For a peer-to-peer system, it has a lot of heirarchical order.
In what way? Maybe DNS has some hierarchy to it, but ultimately the internet is peer-to-peer. It's certainly not a broadcast network.
The FTP servers on the Internet constitute P2P file sharing. Same with web servers. You can install apache on your computer and I can install it on mine, and then we be peers who have access to share each others' files. Google's search engine is the tool that most of us use to indicate what files are available, as well as to find them-- but really, Google is just another peer on the network.
But now, watch out, if you use a protocol other than HTTP and a search engine that's not Google (and maybe decentralized), suddenly you're everyone assumes you're doing something illegal. They want to make peer-to-peer communications illegal, but they're failing to understand that there's no meaningful technical distinction between that "suspicious P2P file sharing" and "normal legitimate Internet traffic."
Well, cheap enough that people don't have to worry when they need to be replaced, or else durable enough that people don't have to worry about them needing to be replaced. Really either one will do.
Any security breaches are not the fault of P2P per se. Why was there a computer with classified documents where the user was allowed to install software and connect directly to the Internet? The user could have installed Apache and made the entire hard drive accessible through HTTP at that point.
Ultimately the entire Internet is peer-to-peer. All these "P2P" applications do is make it easier for the peers to find each other.
As someone with some
libertarian leanings, I definitely don't agree with the idea of the government subsidizing antivirus software. Make all the arguments you want, but I'd sooner be convinced that the government should provide health care.
Why? Because at least there's a genuine need for health care. I have never, not in my entire life, been infected with a virus myself. I run Windows machines, Linux machines, OSX machines. I've seen plenty of infections on Windows, but most of them, the overwhelming majority, started from someone running something they shouldn't have. Most of the rest could have been fixed with a good firewall.
So if the government is going to subsidize something in order to improve security, I'd much rather they subsidize computer education. You can't get an antivirus that will protect you from every stupid user action, but maybe if you educate the users, they'll commit fewer stupid actions.
If you subsidize the antivirus vendors, you're really only subsidizing the patching of security holes that arguably shouldn't exist in the first place. You'll also effectively be removing the economic incentive to remove those holes or to make software better, more secure, and more immune. You'll be saying, "Don't fix these problems, or else the government will give you less money."
And that's the biggest problem with government subsidies. It's not the cost itself, it's that they often provide an incentive to not-fix problems.
Nah... too much whacky geometry IMO. Interesting enough if you're going to do the whole Copernicus->Kepler->Newton thing, but otherwise I'd leave it be.
I'm not claiming the concept is useless or anything. I would just wonder, if Apple really wanted to provide a service like that, how long would it take them to build it from scratch? I find it hard to believe that the technology is worth very much.
Then there's the usebase. This year everyone is on Twitter, but what about next year? Just a few years ago, Friendster was huge. Then it was MySpace. Now it's Facebook. There's no reason to think that someon can't supplant Twitter and take their userbase.
And even if CRA was the cause of the mortgage foreclosures, that still wouldn't make it sufficient to cause the financial collapse. The banks were playing fast and loose with the rules, including selling more insurance than they could afford to pay out to people who had nothing to do with what was being insured (credit default swaps).
Now I'm not going to pretend that I have a deep knowledge of what I'm talking about, but I've heard/read several respected economists who claim that all the mortgage foreclosures that we've had would still have been manageable, but the real problem is the credit default swaps.
Is there really much point in buying twitter? How difficult a thing is it to write that application? Or is the purpose almost entirely to grab the existing users?
And how would this fit into Apple's strategy? I could think of much better ways that Apple could extend their MobileMe service.
I can't remember the names of the books, but I remember finding William Harvey and Antoine Lavoisier interesting back when I was in school. Harvey studied the circulation of blood and Lavoisier did some early work in chemistry, including the discovery of oxygen.
I was also pretty interested in the electromagnetism work by Faraday and Benjamin Franklin, but I remember less about them. By the time it got to Maxwell it was a little too much work. Strangely, I found Einstein much easier to understand than Maxwell, even though the theory itself is a bit whackier.
But if you're more interested in the process of figuring things out than the actual discoveries, then I think Harvey, Lavoisier, and Einstein were all pretty interesting.
But how are they getting the 1% number for "internet usage"? Which sites are they surveying? How are they determining what constitutes a unique machine? If you track IP addresses, you might get one visitor recorded when lots of people are behind a NAT. Even if you track cookies, that can be thrown off if people aren't storing cookies.
The problem with these sorts of stats is that you'll get some sites reporting 40% of their visitors use Linux while others report 0% use Linux. It's not that either is "wrong", but they're attracting different visitors.
What do big dogs do when small dogs start to threaten their dominance? They try to kill them. I actually prefer the "slowly but surely until it's too late" scenario.
Are you sure it's not already too late? Microsoft has been trying to kill Linux for years, and the damn thing won't die. And how can you kill a think like that? Yeah, you could drive Redhat and Canonical out of business, I suppose... and then anyone on earth would be free to pick up where those companies left off and keep developing.
Well I'm not sure "rounding" is quite the word I'd use, but definitely something along those lines. At least in my sense of the word, the idea of an "estimate" is that it's something that you know the answer to, and you might know very accurately, but you don't know very precisely.
Like if you asked me how many computers are in the room I'm in now, without counting, I'd estimate 12. Now it may only be 10 or it may be 14, but when I say 12, that's not simply a guess. I know every computer in the room and could shut my eyes and count in my head and probably come up with an exact number, and I know for absolute sure that it's no more than 14, and definitely at least 10.
So that's an estimate. Not that I particularly like the word "guesstimate".
What you're saying would be absolutely correct if economics were a zero-sum game, and all of us were making the same amount of money. But the reality is that the distribution of wealth matters, and what is done with that money matters.
Great, all the companies can move to poor countries where they can pay the workers next to nothing and pay very little in taxes. Then they can make all the money they want without paying anyone, selling their products to...
Minimizing revenue is a silly goal. It's trivially easy: stop collecting taxes. However, you're left with the choice, then, of unsupportable deficits or anarchy.
I think the question of "what format is suitable" has been answered repeatedly: HTML is fine. PDF is OK. RTF is probably OK.
Each of those formats may do well enough, but none of them are perfect or without problems. Only PDF really allows for sufficient layout control (depending, of course, on your view of "sufficient"), and yet at the same time it doesn't really allow for the content to be maximized on the fly for different sized/shaped displays. Even HTML, which you might argue is the best option, isn't perfect. It's being revised for newer browsers because of various problems.
But regardless, I'm not making excuses. I'm not saying that they can't just pick one of these formats and get to work. I'm saying that if you want to have a single device that can read all the various newspapers with uniform functionality, they need to get together and decide how they're going to do it. It will work best if they standardize.
If it's all HTML, that's fine. They'll probably want to set guidelines for styling. Do you set font size by a set number of px or em? Do you want to set anything as fixed width, or should it all be liquid? What I'm trying to get at is that the web is great, but it's also a mess. If you want to productize newspapers as a subscription service, you probably want to standardize to make the experience as clean as possible.
Part of the problem with that is that people like to control layout, and some kinds of information almost demand a particular presentation. So eventually, you're going to want something more than RTF, probably including something where pictures can be dropped in easily.
That leaves HTML as an obvious option, but then you're still going to want to standardize on a particular version of HTML and CSS, if not a particular rendering engine with particular options enabled to allow consistent rendering. Do you want to make it simplified to avoid battery drain? No javascript support, maybe?
I'm really not trying to advocate a particular choice. I'm just trying to point out that these newspapers need to get into a room with tech people and hash out a standard that provides all the functionality they're going to want. Once they have that, they can essentially publish a spec for what constitutes a "supported reader".
Neither. You'd specify a generic webservice interface with which each reader can specify which features it supports, and the 'newspaper' server would send it a PDF that comes closest.
Sure, that'd be one way to attack the problem. Still, my larger point was that you'll want to standardize things across newspapers. Getting everyone to agree to an interface for providing the appropriate PDF from various versions (which in itself implies the newspapers will be generating various standard PDF versions according to some defined spec) would be the sort of standardization I'm talking about.
It's not that SMS is bad. SMS is basically fine, but it's limited and expensive. That's right, contrary to your point about them being cheaper (than full cellular data plans), SMS messages are expensive for consumers. Last I checked, it cost something like $0.10/message, or something like $20 for an unlimited texting plans. Unlimited data plans aren't necessarily much more than that. And for what? As you point out, they're already built-in with the phones and the cell networks and have very little overhead.
And what format do you suggest? TXT doesn't have formatting, RTF doesn't have very good formating. Even HTML is difficult to get complex formatting that will render reliably. PDF? Ok, that sounds pretty good, but what if PDF lacks some feature that some of the newspapers really want to be able to include?
My point is, it doesn't really need to be a pre-existing format, but whatever format they use, they should standardize so that ANY reader can use it, or at bare minimum, there is some reader out there that can read all the newspapers.
And even if they settled on PDF, for example, there might still be issues to hash out. For example, do all readers' screens use the same aspect ratio? The same resolution? The same number of colors? I bet you I could make a PDF that would look great on 1024x768 with 4 shades of gray but would be flat-out illegible at 800x600 monochrome. So that raises the question, do you try to force papers to shoot for the lowest common denominator, or do you let them establish a baseline standards that readers would have to meet in order to be supported?
Yeah, my point wasn't that resolution doesn't matter, but rather when you start talking about resolutions like "1024x1280", I think you might be taking the wrong approach. Instead of treating it like a computer screen and bumping up through the normal screen resolutions, it seems to me like the better thing to do is start from the question, "What are people going to be reading on these things, and given that, what is the optimal page size?" Once you have a page size, you figure out the DPI necessary for comfortable reading, and go from there.
Slashdot Mirror
Yeah, I heard it's made out of Element 8 as well as a highly explosive chemical used in H-bombs.
That sounds almost as dangerous as dihydrogen monoxide.
For a peer-to-peer system, it has a lot of heirarchical order.
In what way? Maybe DNS has some hierarchy to it, but ultimately the internet is peer-to-peer. It's certainly not a broadcast network.
The FTP servers on the Internet constitute P2P file sharing. Same with web servers. You can install apache on your computer and I can install it on mine, and then we be peers who have access to share each others' files. Google's search engine is the tool that most of us use to indicate what files are available, as well as to find them-- but really, Google is just another peer on the network.
But now, watch out, if you use a protocol other than HTTP and a search engine that's not Google (and maybe decentralized), suddenly you're everyone assumes you're doing something illegal. They want to make peer-to-peer communications illegal, but they're failing to understand that there's no meaningful technical distinction between that "suspicious P2P file sharing" and "normal legitimate Internet traffic."
Well, cheap enough that people don't have to worry when they need to be replaced, or else durable enough that people don't have to worry about them needing to be replaced. Really either one will do.
Any security breaches are not the fault of P2P per se. Why was there a computer with classified documents where the user was allowed to install software and connect directly to the Internet? The user could have installed Apache and made the entire hard drive accessible through HTTP at that point.
Ultimately the entire Internet is peer-to-peer. All these "P2P" applications do is make it easier for the peers to find each other.
As someone with some libertarian leanings, I definitely don't agree with the idea of the government subsidizing antivirus software. Make all the arguments you want, but I'd sooner be convinced that the government should provide health care.
Why? Because at least there's a genuine need for health care. I have never, not in my entire life, been infected with a virus myself. I run Windows machines, Linux machines, OSX machines. I've seen plenty of infections on Windows, but most of them, the overwhelming majority, started from someone running something they shouldn't have. Most of the rest could have been fixed with a good firewall.
So if the government is going to subsidize something in order to improve security, I'd much rather they subsidize computer education. You can't get an antivirus that will protect you from every stupid user action, but maybe if you educate the users, they'll commit fewer stupid actions.
If you subsidize the antivirus vendors, you're really only subsidizing the patching of security holes that arguably shouldn't exist in the first place. You'll also effectively be removing the economic incentive to remove those holes or to make software better, more secure, and more immune. You'll be saying, "Don't fix these problems, or else the government will give you less money."
And that's the biggest problem with government subsidies. It's not the cost itself, it's that they often provide an incentive to not-fix problems.
Nah... too much whacky geometry IMO. Interesting enough if you're going to do the whole Copernicus->Kepler->Newton thing, but otherwise I'd leave it be.
I'm not claiming the concept is useless or anything. I would just wonder, if Apple really wanted to provide a service like that, how long would it take them to build it from scratch? I find it hard to believe that the technology is worth very much.
Then there's the usebase. This year everyone is on Twitter, but what about next year? Just a few years ago, Friendster was huge. Then it was MySpace. Now it's Facebook. There's no reason to think that someon can't supplant Twitter and take their userbase.
And even if CRA was the cause of the mortgage foreclosures, that still wouldn't make it sufficient to cause the financial collapse. The banks were playing fast and loose with the rules, including selling more insurance than they could afford to pay out to people who had nothing to do with what was being insured (credit default swaps).
Now I'm not going to pretend that I have a deep knowledge of what I'm talking about, but I've heard/read several respected economists who claim that all the mortgage foreclosures that we've had would still have been manageable, but the real problem is the credit default swaps.
Is there really much point in buying twitter? How difficult a thing is it to write that application? Or is the purpose almost entirely to grab the existing users?
And how would this fit into Apple's strategy? I could think of much better ways that Apple could extend their MobileMe service.
The whole thing seems slightly fishy to me.
I can't remember the names of the books, but I remember finding William Harvey and Antoine Lavoisier interesting back when I was in school. Harvey studied the circulation of blood and Lavoisier did some early work in chemistry, including the discovery of oxygen.
I was also pretty interested in the electromagnetism work by Faraday and Benjamin Franklin, but I remember less about them. By the time it got to Maxwell it was a little too much work. Strangely, I found Einstein much easier to understand than Maxwell, even though the theory itself is a bit whackier.
But if you're more interested in the process of figuring things out than the actual discoveries, then I think Harvey, Lavoisier, and Einstein were all pretty interesting.
But how are they getting the 1% number for "internet usage"? Which sites are they surveying? How are they determining what constitutes a unique machine? If you track IP addresses, you might get one visitor recorded when lots of people are behind a NAT. Even if you track cookies, that can be thrown off if people aren't storing cookies.
The problem with these sorts of stats is that you'll get some sites reporting 40% of their visitors use Linux while others report 0% use Linux. It's not that either is "wrong", but they're attracting different visitors.
What do big dogs do when small dogs start to threaten their dominance? They try to kill them. I actually prefer the "slowly but surely until it's too late" scenario.
Are you sure it's not already too late? Microsoft has been trying to kill Linux for years, and the damn thing won't die. And how can you kill a think like that? Yeah, you could drive Redhat and Canonical out of business, I suppose... and then anyone on earth would be free to pick up where those companies left off and keep developing.
Well I'm not sure "rounding" is quite the word I'd use, but definitely something along those lines. At least in my sense of the word, the idea of an "estimate" is that it's something that you know the answer to, and you might know very accurately, but you don't know very precisely.
Like if you asked me how many computers are in the room I'm in now, without counting, I'd estimate 12. Now it may only be 10 or it may be 14, but when I say 12, that's not simply a guess. I know every computer in the room and could shut my eyes and count in my head and probably come up with an exact number, and I know for absolute sure that it's no more than 14, and definitely at least 10.
So that's an estimate. Not that I particularly like the word "guesstimate".
What you're saying would be absolutely correct if economics were a zero-sum game, and all of us were making the same amount of money. But the reality is that the distribution of wealth matters, and what is done with that money matters.
Great, all the companies can move to poor countries where they can pay the workers next to nothing and pay very little in taxes. Then they can make all the money they want without paying anyone, selling their products to...
Minimizing revenue is a silly goal. It's trivially easy: stop collecting taxes. However, you're left with the choice, then, of unsupportable deficits or anarchy.
Oh, OK, go ahead, raise those taxes. Raise $200 billion more in taxes and pay $200 billion more in bailouts, is that how it's supposed to work?
No, you're right, we should have $200 billion is tax cuts to pay for $200 billion more in bailouts. That math is much better.
All taxes punish success. You can't get around that fact.
Well that's overstating it; it's not punishing success. It's not like getting a raise is going to result in less take-home pay.
I think the question of "what format is suitable" has been answered repeatedly: HTML is fine. PDF is OK. RTF is probably OK.
Each of those formats may do well enough, but none of them are perfect or without problems. Only PDF really allows for sufficient layout control (depending, of course, on your view of "sufficient"), and yet at the same time it doesn't really allow for the content to be maximized on the fly for different sized/shaped displays. Even HTML, which you might argue is the best option, isn't perfect. It's being revised for newer browsers because of various problems.
But regardless, I'm not making excuses. I'm not saying that they can't just pick one of these formats and get to work. I'm saying that if you want to have a single device that can read all the various newspapers with uniform functionality, they need to get together and decide how they're going to do it. It will work best if they standardize.
If it's all HTML, that's fine. They'll probably want to set guidelines for styling. Do you set font size by a set number of px or em? Do you want to set anything as fixed width, or should it all be liquid? What I'm trying to get at is that the web is great, but it's also a mess. If you want to productize newspapers as a subscription service, you probably want to standardize to make the experience as clean as possible.
Part of the problem with that is that people like to control layout, and some kinds of information almost demand a particular presentation. So eventually, you're going to want something more than RTF, probably including something where pictures can be dropped in easily.
That leaves HTML as an obvious option, but then you're still going to want to standardize on a particular version of HTML and CSS, if not a particular rendering engine with particular options enabled to allow consistent rendering. Do you want to make it simplified to avoid battery drain? No javascript support, maybe?
I'm really not trying to advocate a particular choice. I'm just trying to point out that these newspapers need to get into a room with tech people and hash out a standard that provides all the functionality they're going to want. Once they have that, they can essentially publish a spec for what constitutes a "supported reader".
Neither. You'd specify a generic webservice interface with which each reader can specify which features it supports, and the 'newspaper' server would send it a PDF that comes closest.
Sure, that'd be one way to attack the problem. Still, my larger point was that you'll want to standardize things across newspapers. Getting everyone to agree to an interface for providing the appropriate PDF from various versions (which in itself implies the newspapers will be generating various standard PDF versions according to some defined spec) would be the sort of standardization I'm talking about.
It's not that SMS is bad. SMS is basically fine, but it's limited and expensive. That's right, contrary to your point about them being cheaper (than full cellular data plans), SMS messages are expensive for consumers. Last I checked, it cost something like $0.10/message, or something like $20 for an unlimited texting plans. Unlimited data plans aren't necessarily much more than that. And for what? As you point out, they're already built-in with the phones and the cell networks and have very little overhead.
And what format do you suggest? TXT doesn't have formatting, RTF doesn't have very good formating. Even HTML is difficult to get complex formatting that will render reliably. PDF? Ok, that sounds pretty good, but what if PDF lacks some feature that some of the newspapers really want to be able to include?
My point is, it doesn't really need to be a pre-existing format, but whatever format they use, they should standardize so that ANY reader can use it, or at bare minimum, there is some reader out there that can read all the newspapers.
And even if they settled on PDF, for example, there might still be issues to hash out. For example, do all readers' screens use the same aspect ratio? The same resolution? The same number of colors? I bet you I could make a PDF that would look great on 1024x768 with 4 shades of gray but would be flat-out illegible at 800x600 monochrome. So that raises the question, do you try to force papers to shoot for the lowest common denominator, or do you let them establish a baseline standards that readers would have to meet in order to be supported?
Yeah, my point wasn't that resolution doesn't matter, but rather when you start talking about resolutions like "1024x1280", I think you might be taking the wrong approach. Instead of treating it like a computer screen and bumping up through the normal screen resolutions, it seems to me like the better thing to do is start from the question, "What are people going to be reading on these things, and given that, what is the optimal page size?" Once you have a page size, you figure out the DPI necessary for comfortable reading, and go from there.