This fine is by the FCC. It is their job to regulate the use of the spectrum, and thus they only have jurisdiction over those who license the spectrum. The major labels have been convicted of payola and fined for it. The charger were brought forth by Eliot Spitzer on behalf of the state of NY - see here and here
If the police see a drug deal, both the buyer and the seller will be arrested.
But they both get their own day in court. This is the same thing.
I never bought the lowest cost components, and still had problems. I used to be able to trust ASUS to always have good boards, and then they started putting out some flacky ones in pursuit of looking good on those silly motherboard benchmark tests. IBM used make great drives, then the deskstar debacle happened (okay that example is fairly old). Seagate is still pretty consistant, but they have had a small number of bad batches. The case fans that I used to get (forget the name of the company, but they were nice and quiet) - never had problems with them, then I had several seize up within a few months of buying them. I can't think of a single reliable manufaturer that hasn't put out at least one crummy product recently. Name alone isn't enough anymore.
Prior to 2002, Diebold did not produce any electronic voting machines. In that year they acquired Global Election Systems, which became their voting machine division. GES had always produced garbage, and it is no suprise that they continued to do so under new ownership. The incompetance of that division shouldn'd reflect onto the abilities of the engineers in other divisions, although it does certainly say something about the management of the company as a whole.
As a side note, the reason that Diebold acquired GES in 2002 was because of the expected boom in voting machines sales as a result of the 2001 Help America Vote Act. This act, which was a response to the hanging-chad problems in the 2000 election set forth requirements concering voters with disabilities, which drove many states to buy electronic voting machines as they seemed the easiest way to statisfy the requirements. So next time someone tells you that the federal government cannot pass any laws about electronic voting, remind them that it was federal law that led the situation to begin with.
For me what has changed the most is that the reliability of hardware is much harder to predict. 10 years ago there were good brands and as long as you stuck with them you would have a reliable machine. Now-a-days it isn't anywhere near as clear-cut. Every brand is trying to aim for the low prices and thus has it's occasional stinkers. And the hardware review sites are no help because they only test for performance not reliability. Pricing has also more time consuming as half the online retailers advertise artificially low prices which they make up for with high shipping costs. And if you are running linux, the amount of work necissary to check if the hardware has drivers has increased as well.
So while slapping the components together has gotten easier, shopping for them has gotten harder. Even 10 years ago, shopping for parts was always more time consuming then actually building and configuring the machine, now it is even worse. Not wanting to deal with all that and not wanting to buy a crappy HP/Dell was a big factor in convincing me to get a G5 for my last computer rather than a PC. Probably as much of a factor as the OS.
This may seem tangental to the original discussion, but I don't think it is. Like building computers, it's my opinion thatthe limiting factor in moving to linux isn't lack of technical skill, but being willing to spend the time lookup up how to do things that are different from windows, and scouring mailing lists and forums to find an answer when something goes wrong. So if someone is willing and able to spend the time to learn about everything you need to know to pick out PC components, and deal with windows driver issues on bleeding edge hardware, then they have the skills needed to use linux. The question is one of motivation - why would they do so if linux doesn't do what they want, ie play games.
Before someone catches me on this, MEMs devices like this measure acceleration, not velocity like my previous post implied. Same priciple though, just have to integrate twice. The 6-axis part is important as measuring your rotation in addtion to linear acceleration enables you to keep track of what direction you are pointed.
It is integrating your velocity to estimate where you are between GPS solutions. Navigation and guidance systems for high velocity (read military) devices do this already out of necessity. However, it would also be useful for low velocity situations where you have a spotty GPS signal. In that situation it doesn't have to be perfect to be usefull, especially if the display indicated the approximate error in the estimation by drawing a circle for your position rather than a dot.
Slashdot does have story moderation system now. It is called firehose - you can find a link in the menu at the top of the screen. It allows you to give thumbs up or thumbs down to a story as well as marking a story with feedback such as dupe or typo, in addition to the normal tagging system.
I both gave this story a thumbs down and dupe feedback, however, so many other people moderated the story up that it was at the highest (visible) ranking by the time it got posted. Apparently a bunch of people missed the story the first time around, or didn't realize this was the same study or something. I guess I can't really blame the editors for giving users what they want.
Or am I just missing something? I would hope NASA scientists are far smarter than myself....
They are.
These outposts, as well as Bigalow's hotels, have multiple layers, one of which is essentially kevlar, the same stuff that bullet proof vests are made out of. They actually provide much better protection from micro-meteorites and space junk then our current metal structures do.
In addition, I am very sceptical of the claim that countries are modeling their laws after ours because it is a proven system. I'd wager that it has more to do with being required to have IP laws which are simular to the US and Old Europe in order to participate in various trade groups and treaties.
This is on my G5 (single-1.8Ghz) running 10.3, so it isn't an intel thing. OS X in general seems to put give more control over a window to the application than both Windows and X11. In OS X when an application has the beachball of death going, you often can't minimize / move the windows, while you can in X11. Not that they don't both have problems of their own. On my windows computer here at work, Adobe Acrobat has been locked for days. I can't kill it with the task manager, or taskkill (even when run as LOCALSYSTEM) and don't feel like shutting down all my applications to reboot. So it is just sitting there minding its undead self till I eventually need to read a PDF:)
Finder locks up, because it's a giant ball of shit, but other OS X applications shouldn't lock up enough to force a reboot.
I have had it happen to me a couple times (safari was the offender). In addition, there are many many instances where an application won't lock up the machine per say, but it will lock up the menubar, which unless you know to press cmd-opt-esc to kill the application, has pretty much the same effect.
The open-source programs that use MP3 have always been on uncertain legal ground in many countries including the US. Many of them have received C&D letters from Fraunhofer - the other patent holder, and recognized licenser of MP3 related patents. The way that LAME gets around this is by choosing a license that only allows use for non-commercial research purposes (allowed by US patent laws). Others get around it by basing their operations in countries where the laws are different. It has been this way for years.
That is why Ogg vorbis development was started over decade ago. That is why the codecs for MP3 and other proprietary formats are no longer included in the base install of many distros (including ubuntu last I checked). It is up to the user to install the packages from the non-free repository, and they are responsible for determining if those packages are legal in their country.
In the end the liability is most likely on the user (the Microsoft vs ATT case will make that more clear), not the developer or distributer. The distros and projects may be hit with C&D letters, but there is no reason to think that they will stick anymore than than the ones in the past.
This ruling primarily affect companies that had been paying patent royalties to Fraunhofer, and thought they were in the clear. Groups that have never payed royalties are in the exact same position that they have always been always in.
And most people were used to spending $20 for a walkman when the iPod came out. That's why most people didn't start buying MP3 players until a year or so ago when the price started getting reasonable.
The only people who will buy this are early adopters, just like any other emerging market. However, Apple is genious at is redefining who the early adopter is. MP3 players had been around for a while before the iPod but the only people who bought them were geeks that loved music and had lots of disposible income. When the iPod came out that changed to anybody who loved music and had lots of disposible income. The smart phone is in the exact position that MP3 players were 8 years ago, and I fully expect the iPhone to be just as successfull as the iPod.
Furthermore, as far as price goes, we need to wait and see what Rates Cingular charges. Speculations has been all across the board here, but if the reports are true that the price of the phone is not being subsidised by the cellular plan, then Cingular could choose to offer a discounted data package in order to try and attract more people. Or they could keep charging rediculous amounts for data.
Newspaper headlines are horrible. Between the fact that english has far too many words that could be both nouns or verbs depending on context, that proper often nouns cannot be discerned from normal words when everything is capitalized, and journalists being way too clever for their own good leads to monstrosities of randomly juxtaposed words that cannot be parsed until you have read at least the first couple paragraphs of the article.
Re:"source would have to be made available" ?
on
iPhone Not Running OS X
·
· Score: 4, Insightful
No the parent is right. There is very little GPL or otherwise copyleft code shipped with OS X, and what is there is all userland stuff that really doesn't need to be on a phone. The vast majority of the stuff that Apple/NeXT didn't write is licensed under BSD-like terms, and therefore allows them to do whatever they want with it.
I read that as saying that he could tell which of his books she had been handling by recognizing their scent on her hands. This didn't phase me a bit, although I was a bit surprised that bottles would have a unique enough smell. It wasn't until I read another poster's description that I realized he was recognizing the scent of his wife on the books.
In general, the client isn't as important as the provider when it comes to interoperability. The problem is one of initiating the connection. Like IM, or email or a regular phone call, you need some way of locating the person you want to talk to on the internet. If the person you are calling has a static IP address, and has their software / firewall setup to allow direct connections then you can call them directly using their IP address. But the vast majority of people who use the internet do not have a static IP address and worse are behind a NAT, or would like their SIP number to move with them to multiple locations, and hence IPs. In these case, you have to log into a provider who then associates your account name with your current dynamic IP. Then when someone calls you they send your account name to the provider who then either provides them with your current IP, or in the case where you are both behind a firewall, acts as a mediator to initiate the connection.
Some providers allow connections from outside networks, some don't. As far as I can tell: iChat only allows connections to/from AIM or.Mac accounts. Wengo (the provider) only allows connections from other Wengo accounts. Sametime will depend on how your IT folks set things up. Gizmo appears to allow connections from any SIP number. Most telcom providers allow connections from any SIP number.
The address that you use to connect to people can vary. Sometimes you can just use SIP:username@provider.org, but sometimes you have to use SIP:sipaddress@provider.org, where sipaddress is a long number that you'll have to ask your provider for.
The software packages can sometimes be used with any providers. This is definitely true for the WengoPhone, which is libreware. Other features of the various software packages are often specific to that provider such as account presence (shows you if people in your buddy list are online).
WengoPhone is an SIP based application that can be used with whatever SIP provider you want, as well as being able to interoperate with AIM, MSN, Yahoo, and Jabber (including Google Chat). Who you can talk to then depends on which SIP networks your SIP provider allows calls to. If you use Wengo as your provider, they only allow calls to other Wengo users. It is one of the more frustrating things about SIP - from a technical point-of-view there is little preventing networks from interoperating, but so many are walled off to varying degrees for business reasons.
As for Wengo Visio, it's sole purpose it to let people without VOIP software/accounts to talk to you, so interoperability only affects what software you have to run. From the sparse amount of information on the website, it appears to require a wengo account and WengoPhone software, but it is SIP based and therefore might be able to be setup to connect to any SIP account / software.
My (somewhat limited) experience with.NET was that it's performance was somewhat uneven. Most of the time it would run about about on par with java, but every now and then I would run into something that was dog slow, as in 10-100 times slower. All those problem areas seemed to be in the libraries however, not the VM. This is understandable as java has had much more time to be polished.
In addition to the fact that it is fraud, it might also not be a good investment. There is not a big market for used rings, at least wedding / engagement rings, simply because of the social stigma of giving your fiancé used goods. So a single rock on a plain ring would not be a good way to go. When it comes to purely decorative jewelry, the value of that is largely determined by the fame of the jeweler who designed the piece, in addition to how much the stones (you likely will need several) are worth. I don't know if those factors would decrease the price as much as the fact that it is an artificial diamond, but they would need to be considered.
Obviously he is using IP4.1 which extends each octet to include the Evil Bit, which can be used to mask off packets intended for the spirit metanet. The occult have long hoped that this will lead to breakthroughs in high speed multi-realm data transfer, which has yet to be realised. However, if you matrix the multinet streams, you can get some awesome spooky surround sound.
It could be that Microsoft intentionally limits the message rate to help decrease flooding. Therefore the problem is with GAIM because it doesn't follow the flooding policy designed into the (undocumented) MSN messenger protocol. Regardless of why it doesn't work, it would be nice if GAIM fixed it.
IBM's patent portfolio is truly frightening in that the only thing preventing it from doing massive harm to the industry is self restraint and the enlightened self interest of wanting to remain relevant in the industry. Let's just hope their business never goes south. If you thought that the IP trolls that make money by buying the patents portfolios of failed start-ups was bad, just imagine the hell that will be unleashed if IBM enters a downward spiral and decides to "refocus the company revenue strategies on their intellectual property licensing opportunities".
The reason that we have the credit card fraud protection that we do today is not just because the banks thought it was a good idea, but because federal law makes them liable for all fraudulent charges up to a certain amount. Regardless of what arguments they put forth about who is most at fault, it is the bank and not the consumer who is liable, period. The credit card companies can and do write conditions into it's merchant contract that say they won't pay the merchant for fraudulent charges, especially if they don't have a signed receipt. In this case PIN numbers actually push liability further away from the merchant and back to the bank. When the actual perpetrator of fraud is caught, they are required to pay the money to the bank. But if these don't occur it is the bank who must ultimately swallow the losses.
This is why the banks have invested a ton of money into programs that detect patterns of fraud, and why I am not too worried about these new technologies. If they increase fraud, then the bank will be the one that gets hurt and if it becomes too great a problem they will move back to the old solution or onto better solutions (smart card authentication using PINs and public/private keys). So just stick with regular credit card accounts, not debit cards - the legal protection is the same, but you are out the money until the bank gets around to refunding it - and you will be fine. Of course, this is US specific, YMMV in the EU.
Slashdot Mirror
Wait till he finds out that Jedis don't have corpses.
I never bought the lowest cost components, and still had problems. I used to be able to trust ASUS to always have good boards, and then they started putting out some flacky ones in pursuit of looking good on those silly motherboard benchmark tests. IBM used make great drives, then the deskstar debacle happened (okay that example is fairly old). Seagate is still pretty consistant, but they have had a small number of bad batches. The case fans that I used to get (forget the name of the company, but they were nice and quiet) - never had problems with them, then I had several seize up within a few months of buying them. I can't think of a single reliable manufaturer that hasn't put out at least one crummy product recently. Name alone isn't enough anymore.
Prior to 2002, Diebold did not produce any electronic voting machines. In that year they acquired Global Election Systems, which became their voting machine division. GES had always produced garbage, and it is no suprise that they continued to do so under new ownership. The incompetance of that division shouldn'd reflect onto the abilities of the engineers in other divisions, although it does certainly say something about the management of the company as a whole.
As a side note, the reason that Diebold acquired GES in 2002 was because of the expected boom in voting machines sales as a result of the 2001 Help America Vote Act. This act, which was a response to the hanging-chad problems in the 2000 election set forth requirements concering voters with disabilities, which drove many states to buy electronic voting machines as they seemed the easiest way to statisfy the requirements. So next time someone tells you that the federal government cannot pass any laws about electronic voting, remind them that it was federal law that led the situation to begin with.
For me what has changed the most is that the reliability of hardware is much harder to predict. 10 years ago there were good brands and as long as you stuck with them you would have a reliable machine. Now-a-days it isn't anywhere near as clear-cut. Every brand is trying to aim for the low prices and thus has it's occasional stinkers. And the hardware review sites are no help because they only test for performance not reliability. Pricing has also more time consuming as half the online retailers advertise artificially low prices which they make up for with high shipping costs. And if you are running linux, the amount of work necissary to check if the hardware has drivers has increased as well.
So while slapping the components together has gotten easier, shopping for them has gotten harder. Even 10 years ago, shopping for parts was always more time consuming then actually building and configuring the machine, now it is even worse. Not wanting to deal with all that and not wanting to buy a crappy HP/Dell was a big factor in convincing me to get a G5 for my last computer rather than a PC. Probably as much of a factor as the OS.
This may seem tangental to the original discussion, but I don't think it is. Like building computers, it's my opinion thatthe limiting factor in moving to linux isn't lack of technical skill, but being willing to spend the time lookup up how to do things that are different from windows, and scouring mailing lists and forums to find an answer when something goes wrong. So if someone is willing and able to spend the time to learn about everything you need to know to pick out PC components, and deal with windows driver issues on bleeding edge hardware, then they have the skills needed to use linux. The question is one of motivation - why would they do so if linux doesn't do what they want, ie play games.
Before someone catches me on this, MEMs devices like this measure acceleration, not velocity like my previous post implied. Same priciple though, just have to integrate twice. The 6-axis part is important as measuring your rotation in addtion to linear acceleration enables you to keep track of what direction you are pointed.
It is integrating your velocity to estimate where you are between GPS solutions. Navigation and guidance systems for high velocity (read military) devices do this already out of necessity. However, it would also be useful for low velocity situations where you have a spotty GPS signal. In that situation it doesn't have to be perfect to be usefull, especially if the display indicated the approximate error in the estimation by drawing a circle for your position rather than a dot.
Slashdot does have story moderation system now. It is called firehose - you can find a link in the menu at the top of the screen. It allows you to give thumbs up or thumbs down to a story as well as marking a story with feedback such as dupe or typo, in addition to the normal tagging system.
I both gave this story a thumbs down and dupe feedback, however, so many other people moderated the story up that it was at the highest (visible) ranking by the time it got posted. Apparently a bunch of people missed the story the first time around, or didn't realize this was the same study or something. I guess I can't really blame the editors for giving users what they want.
These outposts, as well as Bigalow's hotels, have multiple layers, one of which is essentially kevlar, the same stuff that bullet proof vests are made out of. They actually provide much better protection from micro-meteorites and space junk then our current metal structures do.
In addition, I am very sceptical of the claim that countries are modeling their laws after ours because it is a proven system. I'd wager that it has more to do with being required to have IP laws which are simular to the US and Old Europe in order to participate in various trade groups and treaties.
This is on my G5 (single-1.8Ghz) running 10.3, so it isn't an intel thing. OS X in general seems to put give more control over a window to the application than both Windows and X11. In OS X when an application has the beachball of death going, you often can't minimize / move the windows, while you can in X11. Not that they don't both have problems of their own. On my windows computer here at work, Adobe Acrobat has been locked for days. I can't kill it with the task manager, or taskkill (even when run as LOCALSYSTEM) and don't feel like shutting down all my applications to reboot. So it is just sitting there minding its undead self till I eventually need to read a PDF :)
The open-source programs that use MP3 have always been on uncertain legal ground in many countries including the US. Many of them have received C&D letters from Fraunhofer - the other patent holder, and recognized licenser of MP3 related patents. The way that LAME gets around this is by choosing a license that only allows use for non-commercial research purposes (allowed by US patent laws). Others get around it by basing their operations in countries where the laws are different. It has been this way for years.
That is why Ogg vorbis development was started over decade ago. That is why the codecs for MP3 and other proprietary formats are no longer included in the base install of many distros (including ubuntu last I checked). It is up to the user to install the packages from the non-free repository, and they are responsible for determining if those packages are legal in their country.
In the end the liability is most likely on the user (the Microsoft vs ATT case will make that more clear), not the developer or distributer. The distros and projects may be hit with C&D letters, but there is no reason to think that they will stick anymore than than the ones in the past.
This ruling primarily affect companies that had been paying patent royalties to Fraunhofer, and thought they were in the clear. Groups that have never payed royalties are in the exact same position that they have always been always in.
And most people were used to spending $20 for a walkman when the iPod came out. That's why most people didn't start buying MP3 players until a year or so ago when the price started getting reasonable.
The only people who will buy this are early adopters, just like any other emerging market. However, Apple is genious at is redefining who the early adopter is. MP3 players had been around for a while before the iPod but the only people who bought them were geeks that loved music and had lots of disposible income. When the iPod came out that changed to anybody who loved music and had lots of disposible income. The smart phone is in the exact position that MP3 players were 8 years ago, and I fully expect the iPhone to be just as successfull as the iPod.
Furthermore, as far as price goes, we need to wait and see what Rates Cingular charges. Speculations has been all across the board here, but if the reports are true that the price of the phone is not being subsidised by the cellular plan, then Cingular could choose to offer a discounted data package in order to try and attract more people. Or they could keep charging rediculous amounts for data.
Newspaper headlines are horrible. Between the fact that english has far too many words that could be both nouns or verbs depending on context, that proper often nouns cannot be discerned from normal words when everything is capitalized, and journalists being way too clever for their own good leads to monstrosities of randomly juxtaposed words that cannot be parsed until you have read at least the first couple paragraphs of the article.
No the parent is right. There is very little GPL or otherwise copyleft code shipped with OS X, and what is there is all userland stuff that really doesn't need to be on a phone. The vast majority of the stuff that Apple/NeXT didn't write is licensed under BSD-like terms, and therefore allows them to do whatever they want with it.
I read that as saying that he could tell which of his books she had been handling by recognizing their scent on her hands. This didn't phase me a bit, although I was a bit surprised that bottles would have a unique enough smell. It wasn't until I read another poster's description that I realized he was recognizing the scent of his wife on the books.
In general, the client isn't as important as the provider when it comes to interoperability. The problem is one of initiating the connection. Like IM, or email or a regular phone call, you need some way of locating the person you want to talk to on the internet. If the person you are calling has a static IP address, and has their software / firewall setup to allow direct connections then you can call them directly using their IP address. But the vast majority of people who use the internet do not have a static IP address and worse are behind a NAT, or would like their SIP number to move with them to multiple locations, and hence IPs. In these case, you have to log into a provider who then associates your account name with your current dynamic IP. Then when someone calls you they send your account name to the provider who then either provides them with your current IP, or in the case where you are both behind a firewall, acts as a mediator to initiate the connection.
.Mac accounts.
Some providers allow connections from outside networks, some don't. As far as I can tell:
iChat only allows connections to/from AIM or
Wengo (the provider) only allows connections from other Wengo accounts.
Sametime will depend on how your IT folks set things up.
Gizmo appears to allow connections from any SIP number.
Most telcom providers allow connections from any SIP number.
The address that you use to connect to people can vary. Sometimes you can just use SIP:username@provider.org, but sometimes you have to use SIP:sipaddress@provider.org, where sipaddress is a long number that you'll have to ask your provider for.
The software packages can sometimes be used with any providers. This is definitely true for the WengoPhone, which is libreware.
Other features of the various software packages are often specific to that provider such as account presence (shows you if people in your buddy list are online).
The Wengo Visio product does include video chat.
WengoPhone is an SIP based application that can be used with whatever SIP provider you want, as well as being able to interoperate with AIM, MSN, Yahoo, and Jabber (including Google Chat). Who you can talk to then depends on which SIP networks your SIP provider allows calls to. If you use Wengo as your provider, they only allow calls to other Wengo users. It is one of the more frustrating things about SIP - from a technical point-of-view there is little preventing networks from interoperating, but so many are walled off to varying degrees for business reasons.
As for Wengo Visio, it's sole purpose it to let people without VOIP software/accounts to talk to you, so interoperability only affects what software you have to run. From the sparse amount of information on the website, it appears to require a wengo account and WengoPhone software, but it is SIP based and therefore might be able to be setup to connect to any SIP account / software.
My (somewhat limited) experience with .NET was that it's performance was somewhat uneven. Most of the time it would run about about on par with java, but every now and then I would run into something that was dog slow, as in 10-100 times slower. All those problem areas seemed to be in the libraries however, not the VM. This is understandable as java has had much more time to be polished.
In addition to the fact that it is fraud, it might also not be a good investment. There is not a big market for used rings, at least wedding / engagement rings, simply because of the social stigma of giving your fiancé used goods. So a single rock on a plain ring would not be a good way to go. When it comes to purely decorative jewelry, the value of that is largely determined by the fame of the jeweler who designed the piece, in addition to how much the stones (you likely will need several) are worth. I don't know if those factors would decrease the price as much as the fact that it is an artificial diamond, but they would need to be considered.
Obviously he is using IP4.1 which extends each octet to include the Evil Bit, which can be used to mask off packets intended for the spirit metanet. The occult have long hoped that this will lead to breakthroughs in high speed multi-realm data transfer, which has yet to be realised. However, if you matrix the multinet streams, you can get some awesome spooky surround sound.
It could be that Microsoft intentionally limits the message rate to help decrease flooding. Therefore the problem is with GAIM because it doesn't follow the flooding policy designed into the (undocumented) MSN messenger protocol. Regardless of why it doesn't work, it would be nice if GAIM fixed it.
IBM's patent portfolio is truly frightening in that the only thing preventing it from doing massive harm to the industry is self restraint and the enlightened self interest of wanting to remain relevant in the industry. Let's just hope their business never goes south. If you thought that the IP trolls that make money by buying the patents portfolios of failed start-ups was bad, just imagine the hell that will be unleashed if IBM enters a downward spiral and decides to "refocus the company revenue strategies on their intellectual property licensing opportunities".
The reason that we have the credit card fraud protection that we do today is not just because the banks thought it was a good idea, but because federal law makes them liable for all fraudulent charges up to a certain amount. Regardless of what arguments they put forth about who is most at fault, it is the bank and not the consumer who is liable, period. The credit card companies can and do write conditions into it's merchant contract that say they won't pay the merchant for fraudulent charges, especially if they don't have a signed receipt. In this case PIN numbers actually push liability further away from the merchant and back to the bank. When the actual perpetrator of fraud is caught, they are required to pay the money to the bank. But if these don't occur it is the bank who must ultimately swallow the losses.
This is why the banks have invested a ton of money into programs that detect patterns of fraud, and why I am not too worried about these new technologies. If they increase fraud, then the bank will be the one that gets hurt and if it becomes too great a problem they will move back to the old solution or onto better solutions (smart card authentication using PINs and public/private keys). So just stick with regular credit card accounts, not debit cards - the legal protection is the same, but you are out the money until the bank gets around to refunding it - and you will be fine. Of course, this is US specific, YMMV in the EU.