government needs to take special measures to protect it, but it does change things.
Pretty sure some combination of 1 and 2 until it gets so bad everyone can see we need to do 3. Given how things are going with climate change, I fear that it will be cities on fire bad before we make that change.
If you make a breakthrough with this at a subsequent job
Sounds like that is not what they are arguing though.
ZeniMax detailed its case that Carmack, while still an employee at Id Software, "designed the specifications and functionality embodied in the Rift SDK and directed its development."... Carmack allegedly used "copyrighted computer code, trade secret information, and technical know-how" from his time at ZeniMax after he moved to Oculus as CTO in 2013.
They could run over someone and that makes them dangerous enough.
If we use that standard, the technology will never be permitted (and neither would any other new thing). Testing is necessary, yes, but we need not fall victim to the perfect solution fallacy.
Apparently it's hard to read street signs. The other day picking up my kid from school I saw two people parked right in front of a sign that said "No parking, stopping or standing at any time". Derp. Fortunately a cop came along and made them move.
I have no idea what you're talking about. What "pass" am I giving them? I said that if a company is found liable they will pay. If individuals are found guilty of crimes, they will be sentenced. I said the technology is not ready for mass adoption but that eventually it's going to be good enough to save a lot of lives. You seem to be responding to something you wish I said rather than what I actually said.
An automatic car will pretty much need to understand the world to be as safe as a human. For example, when a human sees a child's toy suddenly cast out into the road, they slow down and look for the child that might run after it.
For a while, deficiencies in their understanding will be more than made up for by improvements in other areas, such as sensors and reaction time.
So now you're saying automated cars today are as safe as anything else on public roads?
I never said any such thing. I don't know why you think I did.
These cars could and should be tested in a controlled environment.
They just need to understand a subset of the world well enough to safely navigate. There is a huge amount of information you take in about your surroundings (and mostly discard) that an automatic car just doesn't need to be interested in or understand.
Why would we allow AI on the market that is in some way not safe?
Every product on the market is in some way not safe, including cars. They just need to be safe enough, with the standards for safe enough continually changing.
I don't care how many people it might save one day, you can't guarantee that result so people should not be injured or killed today.
I'm not sure what you're saying. That the technology is not ready for widespread adoption? That's obvious.
What are these companies going to say to the parents when one of these vehicles kills a child who was sitting on a curb with the sun shining a certain way so as to be in a blind spot?
If found liable, they'll make a payment (or perhaps just settle out of court). Exactly like what happens now when a company is responsible for someone being injured.
Maybe now, but my first computer I got for college cost $2000 and it wasn't top of the line. You could argue that college pays for itself and so it was for profit, but it was a personal use machine, not for business or other organization. That's just what computers cost back then, you paid it or you went without (and by without I don't mean just use your phone, I mean no computing whatsoever).
Slashdot Mirror
For similar reasons, I'm still not sold on the idea of cloud-based password managers. That seems like a problem just waiting to happen.
If you have a highly secure password/passphrase, which you really should, what's the risk? Unless the encryption is not as good as I've been led to believe, it's not hard to make a password that would take hundreds to thousands of years to crack with current technology. Change your master password every ten or twenty years and you should be OK even if someone gets hold of your encrypted password storage.
As long as manufacturing and distributing parts for fossil fuel power plants is counted too, that's legitimate. But I don't know if it is.
A hundred people can operate a multi gigawatt nuclear reactor. How many installers / maintainers do you need to operate 100000 individual solar installations on rooftops to do the same thing?
I don't think you need anyone to operate them.
I couldn't tell you what the ramifications are but the higher measurement is about 3.5% higher than the lower one. So not something that clearly indicates some kind of massive error like one group was measuring the wrong universe or something, but sounds pretty significant.
Umm, have a basic map on the phone. It doesn't have to be super detailed though it could have a cache of your neighborhood.
I doubt many people would have much use for a nationwide map with nothing more than interstate highways, and preloading the device with a detailed map of the US, let alone the world, would be impossible or at least prohibitive. You can certainly instruct the app to cache an area for offline use, but it's not going to do that without you instructing it to (which is appropriate), and the first time it gets that map it has to come from somewhere.
There are map apps that work differently if you don't like the ones from Google and Apple. You can choose what maps to download ahead of time (oh no, FROM THE CLOUD!) and then you don't need an internet connection to use them. Google and Apple have designed their apps to match how they think most people want to use them, and I think they're probably right. But no solution will work for everyone.
You're expecting your phone to show a map of your location without downloading it from the internet? How would that work exactly?
The question is how to account for "for now"? How long does a person have to stop looking for a job before they should no longer be considered unemployed? I don't think there's any clearly correct answer, but maybe the government's accounting is too short.
'very close' is the point I was making.
I acknowledged that. It is possible internet service is by the byte, in which case there is some extremely small marginal cost.
And is the unit of measure a single download, or a thousand, of for hugely popular items literally millions?
When talking marginal cost, it's a single download.
If your perspective is that the pricing is to be challenged, I expect you to define costs as largely fixed and minimal.
I'm not looking to challenge anyone's pricing, just describing the nature of the product.
If you examine the 'production' effort, you can readily find that costs scale with sales. For digital goods, this is mostly hugely nonlinear. And so it's hard to accept.
If you can find a cost that goes up when one additional person downloads a book or song or app (other than that tiny bandwidth charge) then that would be evidence that the good is rivalrous. But I haven't seen that yet.
Ps- we are also blithely giving market forces and monopolistic market share a pass.
If by giving them a pass you mean not discussing them, yes. There's lots of stuff we're not discussing.
Microsoft Office is a virtual monopoly. Candy Crush is a phenomenal but brief success. Both have huge total distribution costs, but marginal costs seem low. One finances continued development and less profitable projects, the other is a windfall...
I'm not sure what your point is. That fixed costs matter? That monopolies affect markets? Something else?
If it costs more to serve a million downloads a month than it does one because you need a CDN or a bigger pipe or whatever, that isn't a marginal cost, it's still a fixed cost. A marginal cost is how much it costs to serve ONE additional unit. For a physical product, if you need a bigger truck because you're shipping more units, that truck is not a marginal cost. If you need to use another box to ship another unit, that box is a marginal cost, because it's a cost associated with a single additional sale.
Digital goods are not without costs. But they are very close to without marginal costs.
No, successive downloading of digital goods is not without cost. Servers must be maintained and connections as well. Security is an expense.
Those are fixed costs that are incurred whether there are zero downloads or 10,000. Rivalry is about marginal costs. In fact the definition I quoted even has the word "marginal" in it. If you don't know the difference between fixed and marginal costs, you should look that up because you are not going to understand rivalrous and non-rivalrous goods without understanding that first.
A good is considered non-rivalrous or non-rival if, for any level of production, the cost of providing it to a marginal (additional) individual is zero.
You could quibble about companies being charged for internet service by the megabyte or something like that, but in any meaningful sense the cost of providing another copy of a digital good is essentially zero. They're about as non-rivalrous as anything can get.
https://en.wikipedia.org/wiki/...
The only way those IP laws will help you is if you successfully sue your competitor (and the next competitor who tries it, and the next...). And if you think marketing is expensive, try lawsuits.
This is a great fallacy with IP, trying to categorize it as somehow different form all other products.
It's not different from all other products, but it is different from medicine and many other products in that it's non-excludable and non-rivalrous. That doesn't mean the government needs to take special measures to protect it, but it does change things.
Pretty sure some combination of 1 and 2 until it gets so bad everyone can see we need to do 3. Given how things are going with climate change, I fear that it will be cities on fire bad before we make that change.
Sometimes even high quality ones.
My Win7 computer has almost zero CPU usage when idle, across all cores. Maybe I'm the exception.
I don't know about yours but my cell phone does a lot more than browse the web and access email.
If you make a breakthrough with this at a subsequent job
Sounds like that is not what they are arguing though.
ZeniMax detailed its case that Carmack, while still an employee at Id Software, "designed the specifications and functionality embodied in the Rift SDK and directed its development." ... Carmack allegedly used "copyrighted computer code, trade secret information, and technical know-how" from his time at ZeniMax after he moved to Oculus as CTO in 2013.
They could run over someone and that makes them dangerous enough.
If we use that standard, the technology will never be permitted (and neither would any other new thing). Testing is necessary, yes, but we need not fall victim to the perfect solution fallacy.
https://yandoo.wordpress.com/2...
OK, I thought your point was that "low end builds will rise to the price of high end builds".
Apparently it's hard to read street signs. The other day picking up my kid from school I saw two people parked right in front of a sign that said "No parking, stopping or standing at any time". Derp. Fortunately a cop came along and made them move.
I have no idea what you're talking about. What "pass" am I giving them? I said that if a company is found liable they will pay. If individuals are found guilty of crimes, they will be sentenced. I said the technology is not ready for mass adoption but that eventually it's going to be good enough to save a lot of lives. You seem to be responding to something you wish I said rather than what I actually said.
An automatic car will pretty much need to understand the world to be as safe as a human. For example, when a human sees a child's toy suddenly cast out into the road, they slow down and look for the child that might run after it.
For a while, deficiencies in their understanding will be more than made up for by improvements in other areas, such as sensors and reaction time.
So now you're saying automated cars today are as safe as anything else on public roads?
I never said any such thing. I don't know why you think I did.
These cars could and should be tested in a controlled environment.
And they are.
Why the hell not?
They just need to understand a subset of the world well enough to safely navigate. There is a huge amount of information you take in about your surroundings (and mostly discard) that an automatic car just doesn't need to be interested in or understand.
Why would we allow AI on the market that is in some way not safe?
Every product on the market is in some way not safe, including cars. They just need to be safe enough, with the standards for safe enough continually changing.
I don't care how many people it might save one day, you can't guarantee that result so people should not be injured or killed today.
I'm not sure what you're saying. That the technology is not ready for widespread adoption? That's obvious.
What are these companies going to say to the parents when one of these vehicles kills a child who was sitting on a curb with the sun shining a certain way so as to be in a blind spot?
If found liable, they'll make a payment (or perhaps just settle out of court). Exactly like what happens now when a company is responsible for someone being injured.
Maybe now, but my first computer I got for college cost $2000 and it wasn't top of the line. You could argue that college pays for itself and so it was for profit, but it was a personal use machine, not for business or other organization. That's just what computers cost back then, you paid it or you went without (and by without I don't mean just use your phone, I mean no computing whatsoever).