A more interesting type of system would use a QR code challenge-response. A small screen on the gumball machine, or at the supermarket checkout flashes a QR code. You point your phone camera at it and details of the transaction come up on the screen. If you hit "confirm", your private key is used to sign the transaction and produce a response QR code which appears on your screen and is read back by the merchant.
This way, your phone doesn't need to connect back to the payment gateway provider at all. This is an advantage if there is bad reception inside the store, or your provider is having a bad day, or your pre-paid plan ran out, or you only have an iPod and not a smart phone. Banks could probably even produce dedicated devices that performed only this function and provide them to customers.
Actually, I'm pretty sure EFTPOS does cost the merchant, but it's a small flat fee per transaction (I think something on the order of 10c), rather than a percentage of the sale that a credit card company charges. This is why most merchants do have a minimum EFTPOS transaction amount (often $5 or $10).
Does anyone know of an open source stack that can achieve the server side of this functionality? In other words, Grab an OpenGL window, encode as video in real time, and stream it with low latency while using a protocol like VNC or NX to feed user controls back to the server.
Seems like it would be a fun thing to experiment with, maybe play games from a tablet in the living room while the more capable PC chugs away in another room. Mobile devices have much better video decoding than 3d capability.
What the GP is getting as is that the theoretically most efficient aerostat you can build is one with a rigid shell and an evacuated interior. It's not really a balloon, per se, hence the confusion.
Any actual balloon full of gas will always have less density differential than this, and thus generate less lift.
In practice, the mass of extra material required to build a rigid shell generally outweighs any extra lift you could get over a hydrogen or helium balloon. Hence, you don't see evacuated aerostats outside science fiction (e.g. Diamond Age by Neal Stephenson).
I think you may have problems with your eyes if clear 3D glasses make everything dark for you. They don't do anything like that for me.
Except, of course, that 3D glasses work using polarization, making them effectively 50% neutral-density filters.
You might not notice in some circumstances, because your eyes (brain) are pretty good at compensating for different over-all light levels. However, a bad projection setup could easily make the difference noticeable. For instance if you're sitting at the end of a row and the aisle lighting is shining into the corner of your eye without passing through the glasses, or if the projection brightness is just too low, resulting in a lack of dynamic range when cut by 50%.
Sure they had a chance to read and agree. I modified my browser to send out a custom HTTP header with every request:
X-Binding-Contract: You hereby agree to transfer copyright ownership of any material returned in response to this HTTP request to the requesting party.
Most likely the final character was chopped by the 140 character limit. The published message appears to have only 137 characters, but perhaps there were some double spaces or something in the original that were lost. I know of another instance of exactly this happening recently (to a high-profile online media company).
The collision rate on bit.ly links appears to be quite high. In addition to the truncated URL being valid, running through all 52 possibilities appending a single letter to bNCAV, about two thirds of them are valid too.
On all the internet forums I'm on, people from Australia complain constantly about their slow speeds and Draconian caps.
Unfortunately, these things are mostly the result of expensive data links between Australia and the rest of the world (where the content is). The NBN is not going to change this.
What the NBN will enable is better access to domestic services like IP television, VPNs for telecommuting, VoIP, and so forth.
I have a legitimate question for any Aussies on/. Here in the US, the title "Liberal" refers to spineless douchebags who act like conservatives with their own money, property, etc., but who love to micromanage other people's money, property, and selves. Are Aussie Liberals the same as US Liberals?
Actually, looking in from the outside, it seems to me that in the USA the term "liberal" is a meaningless epithet applied by the conservative media to anyone that they don't like.
In Australia the term "Liberal" means "a member of the Liberal Party of Australia", or a person who regularly votes for the same.
Actually, TFA implies that it's not a DDoS, but some other kind(s) of DoS being used. Perhaps something like Slowloris, or exploiting other unreleased server vulnerabilities.
No, some things are also stolen, then "invented" by Apple too.
Seriously, when 10.04 came out with the buttons in the upper left, I thought it was misguided too. But for kicks, I decided to leave them that way to see if it was actually usable for me. Guess what? After a day or two, I liked it. I haven't changed them.
Actually it is misguided. Given that "my other computer is a Mac" I figured I'd be fine with it and give it a try. Within 10 minutes I'd decided I'd never use it and switched the buttons back to the right. You know why? Because I went for the File menu, overshot and hit the close button by accident.
The top-left close button works in OS X because the menu bar is stuck to the top of the screen, not the window. Unless Gnome Global Menu is added by default, it's a bad UI design choice.
Cases like this are ones where the US government (assuming these are US patents) should step up and use their powers of eminent domain to acquire these patents, declare H.264 a government standard (like AES and DES before it) and release the patents (or a perpetual license thereto) into the public domain.
The developers of H.264 and other codecs have certainly put in a lot of research and hard thinking. I believe their algorithms should be patentable (as opposed to "software" patents that are really on UI patterns and business methods), and I believe the inventors should be justly compensated. However, for maximal furtherance of the creative arts, information interchange formats need to be standardized and unencumbered. The visual entertainment industry contributes far, far more to the US economy than the codec-designing industry, and always will. An indirect subsidy like this would be an excellent stimulus.
If they had any sense, the MPAA would by lobbying the government to make this happen, rather than trying to shore up their old distribution models with copyright crackdowns. Getting free (gratis), standard, H.264 decoders into the hands of billions of people worldwide would give them a huge boost to their audience. Unfortunately, since they represent motion picture distributors, they're probably in favour of steep licensing fees to keep the barrier to entry high for content producers wishing to distribute independently.
The "include service in the price and make tips optional" model works pretty well for everywhere else in the world but North America.
Everyone pays, let's say, 15% more. "Regular" people don't have to pay 20% to make up for the cheapskates and tourists who don't understand the unwritten rules, and floor staff don't get stiffed.
You don't stream to iPhone using RTSP, you use Apple's HTTP Live Streaming. Basically, you break your broadcast up into chunks, and create a playlist file which is updated as new chunks are added. The phone polls this file and plays back the video accordingly.
The trouble is that the latency between content creation and playback is a multiple of the chunk size. Apple recommend about 10s per chunk, which gives a latency of around 30s. That's 10s to fill up the first chunk so the playlist can be created, then 10-20s for the client to grab 1 full chunk for playback, and possibly another to ensure seamless transition between chunks (it's not explicit, but it seems that the decision to download this second chunk download may be dependent on network speed).
A 30s latency is probably fine if you want to use this for streaming your MythTV episodes to your iPhone to watch in bed, but totally useless for video conferencing for instance.
You are actually consuming those services, where as data is not really consumed, it is not gone/used up when you are done, there is still a virtually unlimited amount (as long at there is electricity there can be data).
No. The capacity of the network at any given time is finite. You are using a fraction of that available bandwith for some period of time.
Please stop with the "$/byte" calculations for SMS. It's not a data channel, it's a messaging service. It costs 55c or whatever to send a letter in the post. If you write a short letter to your grandma it costs a silly amount per byte too.
If it irks you, get a phone with Internet and send email.
As someone who's lived with transfer limits for some time, I really don't see this as anything but unavoidable in a mature market. We all know that a connection advertised as "8 Mbps" cannot be used at that capacity by all subscribers at all times. It's an 8 Mbps burstable link, but with no indication of what the sustained capacity is. For home use it makes little sense to the consumer to quote a sustained transfer rate, because home users are rarely using data in a slow steady stream. Rather, it's makes more sense to quote the integral -- i.e. total data transferred.
Networks have a finite capacity, and the people who should be paying for capacity and upgrades are the people who are using it. I don't have an unlimited calls plan on my mobile phone. I make maybe 1-2 calls per month, so I have a $0/mo plan (yes, you read that right) with higher call charges. Some people I know pay $70/mo with a lot of included calls. I'm very happy that these two options exist.
Most of the time, when people object to metered usage it's because they're in the top 1% of users, and want to continue being subsidized by others. The other common complaint is that the ISP will charge too much. Well, greedy companies with monopoly or cartel control over the service can always over charge. It's not something inherent in the billing model. In my market, the large one-time monopoly (Telstra) have a history of selling "starter" plans with ridiculously small limits, and then billing obscenely for excess ($150/GB at one point). Other smaller providers provide better packages, perks like unlimited off-peak usage and unmetered game servers, and reasonable excess usage fees (in the $3/GB range).
The conductivity of hair is very low. I know this because I have inadvertently applied 600 V between 3/4" of hair and my (thankfully dry and unsweaty) skull, yet I live to type about it.
Meh. What was the amperage of those 600V?
Not much. That's the GP's point. High voltage delivering non-lethal current means that the resistance was high*. Assuming the internal resistance of the 600V source was low, that implies the resistance of the hair is high.
* Or maybe the source had very limited energy, like a high voltage, low capacitance capacitor.
The MTA told the Stamford Advocate that without a license, the iPhone application might provide inaccurate information.
Uh huh. If you don't give the MTA money, that causes your data to be inaccurate. Thanks for your concern, MTA, but we'll take our chances...
Reminds me of the landlady I had in London. She told us overnight guests were prohibited because of fire safety. "12 backpackers burned to death in Amsterdam just last week". About a month later we got a message that overnight guests were fine if they paid 10 pounds...
The reason nobody comments on your voting system is firstly because it's trying to solve problems that don't exist. Counting the votes at one polling place is not an onerous task, and election rigging is far more likely to happen by bribing the returning officers at a polling place, or preventing/intimidating people out of voting than by someone messing with the votes themselves.
Also your long-winded justification skims over critical flaws like token forgery. For instance at one point you suggest using 1c pieces as voting tokens. Sure this might be convenient, but having a metal detector tuned to detect the smuggling of a single penny would necessitate voters taking off all shoes, belts, and jewelry, leaving their wallets with an official and all kinds of other nonsense. You do irreversible harm to your credibility by suggesting something so ridiculous.
Your system is also impractical for many real-world elections that have preferential voting and dozens or hundreds of candidate. Think about the European Parliament elections, or here in Australia where we have a preferential voting, and senate elections can have close to 100 candidates on one ballot paper (I'm not saying this is a good thing, but it's just the way things are). Your system would be completely useless in these cases.
As far as I know, all jurisdictions in Australia use good old-fashioned "put a 1 in the box" paper ballots, hand counted. Sure, we have less than 10% of the population of the USA, but vote counting uses a divide-and-conquer method so this is not really meaningful (and you have more people available to count!)
We also have compulsory voting, so if you don't show up at a poling place, the government sends you a letter asking you to explain yourself or be fined. If a lot of people write back saying "my boss wouldn't let me out of work" (unlikely since all elections are on Saturdays), or "the police set up road-blocks and carried out 3 hour 'safety' inspections on everyone's car" then this can be investigated.
Slashdot Mirror
A more interesting type of system would use a QR code challenge-response. A small screen on the gumball machine, or at the supermarket checkout flashes a QR code. You point your phone camera at it and details of the transaction come up on the screen. If you hit "confirm", your private key is used to sign the transaction and produce a response QR code which appears on your screen and is read back by the merchant.
This way, your phone doesn't need to connect back to the payment gateway provider at all. This is an advantage if there is bad reception inside the store, or your provider is having a bad day, or your pre-paid plan ran out, or you only have an iPod and not a smart phone. Banks could probably even produce dedicated devices that performed only this function and provide them to customers.
Actually, I'm pretty sure EFTPOS does cost the merchant, but it's a small flat fee per transaction (I think something on the order of 10c), rather than a percentage of the sale that a credit card company charges. This is why most merchants do have a minimum EFTPOS transaction amount (often $5 or $10).
Does anyone know of an open source stack that can achieve the server side of this functionality? In other words, Grab an OpenGL window, encode as video in real time, and stream it with low latency while using a protocol like VNC or NX to feed user controls back to the server.
Seems like it would be a fun thing to experiment with, maybe play games from a tablet in the living room while the more capable PC chugs away in another room. Mobile devices have much better video decoding than 3d capability.
.. and come back when you've got a port of Space Taxi.
What the GP is getting as is that the theoretically most efficient aerostat you can build is one with a rigid shell and an evacuated interior. It's not really a balloon, per se, hence the confusion.
Any actual balloon full of gas will always have less density differential than this, and thus generate less lift.
In practice, the mass of extra material required to build a rigid shell generally outweighs any extra lift you could get over a hydrogen or helium balloon. Hence, you don't see evacuated aerostats outside science fiction (e.g. Diamond Age by Neal Stephenson).
mmmmmmm fourth!!!
fourth != Forth
I think you may have problems with your eyes if clear 3D glasses make everything dark for you. They don't do anything like that for me.
Except, of course, that 3D glasses work using polarization, making them effectively 50% neutral-density filters.
You might not notice in some circumstances, because your eyes (brain) are pretty good at compensating for different over-all light levels. However, a bad projection setup could easily make the difference noticeable. For instance if you're sitting at the end of a row and the aisle lighting is shining into the corner of your eye without passing through the glasses, or if the projection brightness is just too low, resulting in a lack of dynamic range when cut by 50%.
Sure they had a chance to read and agree. I modified my browser to send out a custom HTTP header with every request:
X-Binding-Contract: You hereby agree to transfer copyright ownership of any material returned in response to this HTTP request to the requesting party.
It seems some people are making progress on the glucose fuel cell.
Current tech can only produce about 6.5 microwatts per cell, which is not nearly enough for a regular LED, but it's a step in the right direction.
Most likely the final character was chopped by the 140 character limit. The published message appears to have only 137 characters, but perhaps there were some double spaces or something in the original that were lost. I know of another instance of exactly this happening recently (to a high-profile online media company).
The collision rate on bit.ly links appears to be quite high. In addition to the truncated URL being valid, running through all 52 possibilities appending a single letter to bNCAV, about two thirds of them are valid too.
On all the internet forums I'm on, people from Australia complain constantly about their slow speeds and Draconian caps.
Unfortunately, these things are mostly the result of expensive data links between Australia and the rest of the world (where the content is). The NBN is not going to change this.
What the NBN will enable is better access to domestic services like IP television, VPNs for telecommuting, VoIP, and so forth.
I have a legitimate question for any Aussies on /. Here in the US, the title "Liberal" refers to spineless douchebags who act like conservatives with their own money, property, etc., but who love to micromanage other people's money, property, and selves. Are Aussie Liberals the same as US Liberals?
Actually, looking in from the outside, it seems to me that in the USA the term "liberal" is a meaningless epithet applied by the conservative media to anyone that they don't like.
In Australia the term "Liberal" means "a member of the Liberal Party of Australia", or a person who regularly votes for the same.
Actually, TFA implies that it's not a DDoS, but some other kind(s) of DoS being used. Perhaps something like Slowloris, or exploiting other unreleased server vulnerabilities.
Save the headphones for the bus, train or killing time.
Indeed. I always put my headphones on when it's killing time.
No, some things are also stolen, then "invented" by Apple too.
Seriously, when 10.04 came out with the buttons in the upper left, I thought it was misguided too. But for kicks, I decided to leave them that way to see if it was actually usable for me. Guess what? After a day or two, I liked it. I haven't changed them.
Actually it is misguided. Given that "my other computer is a Mac" I figured I'd be fine with it and give it a try. Within 10 minutes I'd decided I'd never use it and switched the buttons back to the right. You know why? Because I went for the File menu, overshot and hit the close button by accident.
The top-left close button works in OS X because the menu bar is stuck to the top of the screen, not the window. Unless Gnome Global Menu is added by default, it's a bad UI design choice.
Cases like this are ones where the US government (assuming these are US patents) should step up and use their powers of eminent domain to acquire these patents, declare H.264 a government standard (like AES and DES before it) and release the patents (or a perpetual license thereto) into the public domain.
The developers of H.264 and other codecs have certainly put in a lot of research and hard thinking. I believe their algorithms should be patentable (as opposed to "software" patents that are really on UI patterns and business methods), and I believe the inventors should be justly compensated. However, for maximal furtherance of the creative arts, information interchange formats need to be standardized and unencumbered. The visual entertainment industry contributes far, far more to the US economy than the codec-designing industry, and always will. An indirect subsidy like this would be an excellent stimulus.
If they had any sense, the MPAA would by lobbying the government to make this happen, rather than trying to shore up their old distribution models with copyright crackdowns. Getting free (gratis), standard, H.264 decoders into the hands of billions of people worldwide would give them a huge boost to their audience. Unfortunately, since they represent motion picture distributors, they're probably in favour of steep licensing fees to keep the barrier to entry high for content producers wishing to distribute independently.
The "include service in the price and make tips optional" model works pretty well for everywhere else in the world but North America.
Everyone pays, let's say, 15% more. "Regular" people don't have to pay 20% to make up for the cheapskates and tourists who don't understand the unwritten rules, and floor staff don't get stiffed.
You don't stream to iPhone using RTSP, you use Apple's HTTP Live Streaming. Basically, you break your broadcast up into chunks, and create a playlist file which is updated as new chunks are added. The phone polls this file and plays back the video accordingly.
The trouble is that the latency between content creation and playback is a multiple of the chunk size. Apple recommend about 10s per chunk, which gives a latency of around 30s. That's 10s to fill up the first chunk so the playlist can be created, then 10-20s for the client to grab 1 full chunk for playback, and possibly another to ensure seamless transition between chunks (it's not explicit, but it seems that the decision to download this second chunk download may be dependent on network speed).
A 30s latency is probably fine if you want to use this for streaming your MythTV episodes to your iPhone to watch in bed, but totally useless for video conferencing for instance.
You are actually consuming those services, where as data is not really consumed, it is not gone/used up when you are done, there is still a virtually unlimited amount (as long at there is electricity there can be data).
No. The capacity of the network at any given time is finite. You are using a fraction of that available bandwith for some period of time.
Bandwidth x Time = Bytes Transferred.
Please stop with the "$/byte" calculations for SMS. It's not a data channel, it's a messaging service. It costs 55c or whatever to send a letter in the post. If you write a short letter to your grandma it costs a silly amount per byte too.
If it irks you, get a phone with Internet and send email.
As someone who's lived with transfer limits for some time, I really don't see this as anything but unavoidable in a mature market. We all know that a connection advertised as "8 Mbps" cannot be used at that capacity by all subscribers at all times. It's an 8 Mbps burstable link, but with no indication of what the sustained capacity is. For home use it makes little sense to the consumer to quote a sustained transfer rate, because home users are rarely using data in a slow steady stream. Rather, it's makes more sense to quote the integral -- i.e. total data transferred.
Networks have a finite capacity, and the people who should be paying for capacity and upgrades are the people who are using it. I don't have an unlimited calls plan on my mobile phone. I make maybe 1-2 calls per month, so I have a $0/mo plan (yes, you read that right) with higher call charges. Some people I know pay $70/mo with a lot of included calls. I'm very happy that these two options exist.
Most of the time, when people object to metered usage it's because they're in the top 1% of users, and want to continue being subsidized by others. The other common complaint is that the ISP will charge too much. Well, greedy companies with monopoly or cartel control over the service can always over charge. It's not something inherent in the billing model. In my market, the large one-time monopoly (Telstra) have a history of selling "starter" plans with ridiculously small limits, and then billing obscenely for excess ($150/GB at one point). Other smaller providers provide better packages, perks like unlimited off-peak usage and unmetered game servers, and reasonable excess usage fees (in the $3/GB range).
Meh. What was the amperage of those 600V?
Not much. That's the GP's point. High voltage delivering non-lethal current means that the resistance was high*. Assuming the internal resistance of the 600V source was low, that implies the resistance of the hair is high.
* Or maybe the source had very limited energy, like a high voltage, low capacitance capacitor.
Or would you like your internet connection to be served by a SUV carrying hard drives?
Never underestimate the bandwidth of a fedex truck packed with 250 lbs of hard disks!
Depending on the file size of what you would be downloading and with what technology, overnight shipping might STILL be better latency too!
Right, but remember that full hard drives weigh more than empty ones, so you only want to buy about 200 lb of empty drives if you have a 250 lb limit.
The MTA told the Stamford Advocate that without a license, the iPhone application might provide inaccurate information.
Uh huh. If you don't give the MTA money, that causes your data to be inaccurate. Thanks for your concern, MTA, but we'll take our chances...
Reminds me of the landlady I had in London. She told us overnight guests were prohibited because of fire safety. "12 backpackers burned to death in Amsterdam just last week". About a month later we got a message that overnight guests were fine if they paid 10 pounds...
Hi Paul!
The reason nobody comments on your voting system is firstly because it's trying to solve problems that don't exist. Counting the votes at one polling place is not an onerous task, and election rigging is far more likely to happen by bribing the returning officers at a polling place, or preventing/intimidating people out of voting than by someone messing with the votes themselves.
Also your long-winded justification skims over critical flaws like token forgery. For instance at one point you suggest using 1c pieces as voting tokens. Sure this might be convenient, but having a metal detector tuned to detect the smuggling of a single penny would necessitate voters taking off all shoes, belts, and jewelry, leaving their wallets with an official and all kinds of other nonsense. You do irreversible harm to your credibility by suggesting something so ridiculous.
Your system is also impractical for many real-world elections that have preferential voting and dozens or hundreds of candidate. Think about the European Parliament elections, or here in Australia where we have a preferential voting, and senate elections can have close to 100 candidates on one ballot paper (I'm not saying this is a good thing, but it's just the way things are). Your system would be completely useless in these cases.
As far as I know, all jurisdictions in Australia use good old-fashioned "put a 1 in the box" paper ballots, hand counted. Sure, we have less than 10% of the population of the USA, but vote counting uses a divide-and-conquer method so this is not really meaningful (and you have more people available to count!)
We also have compulsory voting, so if you don't show up at a poling place, the government sends you a letter asking you to explain yourself or be fined. If a lot of people write back saying "my boss wouldn't let me out of work" (unlikely since all elections are on Saturdays), or "the police set up road-blocks and carried out 3 hour 'safety' inspections on everyone's car" then this can be investigated.