Always thought Nintendo kick Apple's ass if they just added GSM to the DS and given it a bluetooth headset with a dialer. DS has games, a web browser, a camera(newer ones), etc. Just doesn't have a phone.
One big problem is that afaict the DS is designed like an old fashioned game console with the games accessing the hardware directly. Also I don't think the DS has any form of multitasking. So I think the only way to build a usable DSi based smartphone would be to have a seperate processor (on top of the two processors the DS has already) for the smartphone functionality and somehow build an emulated wifi chip that passed packets from the DS hardware to the smatphone hardware.
Working at a performance venue, We have signs we have to reprint every year. Also, more often, We have a schedule we post every month that gets a few changes. This could potentially save us Lots of money not having to continuously reprint on paper these schedules.
Firstly afaict printing on an economical laser printer costs arround 5 cents a sheet including paper but excluding electricity so at $2 per sheet and assuming the required thermal printer has no other consumables you would have to reuse this paper 20 times to make a saving. My experience with trying to feed used paper back through printers (to print on the other side) is it has to be in pristine condition or it doesn't feed properly so i'm skeptical about whether this will achive 20 cycles under realistic conditions.
Secondly introducing this into a business environment will mean duplicating your printers, thermal e-paper for the transient documents and regular paper in lasers for the permanent documents, thats a lot of extra infrastructure.
Thirdly it would be easy to inadvertantly give, throw away or destroy hundreds of dollars worth of e-paper just by printing a large document on the wrong printer and sending it to someone or accidently putting a document in the paper recycling bin when it should have gone in the e-paper bin.
It's a nice idea but I just don't see it being practical unless it gets a LOT cheaper.
Depends how long the bursts are. Afaict the norm is one burst per frame but there is no real reason the bursts couldn't be shorter than that if desired.
Powered USB is essentially a USB connector stacked with a seperate power connector. This makes the port more than twice as tall which afaict prevents it being placed on a standard expansion slot backplate and considerablly reduces the number of ports you can put on say an ATX backplate. Worse there are three voltage variants which are all incompatible with each other.
The result of these defeciancies is that it has only found niche appeal.
such a practice could easily be deployed via Steam for example, maybe they are doing it already....
I don't think steam currently does rentals but they do free weekends and guest passes which are much the same thing technically so they could probablly bring in rental service pretty quickly if they wanted to.
The emulators I have experiance with work by running in bursts. The "emulation clock" runs much faster than real time for the duration of the burst then stop until the next burst is due so there is no real problem with the emulation clock being a weird speed relative to PC clocks.
Afaict the main problem is determining all the performance interdependencies of the real hardware and then coding those interdependencies without adding unacceptable performance pentalties to the main emulation loop.
but any of the more popular Linux distro keeps on pumping out security updates and will likely do so for years to come.
ROFL, XP may be nearing the end of it's lifecycle but should keep getting security updates until april 2014. Afaict ubuntu lucid will stop getting security updates for desktop software in april 2013.
If you want long support lifecycles on linux you really need to get into RHEL and it's rebuilds.
IMO a good language makes the safe tools painless and the unsafe ones painful. A poor language makes the safe tools painful and the unsafe ones painless.
A web orientated language designed for security for example could have multiple string types and make it easier to apply appropriate conversion processing than to convert between them without doing the processing.
Right now if you want something better than firewire 800 or gig-e (which afaict is comparable to a single modern desktop hard drive on sequential transfers so will bottleneck an external raid array or SSD) and have a mac (other than the mac pro) you have no choice but to use thunderbolt. If you have a PC you can't use thunderbolt. So right now thunderbolt isn't in direct competition with any other interface.
But if and when thunderbolt crosses over into the PC world it will have to compete with USB3 and eSATA. It's better than them but I have my doubts as to whether it's better by a wide enough margin to displace them.
And I don't like this government a bit, but we had a vote etc.
Indeed we did but the combination of a broken system and political backroom deals meant the government we got doesn't really represent the voters. In particular the lib dems buddied up with the tories despite being traditional regarded as closer to labour.
IIRC when I tried it on XP it ran ok but you couldn't get very far on the modern web because it doesn't work with servers that use name based virtual hosting.
My experiance with microsoft in academia is they like to get us on programs that are a lot cheaper than paying for the software normally but are based on paying a subscription based on the size of the whole institution rather than paying for each individual machine.
The result is that there is no motivation to gradually migrate away from MS software since the only way to reduce the ammount paid to MS would be to virtually eliminate MS software from the institution (which is not realisitically going to happen).
User agents would be a clue (IIRC some mobile browsers spoof those but it wouldn't surprise me if there are subtule differences between the commonly spoofed strings and the common ones on desktops). There is likely other information in the http request that differs too. If the tethering software was making the phone act as a nat router TTLs and other TCP/IP options could be a clue If the tethering software worked by loading a proxy on the phone then an x-forwarded for header could be a clue Large downloads would be a clue, so would very high overall data use. As you say traffic from application that don't exist on the phone would be an especially damning clue
Afaict most mobile phone contracts have a clause allowing termination for any or no reason whatsoever so they don't exactly need absolute proof that someone is teathering to send them a "cease or pay up" letter with a threat of disconnection.
Afaict verizon is CDMA so they have to register any devices onto the network (unlike with GSM where you can drop your SIM into any device) so presumablly they would only allow such a device to be registered on an appropriate plan.
One thing i can see an automated car being bad at is choosing between accidents. Sometimes a car will be put into a situation where it will inevitablly hit something but the operator has a choice of what to hit. Say for example a motorcyclist comes off thier bike, will the automated driving system by able to disginguish between the motorcyclist and the vacant bike and be able to choose to hit the vacant motorbike rather than the motorcyclist?
The original poster's point was that often the incoming water is less sanitary than the discharge water of a sewage treatment plant.
And my point is that just because the outgoing water is "cleaner" (e.g. gets a lower contaminantion score on whatever unspecified weighted list of contaminents the study looked at) doesn't mean you won't have problems if you "close the loop".
That doesn't mean you can't recycle water, it just means you have to be extra specially careful to make sure nasty chemicals don't build up over multiple cycles by adding extra special checks at the recycling plant and/or by ensuring that you have good mixing of recycled and virgin water.
About bloody time that some city in the US starts doing this. Did you know that the outflow from the Los Angeles sewage treatment plant is actually cleaner than the water that they pump (at ridiculous cost) over the mountains to the potable water intake?
By what definition of cleaner? How are different contaminents weighted? are there any contaminents that neither the water processing system or the wastewater processing system can remove and so would build up cycle by cycle in a closed loop system? Can those contaminents be removed with special processing? if so how does that special processing impact the economy of the process? if not what limit does this place on the proportion of water that can be safely recycled to keep overall quality acceptable?
Most serial ports these days will accept a 5V signal, so if it's actually TTL then it works.
IIRC most logic level serial is inverted compared to RS-232 (because most RS-232 level shifters are inverting) sometimes you can reconfigure the logic polarity but if your device doesn't allow that then you would need to add an inverter (at which point you may as well add a level shift chip and do it properly IMO).
Also note that while TTL ran off 5V the logic levels it used were closer to 3.3V cmos than to 5V cmos. Indeed it is pretty common to use 5V cmos devices with "TTL compatible inputs" to convert a signal from 3.3V logic to 5V logic.
Whe you say "landlines" do you mean the physical lines or the service?
The physical lines are indeed antiquated but they neeed to be maintained until such time as they can be replaced with something more modern. Remember DSL is delivered over the same lines as POTs.
The service of high quality and reliable phone calls needs to be maintained somehow. VOIP over the open internet works well most of the time but it's sufficiantly unreliable (especially with SIP behind NAT) that i'd be wary of depending on it for a buisness.
It's not a bad idea but I can't even think of what it must be like to surf at that speed. It reminds me of dial up.
Beyond a certain point page load times become dominated by latency and server side slowness not by the bandwidth of your own pipe.
Going from dialup to 512K ADSL was a huge leap but my experiance has been that beyond that there is little difference in normal web browsing (file downloads and video are another matter).
Well, as IPv4 addresses become scarce, having a load of customers on IPv6 with NAT64 to access v4 sites may be cheaper.
IIRC comcast are planning to use ds-lite instead of nat64.
With DS-lite the "customer premises equipment" encapsulates v4 the packets and sends them over v6 to a box at the ISP. The box at the ISP performs network address translation on the v4 packets and includes the clients v6 address in the translation tables so multiple customers can use the same private IPs without conflicting with each other.
This is IMO the best solution as it avoids the need for ISPs to allocate private v4 IPs to customers, avoids the mess of protocol translation and allows legacy v4 only equipment to be used behind the customer premisis equipment.
Slashdot Mirror
Always thought Nintendo kick Apple's ass if they just added GSM to the DS and given it a bluetooth headset with a dialer. DS has games, a web browser, a camera(newer ones), etc. Just doesn't have a phone.
One big problem is that afaict the DS is designed like an old fashioned game console with the games accessing the hardware directly. Also I don't think the DS has any form of multitasking. So I think the only way to build a usable DSi based smartphone would be to have a seperate processor (on top of the two processors the DS has already) for the smartphone functionality and somehow build an emulated wifi chip that passed packets from the DS hardware to the smatphone hardware.
Granted, given that the processor is just a 32-bit processor, the capacity of the HDD that could have been accessed would have been limited.
buillshit.
Working at a performance venue, We have signs we have to reprint every year. Also, more often, We have a schedule we post every month that gets a few changes. This could potentially save us Lots of money not having to continuously reprint on paper these schedules.
Firstly afaict printing on an economical laser printer costs arround 5 cents a sheet including paper but excluding electricity so at $2 per sheet and assuming the required thermal printer has no other consumables you would have to reuse this paper 20 times to make a saving. My experience with trying to feed used paper back through printers (to print on the other side) is it has to be in pristine condition or it doesn't feed properly so i'm skeptical about whether this will achive 20 cycles under realistic conditions.
Secondly introducing this into a business environment will mean duplicating your printers, thermal e-paper for the transient documents and regular paper in lasers for the permanent documents, thats a lot of extra infrastructure.
Thirdly it would be easy to inadvertantly give, throw away or destroy hundreds of dollars worth of e-paper just by printing a large document on the wrong printer and sending it to someone or accidently putting a document in the paper recycling bin when it should have gone in the e-paper bin.
It's a nice idea but I just don't see it being practical unless it gets a LOT cheaper.
Depends how long the bursts are. Afaict the norm is one burst per frame but there is no real reason the bursts couldn't be shorter than that if desired.
Because PoE gives you about 25W (up to 50W if you don't care about specs, standards and safety) at, usually, 47V.
And that is the latest "POE plus" stuff, the older POE stuff is quite a bit lower.
at, usually, 47V. Converting that down to 19V probably takes quite a bit of efficiency so you'll be lucky to get 10W.
Only if your converter is awful, 20W should be perfectly achivable if the converter doesn't suck.
Powered USB is essentially a USB connector stacked with a seperate power connector. This makes the port more than twice as tall which afaict prevents it being placed on a standard expansion slot backplate and considerablly reduces the number of ports you can put on say an ATX backplate. Worse there are three voltage variants which are all incompatible with each other.
The result of these defeciancies is that it has only found niche appeal.
such a practice could easily be deployed via Steam for example, maybe they are doing it already....
I don't think steam currently does rentals but they do free weekends and guest passes which are much the same thing technically so they could probablly bring in rental service pretty quickly if they wanted to.
The emulators I have experiance with work by running in bursts. The "emulation clock" runs much faster than real time for the duration of the burst then stop until the next burst is due so there is no real problem with the emulation clock being a weird speed relative to PC clocks.
Afaict the main problem is determining all the performance interdependencies of the real hardware and then coding those interdependencies without adding unacceptable performance pentalties to the main emulation loop.
but any of the more popular Linux distro keeps on pumping out security updates and will likely do so for years to come.
ROFL, XP may be nearing the end of it's lifecycle but should keep getting security updates until april 2014. Afaict ubuntu lucid will stop getting security updates for desktop software in april 2013.
If you want long support lifecycles on linux you really need to get into RHEL and it's rebuilds.
IMO a good language makes the safe tools painless and the unsafe ones painful. A poor language makes the safe tools painful and the unsafe ones painless.
A web orientated language designed for security for example could have multiple string types and make it easier to apply appropriate conversion processing than to convert between them without doing the processing.
Thunderbolt is not competing against USB either.
Right now if you want something better than firewire 800 or gig-e (which afaict is comparable to a single modern desktop hard drive on sequential transfers so will bottleneck an external raid array or SSD) and have a mac (other than the mac pro) you have no choice but to use thunderbolt. If you have a PC you can't use thunderbolt. So right now thunderbolt isn't in direct competition with any other interface.
But if and when thunderbolt crosses over into the PC world it will have to compete with USB3 and eSATA. It's better than them but I have my doubts as to whether it's better by a wide enough margin to displace them.
And I don't like this government a bit, but we had a vote etc.
Indeed we did but the combination of a broken system and political backroom deals meant the government we got doesn't really represent the voters. In particular the lib dems buddied up with the tories despite being traditional regarded as closer to labour.
IIRC when I tried it on XP it ran ok but you couldn't get very far on the modern web because it doesn't work with servers that use name based virtual hosting.
My experiance with microsoft in academia is they like to get us on programs that are a lot cheaper than paying for the software normally but are based on paying a subscription based on the size of the whole institution rather than paying for each individual machine.
The result is that there is no motivation to gradually migrate away from MS software since the only way to reduce the ammount paid to MS would be to virtually eliminate MS software from the institution (which is not realisitically going to happen).
There are lots of clues
User agents would be a clue (IIRC some mobile browsers spoof those but it wouldn't surprise me if there are subtule differences between the commonly spoofed strings and the common ones on desktops). There is likely other information in the http request that differs too.
If the tethering software was making the phone act as a nat router TTLs and other TCP/IP options could be a clue
If the tethering software worked by loading a proxy on the phone then an x-forwarded for header could be a clue
Large downloads would be a clue, so would very high overall data use.
As you say traffic from application that don't exist on the phone would be an especially damning clue
Afaict most mobile phone contracts have a clause allowing termination for any or no reason whatsoever so they don't exactly need absolute proof that someone is teathering to send them a "cease or pay up" letter with a threat of disconnection.
Afaict verizon is CDMA so they have to register any devices onto the network (unlike with GSM where you can drop your SIM into any device) so presumablly they would only allow such a device to be registered on an appropriate plan.
One thing i can see an automated car being bad at is choosing between accidents. Sometimes a car will be put into a situation where it will inevitablly hit something but the operator has a choice of what to hit. Say for example a motorcyclist comes off thier bike, will the automated driving system by able to disginguish between the motorcyclist and the vacant bike and be able to choose to hit the vacant motorbike rather than the motorcyclist?
To be precise, a mere logical inverter won't work, because RS232 uses both negative and positive levels
I was under the impression that despite being non compliant it would generally work because most receivers have an input threshold above zero.
The original poster's point was that often the incoming water is less sanitary than the discharge water of a sewage treatment plant.
And my point is that just because the outgoing water is "cleaner" (e.g. gets a lower contaminantion score on whatever unspecified weighted list of contaminents the study looked at) doesn't mean you won't have problems if you "close the loop".
That doesn't mean you can't recycle water, it just means you have to be extra specially careful to make sure nasty chemicals don't build up over multiple cycles by adding extra special checks at the recycling plant and/or by ensuring that you have good mixing of recycled and virgin water.
About bloody time that some city in the US starts doing this. Did you know that the outflow from the Los Angeles sewage treatment plant is actually cleaner than the water that they pump (at ridiculous cost) over the mountains to the potable water intake?
By what definition of cleaner? How are different contaminents weighted? are there any contaminents that neither the water processing system or the wastewater processing system can remove and so would build up cycle by cycle in a closed loop system? Can those contaminents be removed with special processing? if so how does that special processing impact the economy of the process? if not what limit does this place on the proportion of water that can be safely recycled to keep overall quality acceptable?
Most serial ports these days will accept a 5V signal, so if it's actually TTL then it works.
IIRC most logic level serial is inverted compared to RS-232 (because most RS-232 level shifters are inverting) sometimes you can reconfigure the logic polarity but if your device doesn't allow that then you would need to add an inverter (at which point you may as well add a level shift chip and do it properly IMO).
Also note that while TTL ran off 5V the logic levels it used were closer to 3.3V cmos than to 5V cmos. Indeed it is pretty common to use 5V cmos devices with "TTL compatible inputs" to convert a signal from 3.3V logic to 5V logic.
Whe you say "landlines" do you mean the physical lines or the service?
The physical lines are indeed antiquated but they neeed to be maintained until such time as they can be replaced with something more modern. Remember DSL is delivered over the same lines as POTs.
The service of high quality and reliable phone calls needs to be maintained somehow. VOIP over the open internet works well most of the time but it's sufficiantly unreliable (especially with SIP behind NAT) that i'd be wary of depending on it for a buisness.
It's not a bad idea but I can't even think of what it must be like to surf at that speed. It reminds me of dial up.
Beyond a certain point page load times become dominated by latency and server side slowness not by the bandwidth of your own pipe.
Going from dialup to 512K ADSL was a huge leap but my experiance has been that beyond that there is little difference in normal web browsing (file downloads and video are another matter).
Well, as IPv4 addresses become scarce, having a load of customers on IPv6 with NAT64 to access v4 sites may be cheaper.
IIRC comcast are planning to use ds-lite instead of nat64.
With DS-lite the "customer premises equipment" encapsulates v4 the packets and sends them over v6 to a box at the ISP. The box at the ISP performs network address translation on the v4 packets and includes the clients v6 address in the translation tables so multiple customers can use the same private IPs without conflicting with each other.
This is IMO the best solution as it avoids the need for ISPs to allocate private v4 IPs to customers, avoids the mess of protocol translation and allows legacy v4 only equipment to be used behind the customer premisis equipment.
OTOH if you lied and the cops found out you had lied then I would think that could put the project in even more serious jeopardy.