While I've never heard the term "micro-expression" before, I'm very familiar with the concept, although from a venue very different from an airport: in poker, these are called "tells" and it is very well established that they really do work
First of all, you're jumping to conclusions. Microexpressions and "tells" are different.
More importantly, "working" in poker and "working" in security are very different. In poker, even something slightly better than chance is good. In security, you need to be pretty close to certain before you retain people.
You'd say IDSes, firewalls, security policies, password changing, WPA for wireless, etc., were all just security theater and would disable the lot.
No. I'm saying that security needs to be based on scientific and engineering principles: peer reviewed results, repeatable experiments, statistical validity, and cost/benefit analyses.
That's as opposed to the kind of irrational handwaving and knee-jerk reactions people like you engage in; thank you for giving us such a clear demonstration why security is so poor.
Sound, hard-to-forge identity cards are a good thing for privacy: they make it harder for other people to impersonate you.
What is bad is centralized databases.
It's odd that the two are so often linked, because a good system of identity cards actually greatly reduces the need for centralized databases: if important information about you is on your id card, then showing the id card is sufficient--there is no need to go to a centralized database to verify that information or retrieve additional information. Furthermore, with an id card, you control who sees the information, whereas with a centralized database, you don't.
What's the typical way to negotiate this term out of the contract?
Simple: you buy a business line; it's only a question of paying for it.
Alternatively, you can co-locate and connect to your server from home. That's outgoing connections only, so there is no problem with terms of service, provided you stay under the implicit volume limits.
I'm sorry you are having trouble understanding that term, but they get quite a bit more explicit: no "servers", no "public services", no "file sharing", no "e-mail servers", no "web hosting", no "proxy services". BitTorrent is clearly prohibited. Furthermore, they can turn off anything else they don't like; it says so in the TOS. That's what you're paying for. If you don't like it, get a different subscription.
Comcast is purposely keeping things vague because that allows them to badger their customers into submission more easily.
They are keeping things vague because it's a low-cost consumer service with no guarantees whatsoever.
If you don't like it, get a different subscription; Comcast offers subscriptions that let you run servers.
Switching to a metered system would allow them to have a precise notion of what overusing the service means.
If you don't like the "imprecise notion" they are giving you, get a subscription that gives you the guarantees you want and pay for it.
Now, I agree that volume-priced models would be nice even for consumer cable (I've said so before). Of course, I think P2P and business models based on it would come to a crashing halt, but that's just fine by me.
I reject your notion of normal. We have a media room in our house, projector screen and home theater, etc.
OK, so you're a rich guy with tens of thousands of dollars of home entertainment equipment, and you're subscribing to an IPTV service that makes you abuse your network connection instead of them paying for a content distribution network. I just use cable for reading news and occasionally connecting to Subversion. Why should I subsidize your bandwidth usage?
It's a normal usage
Statistically, it's not "normal" usage. Contractually, it's not permitted usage. And economically, it just wouldn't work out.
I pay for unlimited data, and if they were to start restricting *my* usage by packet type to save themselves money I would be extremely pissed.
No, you do not. If you're a residential Comcast customer, you pay for unlimited data subject to the TOS. Those TOS specifically prohibit file sharing. If you want to be able to do file sharing, ISPs offer other plans for that as well--more expensive ones.
Alternatively, you can co-locate, which is what I do for my servers.
I noticed you left out, "getting their own dedicated line". Why?
Because there is no such thing as a "dedicated line": sooner or later, the wires come together. What is being sold as a "dedicated line" is a shared line with specific bandwidth guarantees, guarantees that are useless and unnecessarily expensive for almost every user.
Volume-based pricing is the right way to go, with appropriate peak and off-peak rates to shift the traffic around. Among other things, P2P traffic could probably be accommodated well at very low prices during off-peak hours, and shifting P2P traffic to off-peak hours could even be automated.
Until they define what they mean, all arguments that rely on the notion of "server" are specious.... and, incidentally, they do exactly that:
run programs, equipment, or servers from the Premises that provide network content or any other services to anyone outside of your Premises LAN (Local Area Network), also commonly referred to as public services or servers. Examples of prohibited services and servers include, but are not limited to, e-mail, Web hosting, file sharing, and proxy services and servers;
No, it depends on how they define "server"; it says so in the TOS.
Until they define what they mean, all arguments that rely on the notion of "server" are specious.
Actually, in the TOS, it pretty much says that they can restrict your traffic in any way they like for any reason they like. So, what is "specious" is your kinds of attempts to argue that they are obligated to let your run P2P nets.
And for those of us who actually like to use our networks to do web browsing and the occasionally download something, P2P usage by others is a big nuisance.
In any case, I think the solution is simply to go to volume based pricing. And, in fact, I think it would be good if ISPs were regulated and forced to offer transparent, volume-based pricing. That way, you can run all the P2P software you like and pay for the bandwidth, rather than have low-volume users like me subsidize you.
run programs, equipment, or servers from the Premises that provide network content or any other services to anyone outside of your Premises LAN (Local Area Network), also commonly referred to as public services or servers. Examples of prohibited services and servers include, but are not limited to, e-mail, Web hosting, file sharing, and proxy services and servers;
Also:
restrict, inhibit, interfere with, or otherwise disrupt or cause a performance degradation, regardless of intent, purpose or knowledge, to the Service or any Comcast (or Comcast supplier) host, server, backbone network, node or service, or otherwise cause a performance degradation to any Comcast (or Comcast supplier) facilities used to deliver the Service;
Furthermore, they are free to define any other activities as "prohibited" any time they like.
It means that Comcast isn't obligated under their contract with you to let you do any of that.
Now, you might want to argue that they ought to be obligated, that they are abusing their monopoly by writing such terms into your contract, etc. You might even be right. But those are different arguments and they need to be addressed at different people.
The kind of regulation I would favor is requiring ISPs to do simple volume-based pricing above a connection fee, and to make pricing transparent.
If you'd read your history, you'd know that all of this is, in fact, quite analogous to railroads about a century ago.
If you look at your Comcast terms of service, you will find that you're not allowed to run any servers at all. So, to all those people complaining about Comcast not honoring their "unlimited" plans, you're not honoring your part of the contract either. Comcast will tolerate a limited amount of "serving" (webcams, remote access), but torrents go far beyond that.
Now, this situation sucks. I don't use a lot of volume, but I want fast speed when I need it and I'm basically subsidizing people who send around porn and warez. Furthermore, abuses by high volume server and p2p operators on the network not only limit the performance I paid for, they also may force Comcast to institute filters that also interfere with my usage.
I think the solution is to go to volume pricing and drop unlimited plans altogether. That way, if you want to run p2p or some server over cable, you pay for what you use. Or you can do what what I do for my high-volume server needs: colocate.
Simple economics tells you that the monthly payment you make must more than cover the average monthly bandwidth costs, with a tidy profit. So, normal iTunes and YouTube usage are covered. And, in fact, neither of those comes close to what a serious Torrent user will consume.
Come on, haven't you figured out that "unlimited plan" doesn't mean "unlimited"? Just like "satisfaction guaranteed" doesn't actually mean that. It's just one of those phrases.
In fact, the best thing to do would be for ISPs to scrap unlimited plans and return to volume pricing, preferably peak/off-peak. That way, people pay for what they actually use and we can eliminate the guesswork.
Indeed: volume in practice isn't free or unlimited; it's a scarce resource that different users compete for. In fact, a well-priced volume plan, perhaps with peak and off-peak rates and the ability to pre-purchase gigabytes, would remove a lot of the uncertainty and unpredictability surrounding hosting.
Unfortunately, the market isn't really competitive, and the few providers that have wires into your home can afford to cater to just the average consumer and not worry about offering something for geeks and high bandwidth users.
The editorialist calls that a pseudo-science, but in fact it's a well-understood skill that can be taught and learned.
In fact the "editorialist" is right: for the time being, "microexpressions" are a pseudo-science. That is, the claims for their predictive ability are not sufficiently well-founded, there haven't been large enough studies, and the experiments have not been replicated enough, to justify widespread deployment.
Techniques used by law enforcement should really be studied with the same experimental and statistical rigor as drugs. People get upset, and there are big lawsuits, when a drug causes causes a few deaths per million, but the rate at which some bad law enforcement techniques disable people, lead to injury or death is several orders of magnitude worse.
If you discount a hypothesis because you haven't figured out how to test it yet, then you are the one being closed minded.
A statement only becomes a scientific hypothesis once there is a way to test it experimentally. Prior to that, it's speculation, fiction, a tautology, or often simply meaningless.
Secondly, science needs to be large enough to handle any concept that might be true. This includes things like living in "The Matrix", or being in a glass ball (called a universe) on someone's (God's) desk. Science is strong enough to show these things.
No, it is not. Those statements are just meaningless. For examples, at the scales of the universe, concepts like "glass ball" or "desk" don't make sense.
There is no reason to believe that all memories work the same way. Taste and smell aversion, in particular, are so different from other kinds of memories that there's a good chance that they work differently too.
Well, according to the summary, its not just the ads that are the issue, its the behavior of the users, too.
Except he has no convincing evidence for that assertion. All the studies I have seen suggest that Firefox users are more affluent and more educated. Of course, given the widespread use of Firefox in Europe, US sites may see disproportionately fewer purchases from Firefox users.
Anyway, few businesses can afford to lose a few percent of their potential customers.
Yes, but let's be clear why. Solaris used to be full of bugs, memory leaks, crashes, and data corruption. Several of the systems Sun designed (e.g., NFS, NIS) had serious design, performance, and security problems, in addition to data corruption bugs. So, the Solaris kernel is "good and stable" because it's old and because Sun let their users do the debugging (the user land still sucks, of course). So, Solaris may solve your problem, but I wouldn't trust any new code or designs coming out of Sun until it's been in real-world use and debugged for several years.
I can't see how it's any less open than (say) Linux.
Solaris is proprietary because Sun owns the copyright and retains special rights to the code. The fact that they have released it under a dual license doesn't change that. The same is true for Java. And the distinction is not academic: Sun gets to decide what happens with Solaris and Java in the future, not the end users.
Linux, in contrast, is not proprietary: everybody has the same rights to Linux as everybody else, and, for practical purposes, everybody owns it.
Slashdot Mirror
While I've never heard the term "micro-expression" before, I'm very familiar with the concept, although from a venue very different from an airport: in poker, these are called "tells" and it is very well established that they really do work
First of all, you're jumping to conclusions. Microexpressions and "tells" are different.
More importantly, "working" in poker and "working" in security are very different. In poker, even something slightly better than chance is good. In security, you need to be pretty close to certain before you retain people.
You'd say IDSes, firewalls, security policies, password changing, WPA for wireless, etc., were all just security theater and would disable the lot.
No. I'm saying that security needs to be based on scientific and engineering principles: peer reviewed results, repeatable experiments, statistical validity, and cost/benefit analyses.
That's as opposed to the kind of irrational handwaving and knee-jerk reactions people like you engage in; thank you for giving us such a clear demonstration why security is so poor.
Sound, hard-to-forge identity cards are a good thing for privacy: they make it harder for other people to impersonate you.
What is bad is centralized databases.
It's odd that the two are so often linked, because a good system of identity cards actually greatly reduces the need for centralized databases: if important information about you is on your id card, then showing the id card is sufficient--there is no need to go to a centralized database to verify that information or retrieve additional information. Furthermore, with an id card, you control who sees the information, whereas with a centralized database, you don't.
What's the typical way to negotiate this term out of the contract?
Simple: you buy a business line; it's only a question of paying for it.
Alternatively, you can co-locate and connect to your server from home. That's outgoing connections only, so there is no problem with terms of service, provided you stay under the implicit volume limits.
Again an undefined term: "network content".
I'm sorry you are having trouble understanding that term, but they get quite a bit more explicit: no "servers", no "public services", no "file sharing", no "e-mail servers", no "web hosting", no "proxy services". BitTorrent is clearly prohibited. Furthermore, they can turn off anything else they don't like; it says so in the TOS. That's what you're paying for. If you don't like it, get a different subscription.
Comcast is purposely keeping things vague because that allows them to badger their customers into submission more easily.
They are keeping things vague because it's a low-cost consumer service with no guarantees whatsoever.
If you don't like it, get a different subscription; Comcast offers subscriptions that let you run servers.
Switching to a metered system would allow them to have a precise notion of what overusing the service means.
If you don't like the "imprecise notion" they are giving you, get a subscription that gives you the guarantees you want and pay for it.
Now, I agree that volume-priced models would be nice even for consumer cable (I've said so before). Of course, I think P2P and business models based on it would come to a crashing halt, but that's just fine by me.
I reject your notion of normal. We have a media room in our house, projector screen and home theater, etc.
OK, so you're a rich guy with tens of thousands of dollars of home entertainment equipment, and you're subscribing to an IPTV service that makes you abuse your network connection instead of them paying for a content distribution network. I just use cable for reading news and occasionally connecting to Subversion. Why should I subsidize your bandwidth usage?
It's a normal usage
Statistically, it's not "normal" usage. Contractually, it's not permitted usage. And economically, it just wouldn't work out.
I pay for unlimited data, and if they were to start restricting *my* usage by packet type to save themselves money I would be extremely pissed.
No, you do not. If you're a residential Comcast customer, you pay for unlimited data subject to the TOS. Those TOS specifically prohibit file sharing. If you want to be able to do file sharing, ISPs offer other plans for that as well--more expensive ones.
Alternatively, you can co-locate, which is what I do for my servers.
I noticed you left out, "getting their own dedicated line". Why?
Because there is no such thing as a "dedicated line": sooner or later, the wires come together. What is being sold as a "dedicated line" is a shared line with specific bandwidth guarantees, guarantees that are useless and unnecessarily expensive for almost every user.
Volume-based pricing is the right way to go, with appropriate peak and off-peak rates to shift the traffic around. Among other things, P2P traffic could probably be accommodated well at very low prices during off-peak hours, and shifting P2P traffic to off-peak hours could even be automated.
Clear enough for you?
http://www.comcast.net/terms/use.jsp
Your argument depends on how you define "server".
No, it depends on how they define "server"; it says so in the TOS.
Until they define what they mean, all arguments that rely on the notion of "server" are specious.
Actually, in the TOS, it pretty much says that they can restrict your traffic in any way they like for any reason they like. So, what is "specious" is your kinds of attempts to argue that they are obligated to let your run P2P nets.
And for those of us who actually like to use our networks to do web browsing and the occasionally download something, P2P usage by others is a big nuisance.
In any case, I think the solution is simply to go to volume based pricing. And, in fact, I think it would be good if ISPs were regulated and forced to offer transparent, volume-based pricing. That way, you can run all the P2P software you like and pay for the bandwidth, rather than have low-volume users like me subsidize you.
Under prohibited activities, you find:
Also:
Furthermore, they are free to define any other activities as "prohibited" any time they like.
It means that Comcast isn't obligated under their contract with you to let you do any of that.
Now, you might want to argue that they ought to be obligated, that they are abusing their monopoly by writing such terms into your contract, etc. You might even be right. But those are different arguments and they need to be addressed at different people.
The kind of regulation I would favor is requiring ISPs to do simple volume-based pricing above a connection fee, and to make pricing transparent.
If you'd read your history, you'd know that all of this is, in fact, quite analogous to railroads about a century ago.
If you look at your Comcast terms of service, you will find that you're not allowed to run any servers at all. So, to all those people complaining about Comcast not honoring their "unlimited" plans, you're not honoring your part of the contract either. Comcast will tolerate a limited amount of "serving" (webcams, remote access), but torrents go far beyond that.
Now, this situation sucks. I don't use a lot of volume, but I want fast speed when I need it and I'm basically subsidizing people who send around porn and warez. Furthermore, abuses by high volume server and p2p operators on the network not only limit the performance I paid for, they also may force Comcast to institute filters that also interfere with my usage.
I think the solution is to go to volume pricing and drop unlimited plans altogether. That way, if you want to run p2p or some server over cable, you pay for what you use. Or you can do what what I do for my high-volume server needs: colocate.
Simple economics tells you that the monthly payment you make must more than cover the average monthly bandwidth costs, with a tidy profit. So, normal iTunes and YouTube usage are covered. And, in fact, neither of those comes close to what a serious Torrent user will consume.
Come on, haven't you figured out that "unlimited plan" doesn't mean "unlimited"? Just like "satisfaction guaranteed" doesn't actually mean that. It's just one of those phrases.
In fact, the best thing to do would be for ISPs to scrap unlimited plans and return to volume pricing, preferably peak/off-peak. That way, people pay for what they actually use and we can eliminate the guesswork.
Indeed: volume in practice isn't free or unlimited; it's a scarce resource that different users compete for. In fact, a well-priced volume plan, perhaps with peak and off-peak rates and the ability to pre-purchase gigabytes, would remove a lot of the uncertainty and unpredictability surrounding hosting.
Unfortunately, the market isn't really competitive, and the few providers that have wires into your home can afford to cater to just the average consumer and not worry about offering something for geeks and high bandwidth users.
The editorialist calls that a pseudo-science, but in fact it's a well-understood skill that can be taught and learned.
In fact the "editorialist" is right: for the time being, "microexpressions" are a pseudo-science. That is, the claims for their predictive ability are not sufficiently well-founded, there haven't been large enough studies, and the experiments have not been replicated enough, to justify widespread deployment.
Techniques used by law enforcement should really be studied with the same experimental and statistical rigor as drugs. People get upset, and there are big lawsuits, when a drug causes causes a few deaths per million, but the rate at which some bad law enforcement techniques disable people, lead to injury or death is several orders of magnitude worse.
The iPhone interface is very nice. But the best interface doesn't help me if the thing just doesn't do what I need my phone to do, and it doesn't.
If you discount a hypothesis because you haven't figured out how to test it yet, then you are the one being closed minded.
A statement only becomes a scientific hypothesis once there is a way to test it experimentally. Prior to that, it's speculation, fiction, a tautology, or often simply meaningless.
Secondly, science needs to be large enough to handle any concept that might be true. This includes things like living in "The Matrix", or being in a glass ball (called a universe) on someone's (God's) desk. Science is strong enough to show these things.
No, it is not. Those statements are just meaningless. For examples, at the scales of the universe, concepts like "glass ball" or "desk" don't make sense.
It depends on how the page is written.
If you want to know more about what the page is hanging on, get YSlow.
I'm assuming that most sites actually want to serve the ads first.
There is no reason to believe that all memories work the same way. Taste and smell aversion, in particular, are so different from other kinds of memories that there's a good chance that they work differently too.
Tell us: you get paid for spreading vitriol about global warming and the people working in the area? Or do you simply not know any better?
Well, according to the summary, its not just the ads that are the issue, its the behavior of the users, too.
Except he has no convincing evidence for that assertion. All the studies I have seen suggest that Firefox users are more affluent and more educated. Of course, given the widespread use of Firefox in Europe, US sites may see disproportionately fewer purchases from Firefox users.
Anyway, few businesses can afford to lose a few percent of their potential customers.
* Solaris is good and stable.
Yes, but let's be clear why. Solaris used to be full of bugs, memory leaks, crashes, and data corruption. Several of the systems Sun designed (e.g., NFS, NIS) had serious design, performance, and security problems, in addition to data corruption bugs. So, the Solaris kernel is "good and stable" because it's old and because Sun let their users do the debugging (the user land still sucks, of course). So, Solaris may solve your problem, but I wouldn't trust any new code or designs coming out of Sun until it's been in real-world use and debugged for several years.
I can't see how it's any less open than (say) Linux.
Solaris is proprietary because Sun owns the copyright and retains special rights to the code. The fact that they have released it under a dual license doesn't change that. The same is true for Java. And the distinction is not academic: Sun gets to decide what happens with Solaris and Java in the future, not the end users.
Linux, in contrast, is not proprietary: everybody has the same rights to Linux as everybody else, and, for practical purposes, everybody owns it.
The author of those pages is a right wing religious nut; have a look at his diatribe against atheists:
e ism-and-Morality
http://www.udolpho.com/weblog/?id=01170&title=Ath
Well, then block people who block ads. Or integrate your ads into your site.
Blocking a specific browser is stupid.