Embed the token into the cards. They don't have a significant cost these days, and it would make the cards significantly more secure. Yes, it makes the cards more expensive than a piece of plastic and a magstripe, but really, it's not THAT much. Particularly when amortized over all the cards in circulation.
If you're going that far, you could also include the PIN entry keypad on the card and use a secure link to make it nearly impossible for an attacker to get your PIN via the capture device.
And, if designed properly, they won't wear out as fast as the old style ones, and they are more secure, so don't have to expire as often. The real expiration is on the CC company servers anyway, and checked when you try to use the card.
The really painful part isn't the cards really, it's the readers. And internet transactions, but that can be handled reasonably if you have a display on the card. It can show you a bunch of numbers to type into the computer after you tell it how much you want to allow the merchant to charge you. Generates a time limited code (one use, good for one minute?) that allows the transaction to process.
There's also Germanium diodes with a forward voltage drop of about 0.2V. That might still be too much as a typical bridge has to pass current through 2 of them though. I suppose you could then use a boost converter, but that all seems quite wasteful.
So you build a bigger server for the primary, and demote the old disks to backup duty.:)
As mentioned, RAID is better than nothing. Particularly with a good filesystem like ZFS. Another RAID to store copies of all the really important data on another server, preferably offsite, and your backup is quite good. Snapshots and/or a versioning backup app to handle things like the user deleting or changing a file and wanting the old version back, and you're in pretty good shape.
Exactly, which is one of the first features I look for in a motherboard. Asus has had an excellent implementation for years, Gigabyte does quite well too.
How much for the drives? And you need at least 2 for DR as the first one could die or get destroyed in the fire that killed the servers. For large companies that can pay for those drives over many systems being backed up, LTO might make sense even then. For home users, and even mid-size business users, the costs for the drives kill LTO out of the gate. Tape has its advantages, but the full system cost kills it for a very large market. Of course, I would never have referred to tape as "dead" as there is still good reason to use it, if you can afford it. It would be nice if there were a drive available for those tapes that was affordable, but home/small business tape backup has always been plagued with crap hardware that eats tapes and such.
For most users that can't afford it, it's cheaper and more reliable to run ZFS capable servers in multiple locations with some kind of backup and snapshot jobs running to keep data and history data available. I do this for my personal data using old parts to run the offsite backup server. Seed it locally before taking it to the new location and the incrementals are small enough to go over the internet if needed. In my case it's next door, so a wireless bridge is faster. And with ZFS I can know for sure the data is good with a scrub monthly.
They do show up in "Manage Applications", however, looking at the manifest file, it doesn't look like the app requested them. Interesting. In the logs, I have seen apps granted "implicit" permissions, I wonder if that happened here.
Same here, I just followed a link the developer posted in this story to download the APK file and installed it to my G1. No problem. Nice app, now half my day will be wasted playing it, so thanks to the developer, and in a weird way, to the Tetris company.:)
Of course, you could just not use your carrier's version of the OS. I believe all the Android devices are capabile of being rooted at this point. Certainly a bigger hoop to jump than clicking "Unknown Sources" though. I haven't run TMobile's version of Android on my G1 since the day after a got the thing. Another bonus is that I'm running 2.1, which is not available for G1 from official sources.:)
If only TMO would get a decent spec Android phone with a keyboard. I'm already running faster than even the MT-Slide coming out soon. 650Mhz overclock mod.:)
I can't speak for anyone else, but I'm *MUCH* more willing to respect IP laws when the IP holders respect the purchasers of those copies and the laws are reasonable. No, perpetual copyright is NOT reasonable and I don't respect that. Copyright is supposed to be a TRADE, the creator is able to have a monopoly protected by government for a short time, and in return the public gets open access to it later. The media companies don't want to hold to their end of the bargain. If they want us to respect their rights, perhaps they should start by respecting ours.
If they want to boost sales from many current "pirates", they really only need to do a few things.
1) Provide media in the format people want at a decent price. That means no restrictions on what can play it back. 2) No restrictions on playback and skipping around. No non-skippable ads, previews, warnings, or other content. Preferably, the main movie starts first with the OPTION to use a menu. 3) No DRM 4) Instant high speed and quality downloads 5) Reasonable copyright terms (20 years, max, IMO) 6) Provide re-download ability for damaged media, deleted/corrupted files, etc. Perhaps with a small fee ($5?) for bandwidth and server resources. 7) Stop all the stupid lawsuits 8) Automation friendly interfaces so I could, for example, auto-purchase and download the latest episode of a TV show or queue a movie up for auto-download when it's released.
There will always be people willing to "steal" the content. Get over it and provide a safe, quality, fast service with good prices. No, current DVD/BD prices won't do. I don't get any physical media, cases, liners, etc. It needs to be cheaper than the physical object. Do this, and many current downloaders will download from you and pay for the privilege. Thanks to recent BS pulled by the various companies, getting people to trust your downloads might be difficult. You can thank Sony for the rootkit incident, for example. However, it can be done. Guarantee no executable data will be downloaded from your servers, just the video/audio streams. That might help. Use well known, preferably open container formats with known codecs. MKV containers with h264/AC3/DTS streams + subtitles in various languages would be one option. I believe MP4 containers have the same sorts of capabilities. No EXEs, at least not by default. Provide a player app people can use if they choose to if you like, but it can't be required.
Reading the FCCs plan, they did NOT want to ban QoS. So you could prioritize based on protocol just fine. Large bulk transfers like p2p, and log running http/ftp sessions can stand to be slowed down a bit to let the latency sensitive traffic through as it's generally consuming fewer bytes/sec as well. The catch is you get "tragedy of the commons" effects when users figure out that you can tunnel everything over short lived https or SSH connections to bypass basic QoS. So throttle based on overall usage instead. If you use a lot, you get lower priority. Use a sliding window and you should be able to get everyone's important traffic through without too much trouble without having to deal with protocol specifics or destination/source IP. The important bit is that you disclose this all up front so everyone knows what they are signing up for. Perhaps allowing normal QoS of VOIP streams and similar with a caveat that users caught attempting to abuse those QoS setups will be banned. Or you could throttle throughput for QoS streams. VOIP shouldn't need more than, what, 128kbps? ISDN was 64kbps per phone line and that used minimal compression. Adjust based on real uses and p2p users won't WANT to abuse QoS. If routers can't do this, that is an engineering problem that I don't see as being terribly difficult to work around.
The important bit is to give users something when you take something. Make it a fair bargain, and disclose it up front. No fine print, no hidden TOS, no draconian "cut off when you use an undisclosed amount of transfer". What most supporters of Net Neutrality want to prevent is BS like Comcast degrading service to Vonage, YouTube, Hulu, and others to make Comcast services work better than the competition. Or to prevent them from trying to strongarm Google and other highly used sites into paying protection money to avoid being blocked or throttled into uselessness. That sort of thing is what the FCC plan was about. Go read it, and quote for us the parts you find objectionable so we have a common ground to discuss on if you like.
If you just want ATSC, get an HDHomeRun. Network connected, so no kernel level drivers. IR receiver included, open specs for the protocols. Fully supported in MythTV. I had the same experience with Myth years ago because I did some research and bought hardware that worked. If you do that, getting Myth working is pretty easy really. I set up the hardware, installed MythBuntu, and it "just worked" as far as the basics went. I did a lot of tweaking to get things the way I wanted them, but recording TV, watching recordings, and playing videos worked fine. It's even better now as you can use NVidia's VDPAU to handle playback so you can use low-power CPU devices like the Atom/ION combo. I used to need a fast dual core CPU, now an Atom works great.
All Win7 gives you over Linux/Myth in an HTPC is driver support, IMO. And even that is up to the hardware vendor to get right.
Hardware support is one of the biggest reasons I bought an HDHomeRun. No kernel level drivers required, and an open spec for communication meant that any program could add support for it in userspace without any more info needed. And I got a fully supported IR receiver out of it as well.
If you want CableCard, talk to CableLabs, they are the reason Myth (and all OSS projects) can't support CableCard. It's not "DRM-related poo flinging", it's stating the simple fact. If the government here had any balls, they would have required open access as I can't tune the CableCard device to a channel I'm not authorized for anyway. But they gov caved to the cable and MPAA lobby and here we sit with no access. You can use a normal capture card and an IR blaster, HD-PVR for HD channels. If you want this to change, convince some members of congress.
And it sucks on Windows too, just not as much. When that same machine using a real video player can decode the video just fine when it's captured and extracted from the flash container. Flash video sucks hard on Linux as well, so it's not just OSX users that get screwed by Adobe. Flash blows, and it can't get replaced fast enough for my taste. Adobe has proven that they are unable or unwilling to hire programmers that can code their way out of a wet paper bag.
You might want to add bitrate and codec to the quality requirement. 1080p is worthless if it's low bitrate. Decent 1080p h264 would probably weigh in at about 3Gbyte/hour. Excellent quality is probably double that. I'm fine with the file size and download time. And even the price, provided those requirements are met. I'd also prefer an MKV container, just because it's so flexible, with at least an english subtitle track. Other languages would be nice for our friends in other nations. Oh, they should be able to buy and download as well, with no delay from the US release. I also need download to at least be an option over streaming. Offer a stream if you like, but the file needs to be available to download. I'm even willing to use bittorrent and seed for a while if they would like.
Oh, and no weird download clients. Standard, existing protocols only. It must work in Windows, OSX, and Linux at a minimum. Use a known protocol like HTTP, HTTPS, BitTorrent, FTP, etc. and it's a no brainer.
Not the mentioned hardware accelerated on EVERYTHING. My cell phone has hardware acceleration for h264. OGG? no. VP8? no. Can the CPU do it? no. Well, h264 it is then. It's fine to say we should push for open codecs, but when I can't play the videos encoded with them on my equipment...... Google and VP8 are probably our best chance here, if Google can push for hardware supported VP8 in Android equipment, they might be able to stem the tide. If they care. They already have h264 licenses.
In the case of ISPs, we don't even need that. All ISPs can put data on the wires for a set price. Perhaps with various $/GB $/Mbps rates (decided by a PUC?). Use fiber and there's more than enough bandwidth for everyone to share. The pricing must be the same for all ISPs, that would be my only real requirement on it. Much like the way the government builds roads and everyone pays to use them via the gas taxes etc.. I can have a package shipped to my door by USPS, UPS, Fedex, DHL, Joe down the street, whatever. Open competition.
They already did, and delegated that authority to the FCC. If you think that's wrong, write your congressman I suppose. All the alphabet agencies are created under basically the same rules though, so I don't expect congress to change the rules now.
I agree with the basic principle, but you're overlooking the fact that the telecom industry is NOT a free market and hasn't been for over 100 years. They have been granted monopoly status by government action. That's exactly the opposite of a free market. Cable TV has never been a free market. There is no competition that is legally allowed to come in and fight back. The only hope of any competition is wireless, but they can't compete with the raw speed a wired line can. And the entrenched monopolies can just lower prices and push speeds up to force new competition out of business. With protected monopoly profits no less.
In this industry, there is no freedom, no free market. Even removing the laws preventing competition isn't enough in this case. The existing companies also got huge subsidies and tax breaks to pay for the networks. Can you think of a business plan that can compete with that?
So they get to choose who connects to it. Simple as that. If you want to bring a personal machine in for personal, non job related use, accept that you might not have connectivity. Most of the hospitals around here have a guest wifi, you might be able to use that, or a 3G card. For job related stuff, tell them they have to provide the equipment.
If you have read the HIPPA laws, the penalties for leaking PII are severe. Full-disk encryption for all connected machines is probably the best way to prevent problems with such things. It would be nice if they would let you just use TrueCrypt and install it yourself, but IT departments tend to just set a standard policy for everyone. That way they can audit the policy and such. You wouldn't want to have to support everyone doing their own thing either, to be fair.
Install Autokiller from the market and set it to aggressive mode. It keeps my phone running nice and fast. It might still kill your music player though, I don't use it so I don't know for sure. I've heard there are ways to keep it from killing particular apps, you might ask on XDA about that.
Speed. Stock 2.x ROMs run SLOW on a G1, even with swap/Compcache enabled. Other builds are significantly faster. Try SuperD or SuperEclair, very fast compared to even stock 1.6 ROMs.
Slashdot Mirror
Embed the token into the cards. They don't have a significant cost these days, and it would make the cards significantly more secure. Yes, it makes the cards more expensive than a piece of plastic and a magstripe, but really, it's not THAT much. Particularly when amortized over all the cards in circulation.
If you're going that far, you could also include the PIN entry keypad on the card and use a secure link to make it nearly impossible for an attacker to get your PIN via the capture device.
And, if designed properly, they won't wear out as fast as the old style ones, and they are more secure, so don't have to expire as often. The real expiration is on the CC company servers anyway, and checked when you try to use the card.
The really painful part isn't the cards really, it's the readers. And internet transactions, but that can be handled reasonably if you have a display on the card. It can show you a bunch of numbers to type into the computer after you tell it how much you want to allow the merchant to charge you. Generates a time limited code (one use, good for one minute?) that allows the transaction to process.
http://en.wikipedia.org/wiki/Schottky_diode
There's also Germanium diodes with a forward voltage drop of about 0.2V. That might still be too much as a typical bridge has to pass current through 2 of them though. I suppose you could then use a boost converter, but that all seems quite wasteful.
So you build a bigger server for the primary, and demote the old disks to backup duty. :)
As mentioned, RAID is better than nothing. Particularly with a good filesystem like ZFS. Another RAID to store copies of all the really important data on another server, preferably offsite, and your backup is quite good. Snapshots and/or a versioning backup app to handle things like the user deleting or changing a file and wanting the old version back, and you're in pretty good shape.
Exactly, which is one of the first features I look for in a motherboard. Asus has had an excellent implementation for years, Gigabyte does quite well too.
How much for the drives? And you need at least 2 for DR as the first one could die or get destroyed in the fire that killed the servers. For large companies that can pay for those drives over many systems being backed up, LTO might make sense even then. For home users, and even mid-size business users, the costs for the drives kill LTO out of the gate. Tape has its advantages, but the full system cost kills it for a very large market. Of course, I would never have referred to tape as "dead" as there is still good reason to use it, if you can afford it. It would be nice if there were a drive available for those tapes that was affordable, but home/small business tape backup has always been plagued with crap hardware that eats tapes and such.
For most users that can't afford it, it's cheaper and more reliable to run ZFS capable servers in multiple locations with some kind of backup and snapshot jobs running to keep data and history data available. I do this for my personal data using old parts to run the offsite backup server. Seed it locally before taking it to the new location and the incrementals are small enough to go over the internet if needed. In my case it's next door, so a wireless bridge is faster. And with ZFS I can know for sure the data is good with a scrub monthly.
They do show up in "Manage Applications", however, looking at the manifest file, it doesn't look like the app requested them. Interesting. In the logs, I have seen apps granted "implicit" permissions, I wonder if that happened here.
Same here, I just followed a link the developer posted in this story to download the APK file and installed it to my G1. No problem. Nice app, now half my day will be wasted playing it, so thanks to the developer, and in a weird way, to the Tetris company. :)
Of course, you could just not use your carrier's version of the OS. I believe all the Android devices are capabile of being rooted at this point. Certainly a bigger hoop to jump than clicking "Unknown Sources" though. I haven't run TMobile's version of Android on my G1 since the day after a got the thing. Another bonus is that I'm running 2.1, which is not available for G1 from official sources. :)
If only TMO would get a decent spec Android phone with a keyboard. I'm already running faster than even the MT-Slide coming out soon. 650Mhz overclock mod. :)
I can't speak for anyone else, but I'm *MUCH* more willing to respect IP laws when the IP holders respect the purchasers of those copies and the laws are reasonable. No, perpetual copyright is NOT reasonable and I don't respect that. Copyright is supposed to be a TRADE, the creator is able to have a monopoly protected by government for a short time, and in return the public gets open access to it later. The media companies don't want to hold to their end of the bargain. If they want us to respect their rights, perhaps they should start by respecting ours.
If they want to boost sales from many current "pirates", they really only need to do a few things.
1) Provide media in the format people want at a decent price. That means no restrictions on what can play it back.
2) No restrictions on playback and skipping around. No non-skippable ads, previews, warnings, or other content. Preferably, the main movie starts first with the OPTION to use a menu.
3) No DRM
4) Instant high speed and quality downloads
5) Reasonable copyright terms (20 years, max, IMO)
6) Provide re-download ability for damaged media, deleted/corrupted files, etc. Perhaps with a small fee ($5?) for bandwidth and server resources.
7) Stop all the stupid lawsuits
8) Automation friendly interfaces so I could, for example, auto-purchase and download the latest episode of a TV show or queue a movie up for auto-download when it's released.
There will always be people willing to "steal" the content. Get over it and provide a safe, quality, fast service with good prices. No, current DVD/BD prices won't do. I don't get any physical media, cases, liners, etc. It needs to be cheaper than the physical object. Do this, and many current downloaders will download from you and pay for the privilege. Thanks to recent BS pulled by the various companies, getting people to trust your downloads might be difficult. You can thank Sony for the rootkit incident, for example. However, it can be done. Guarantee no executable data will be downloaded from your servers, just the video/audio streams. That might help. Use well known, preferably open container formats with known codecs. MKV containers with h264/AC3/DTS streams + subtitles in various languages would be one option. I believe MP4 containers have the same sorts of capabilities. No EXEs, at least not by default. Provide a player app people can use if they choose to if you like, but it can't be required.
Reading the FCCs plan, they did NOT want to ban QoS. So you could prioritize based on protocol just fine. Large bulk transfers like p2p, and log running http/ftp sessions can stand to be slowed down a bit to let the latency sensitive traffic through as it's generally consuming fewer bytes/sec as well. The catch is you get "tragedy of the commons" effects when users figure out that you can tunnel everything over short lived https or SSH connections to bypass basic QoS. So throttle based on overall usage instead. If you use a lot, you get lower priority. Use a sliding window and you should be able to get everyone's important traffic through without too much trouble without having to deal with protocol specifics or destination/source IP. The important bit is that you disclose this all up front so everyone knows what they are signing up for. Perhaps allowing normal QoS of VOIP streams and similar with a caveat that users caught attempting to abuse those QoS setups will be banned. Or you could throttle throughput for QoS streams. VOIP shouldn't need more than, what, 128kbps? ISDN was 64kbps per phone line and that used minimal compression. Adjust based on real uses and p2p users won't WANT to abuse QoS. If routers can't do this, that is an engineering problem that I don't see as being terribly difficult to work around.
The important bit is to give users something when you take something. Make it a fair bargain, and disclose it up front. No fine print, no hidden TOS, no draconian "cut off when you use an undisclosed amount of transfer". What most supporters of Net Neutrality want to prevent is BS like Comcast degrading service to Vonage, YouTube, Hulu, and others to make Comcast services work better than the competition. Or to prevent them from trying to strongarm Google and other highly used sites into paying protection money to avoid being blocked or throttled into uselessness. That sort of thing is what the FCC plan was about. Go read it, and quote for us the parts you find objectionable so we have a common ground to discuss on if you like.
If you just want ATSC, get an HDHomeRun. Network connected, so no kernel level drivers. IR receiver included, open specs for the protocols. Fully supported in MythTV. I had the same experience with Myth years ago because I did some research and bought hardware that worked. If you do that, getting Myth working is pretty easy really. I set up the hardware, installed MythBuntu, and it "just worked" as far as the basics went. I did a lot of tweaking to get things the way I wanted them, but recording TV, watching recordings, and playing videos worked fine. It's even better now as you can use NVidia's VDPAU to handle playback so you can use low-power CPU devices like the Atom/ION combo. I used to need a fast dual core CPU, now an Atom works great.
All Win7 gives you over Linux/Myth in an HTPC is driver support, IMO. And even that is up to the hardware vendor to get right.
HDHomeRun, and yes.
Hardware support is one of the biggest reasons I bought an HDHomeRun. No kernel level drivers required, and an open spec for communication meant that any program could add support for it in userspace without any more info needed. And I got a fully supported IR receiver out of it as well.
If you want CableCard, talk to CableLabs, they are the reason Myth (and all OSS projects) can't support CableCard. It's not "DRM-related poo flinging", it's stating the simple fact. If the government here had any balls, they would have required open access as I can't tune the CableCard device to a channel I'm not authorized for anyway. But they gov caved to the cable and MPAA lobby and here we sit with no access. You can use a normal capture card and an IR blaster, HD-PVR for HD channels. If you want this to change, convince some members of congress.
And don't use encryption or conduct business over that link. Hams have somewhat strict use rules in exchange for that RF power output. :)
And it sucks on Windows too, just not as much. When that same machine using a real video player can decode the video just fine when it's captured and extracted from the flash container. Flash video sucks hard on Linux as well, so it's not just OSX users that get screwed by Adobe. Flash blows, and it can't get replaced fast enough for my taste. Adobe has proven that they are unable or unwilling to hire programmers that can code their way out of a wet paper bag.
Also, include an open API so that users can "subscribe" to various shows. Much like the often mentioned RSS + BitTorrent approach.
You might want to add bitrate and codec to the quality requirement. 1080p is worthless if it's low bitrate. Decent 1080p h264 would probably weigh in at about 3Gbyte/hour. Excellent quality is probably double that. I'm fine with the file size and download time. And even the price, provided those requirements are met. I'd also prefer an MKV container, just because it's so flexible, with at least an english subtitle track. Other languages would be nice for our friends in other nations. Oh, they should be able to buy and download as well, with no delay from the US release. I also need download to at least be an option over streaming. Offer a stream if you like, but the file needs to be available to download. I'm even willing to use bittorrent and seed for a while if they would like.
Oh, and no weird download clients. Standard, existing protocols only. It must work in Windows, OSX, and Linux at a minimum. Use a known protocol like HTTP, HTTPS, BitTorrent, FTP, etc. and it's a no brainer.
Not the mentioned hardware accelerated on EVERYTHING. My cell phone has hardware acceleration for h264. OGG? no. VP8? no. Can the CPU do it? no. Well, h264 it is then. It's fine to say we should push for open codecs, but when I can't play the videos encoded with them on my equipment...... Google and VP8 are probably our best chance here, if Google can push for hardware supported VP8 in Android equipment, they might be able to stem the tide. If they care. They already have h264 licenses.
In the case of ISPs, we don't even need that. All ISPs can put data on the wires for a set price. Perhaps with various $/GB $/Mbps rates (decided by a PUC?). Use fiber and there's more than enough bandwidth for everyone to share. The pricing must be the same for all ISPs, that would be my only real requirement on it. Much like the way the government builds roads and everyone pays to use them via the gas taxes etc.. I can have a package shipped to my door by USPS, UPS, Fedex, DHL, Joe down the street, whatever. Open competition.
They already did, and delegated that authority to the FCC. If you think that's wrong, write your congressman I suppose. All the alphabet agencies are created under basically the same rules though, so I don't expect congress to change the rules now.
I agree with the basic principle, but you're overlooking the fact that the telecom industry is NOT a free market and hasn't been for over 100 years. They have been granted monopoly status by government action. That's exactly the opposite of a free market. Cable TV has never been a free market. There is no competition that is legally allowed to come in and fight back. The only hope of any competition is wireless, but they can't compete with the raw speed a wired line can. And the entrenched monopolies can just lower prices and push speeds up to force new competition out of business. With protected monopoly profits no less.
In this industry, there is no freedom, no free market. Even removing the laws preventing competition isn't enough in this case. The existing companies also got huge subsidies and tax breaks to pay for the networks. Can you think of a business plan that can compete with that?
So they get to choose who connects to it. Simple as that. If you want to bring a personal machine in for personal, non job related use, accept that you might not have connectivity. Most of the hospitals around here have a guest wifi, you might be able to use that, or a 3G card. For job related stuff, tell them they have to provide the equipment.
If you have read the HIPPA laws, the penalties for leaking PII are severe. Full-disk encryption for all connected machines is probably the best way to prevent problems with such things. It would be nice if they would let you just use TrueCrypt and install it yourself, but IT departments tend to just set a standard policy for everyone. That way they can audit the policy and such. You wouldn't want to have to support everyone doing their own thing either, to be fair.
Install Autokiller from the market and set it to aggressive mode. It keeps my phone running nice and fast. It might still kill your music player though, I don't use it so I don't know for sure. I've heard there are ways to keep it from killing particular apps, you might ask on XDA about that.
Speed. Stock 2.x ROMs run SLOW on a G1, even with swap/Compcache enabled. Other builds are significantly faster. Try SuperD or SuperEclair, very fast compared to even stock 1.6 ROMs.