Because most people with VoIP on their computers use Skype.
Most people run Windows, most people store their important data in propriatary formats that won't be accessible in a couple of upgrades' time, most people pay their cellular service providers crazy amounts of money each month and think that their frequent phone upgrades are free (and even though they don't particularly want the upgrade, they may as well have it coz it's "free", right?) - just because most people do it doesn't make it a sensible thing to do.
You won't get Skype on it since Skype is completely proprietary (and why on earth would you want it?). I can't see any reason why you couldn't run a SIP user agent on it though. It runs Xorg, so you can probably run Ekiga on it.
GSM is really only of use for making calls. Using it for data is insane - you won't get more than 9.6Kb/s
Not true - GSM allows for HSCSD, which is basically bonding of up to 4 GSM channels, giving you 38.4Kbps.
Also, depending on your application, using GSM CSD instead of GPRS may be very beneficial - GPRS has really high latency, which makes interactive stuff like SSH really painful (also makes establishing SSL connections terminally slow because of the number of round-trips needed for the handshake), whereas CSD is significantly lower latency. The upshot of this is that if you need to do something like administer a server over SSH, you want CSD, not GPRS.
GPRS gives you a maximum of about 5KB/s with 2 second latency in the real world
This is _really_ variable and depends on how busy the cell you are in is. If you're using CSD then once the circuit is established you are guaranteed the bandwidth, whereas the bandwidth available on GPRS will vary. I use GPRS a lot on Orange's network, and I can tell you that it'll go from "reasonable" to "shockingly bad" (~300 bytes per second, 5-45 seconds latency, over 50% packet loss) to "no connection at all" at a moment's notice - I certainly wouldn't want to rely on it for anything important.
Also, I see very frequent dropouts within Orange's network itself (i.e. not on the air-interface), but that is down to Orange's incompetence at running an IP network rather than the technology itself.
I was wondering why these SSDs draw any power at all while idling - I guess they have to leave the controller powered up, but some of the figures quoted were around 2W when idle - you can get x86 processors that run on less than that FFS!
And why shouldn't they make you re-take it regularly to ensure driver's are still fit to drive? Most similar professions demand this.
For most people, driving isn't a profession.
If you don't break the law, you won't get fined. It's incredibly easy.
I have a number of problems with speed cameras:
1. They are seen as a replacement for patrols. This means that you only catch the people who are speeding instead of the people who are being far more dangerous (for example, tailgating, driving with no lights on, etc.) 2. There is no flexibility. There are some roads which change speed limit very frequently and it is very easy to miss a sign. If you were stopped by a patrol in a similar situation, it is likely that you would just get a warning for a completely innocent mistake rather than wilfully exceeding the limit. 3. They are often placed in areas where they *decrease* safety. I know of a number of roads where the only places where it is safe to overtake have had their speed limit reduced and cameras installed - this means that people end up overtaking in unsafe places. Yes, you could argue that the drivers who do this are stupid, but that doesn't get away from the fact that the safety of the road has been decreased as a direct result of the installation of cameras. 4. There is a significant amount of time between being caught by a speed camera and receiving the notice. This can end up in a situation where someone with a miscalibrated speedo ends up losing their licence instead of finding out that there is a problem (and getting it fixed) the first time they are pulled over. That is both unfair, and also dangerous since the driver is (unintentionally) speeding the whole time they are waiting for the first notice to drop through the door.
we have a strict, accurate, reasonable measure of a scientifically-accurate value which is prominently displayed on all roads,
It really isn't that scientifically accurate - the speed limit placed on a specific road is based on a case by case judgement. To achieve a scientifically accurate speed limit you would need to drive a large number of vehicles along each road at various speeds and record how many accidents happened at each speed for each individual road and then set that road's speed limit according to the results.
What other activity do you do that requires you to control a piece of explosive machinery
It's like saying that speed cameras are at fault because people brake heavily before them. They are not, they are exposing the problem that stupid drivers have always existed and yet nothing is done about them. You should ALREADY be at the speed limit (in fact, significantly less than, in almost all circumstances). If you have to brake heavily, the problem is YOU. YOU have created the hazard yourself. In the same way, you can't "blame" a plastic bag flying in front of your car for the accident that meant you hit someone in front, who was not a safe distance away. YOU were too close. YOU shouldn't be. YOU did not have a safe braking distance between you and the car in front. The plastic bag didn't press the throttle for you or cut your brake lines.
But the authority who is installing the speed cameras should be asking themselves why they are installing them. The possibilities that spring to mind are:
1. To make the road safer. 2. To blindly enforce the law for no reason other than it is the law. 3. To make money.
In the case of (1), if installing a speed camera increases the number of accidents then the authority has completely failed to accomplish their objective, no matter who's fault it is that the accidents are happening. There is no point in just blaming people after the accident has happened and saying it wasn't the camera's fault.
As for (2), why are we blindly enforcing laws? If people are safely exceeding the speed limit then that is surely an indication that the speed limit is set wrong, and punishing these drivers is accomplishes nothing.
You will get more functionality than the freerunner and the N800 runs Linux (bigger screen) and is very hackable and stable right now.
And you get to fill your pockets with even more devices... I gave up carrying around a separate PDA and phone a long time ago because I just have far too much crap in my pockets already. At the moment I'm using a Symbian smartphone (which is, frankly, utter crap) - I've been following the OpenMoko project for a couple of years now and I'm sorely tempted to get a Freerunner, but the lack of 3G is putting me off.
I'm not sure why you think the N800 has "more functionality" anyway - the Freerunner runs Linux and Xorg so AFAIK you can run most of your desktop software on it (so long as the software plays nice with a small screen).
Well multi-touch definitely would have been nice for hackability.
Multitouch precludes the ability to use a stylus. (Multitouch uses capacitative coupling, similar to the trackpad on a laptop, rather than presure sensitivity).
It would have been nice if they designed it with a Android upgrade in mind and designed the Freerunner with a ARMv5 chip.
Since much of Android's development has been done behind closed doors, and the Neo 1973 (which the Freerunner is based on) was available before Android was ever announced, I don't see how they could have.
basically requires a stylus, and it isnt multitouch.
They should have included a stylus holder in the phone case (as far as I know there isn't one), but I consider the lack of multitouch to be a really good thing. Multitouch is fundamentally incompatible with a stylus, and the ability to use a stylus on my phone is pretty important to me (and is one of the many many reasons I wouldn't buy an iPhone).
I've thought for quite a long time that the Neo would make a great network debugging tool - plug in a USB ethernet adaptor and you can wander around a site running network tests from your phone.
The iPhone 3G has A-GPS, or assisted GPS. It is a real, honest-to-God GPS receiver, but queries a database of known wifi hotspots and their locations in the event GPS lock can't be obtained.
A-GPS has nothing to do with wifi - it simply means that the GPS chip needs some external support in order to work (i.e. it isn't a stand-alone chip). How much external support is needed is dependent on the exact hardware - in many cases it just requires the host CPU to do a bit of work, in other cases it requires that processing is offloaded to a remote server on the cellular network.
I'm not arguing with this. I'm as surprised as anyone that his name isn't on the list - I'm just pointing out that the "maybe he wants to be neutral" arguments don't hold much water since he has been quite clear on his position previously (and it is anything but neutral).
If Moore's law stated (roughly paraphrasing) that computer performance doubles each 2 years
Except it didn't state that at all. Moore observed that the number of transistors that can be inexpensively placed on an integrated circuit doubles every two years. Equating number of transistors == performance is vastly oversimplifying things. For one thing, it doesn't even take into account changes in clock speeds.
Also, increasingly the performance of a single piece of silicon is less important, since we are offloading some processing to specialist processors (e.g. GPUs, which are many many times faster at performing the calculations required for 3D graphics than your CPU). I do wonder if one of the next steps in general purpose computing will be to include a number of FPGAs on the motherboard (or in the CPU package itself) which can be reconfigured to do specialist operations on the fly - this would massively increase the performance of some operations that are normally run on the CPU.
Attacking? For an attack to be meaningful, you'd have to invade the US and seize US oil fields. I can't see that succeeding for any nation or even continent-sized combination of countries. The US military would decimate any force before it got even close.
You're assuming that the US continues to remain as strong as it is now.
Imagine, for example, if the US found itself at war with most of the rest of the world - the US's forces may well be decimated. At that point, if another country decided they wanted the US's oil, the US would pretty much be at their mercy. (Think: Germany after the second world war).
The first of those was a pretty good example of the shit hitting the fan and nothing majorly bad happening as a result (i.e. a demonstration that fission can be pretty safe, even when things go to hell). The second was an example of why you don't use inherently unsafe reactors that wouldn't have been allowed in the West at the time anyway (not to mention: not allowing complete idiots to run them).
Let's use up the oil resources of the people who hate us while it's still relatively cheap, then tap our own resources at $300 a barrel and make them come crawling.
Actually, with digital TV mandated next year, cable TV is more and more of a rip off and excess expenditure than $9 a gallon gas.
I don't know how it works in the US, but in the UK the cost of upgrading to digital TV is a £15 DVB-T receiver or a £40 DVB-S receiver. That's about the cost of 13 litres of petrol around here.
And, when push comes to shove on the highway, we will survive while the Civic drive bites the dust.
I think that's a pretty bad attitude - "when I plough into another car, I don't care if I kill the family in the car I just hit so long as I'm not hurt."
Slashdot Mirror
Because most people with VoIP on their computers use Skype.
Most people run Windows, most people store their important data in propriatary formats that won't be accessible in a couple of upgrades' time, most people pay their cellular service providers crazy amounts of money each month and think that their frequent phone upgrades are free (and even though they don't particularly want the upgrade, they may as well have it coz it's "free", right?) - just because most people do it doesn't make it a sensible thing to do.
You won't get Skype on it since Skype is completely proprietary (and why on earth would you want it?). I can't see any reason why you couldn't run a SIP user agent on it though. It runs Xorg, so you can probably run Ekiga on it.
GSM is really only of use for making calls. Using it for data is insane - you won't get more than 9.6Kb/s
Not true - GSM allows for HSCSD, which is basically bonding of up to 4 GSM channels, giving you 38.4Kbps.
Also, depending on your application, using GSM CSD instead of GPRS may be very beneficial - GPRS has really high latency, which makes interactive stuff like SSH really painful (also makes establishing SSL connections terminally slow because of the number of round-trips needed for the handshake), whereas CSD is significantly lower latency. The upshot of this is that if you need to do something like administer a server over SSH, you want CSD, not GPRS.
GPRS gives you a maximum of about 5KB/s with 2 second latency in the real world
This is _really_ variable and depends on how busy the cell you are in is. If you're using CSD then once the circuit is established you are guaranteed the bandwidth, whereas the bandwidth available on GPRS will vary. I use GPRS a lot on Orange's network, and I can tell you that it'll go from "reasonable" to "shockingly bad" (~300 bytes per second, 5-45 seconds latency, over 50% packet loss) to "no connection at all" at a moment's notice - I certainly wouldn't want to rely on it for anything important.
Also, I see very frequent dropouts within Orange's network itself (i.e. not on the air-interface), but that is down to Orange's incompetence at running an IP network rather than the technology itself.
I was wondering why these SSDs draw any power at all while idling - I guess they have to leave the controller powered up, but some of the figures quoted were around 2W when idle - you can get x86 processors that run on less than that FFS!
And why shouldn't they make you re-take it regularly to ensure driver's are still fit to drive? Most similar professions demand this.
For most people, driving isn't a profession.
If you don't break the law, you won't get fined. It's incredibly easy.
I have a number of problems with speed cameras:
1. They are seen as a replacement for patrols. This means that you only catch the people who are speeding instead of the people who are being far more dangerous (for example, tailgating, driving with no lights on, etc.)
2. There is no flexibility. There are some roads which change speed limit very frequently and it is very easy to miss a sign. If you were stopped by a patrol in a similar situation, it is likely that you would just get a warning for a completely innocent mistake rather than wilfully exceeding the limit.
3. They are often placed in areas where they *decrease* safety. I know of a number of roads where the only places where it is safe to overtake have had their speed limit reduced and cameras installed - this means that people end up overtaking in unsafe places. Yes, you could argue that the drivers who do this are stupid, but that doesn't get away from the fact that the safety of the road has been decreased as a direct result of the installation of cameras.
4. There is a significant amount of time between being caught by a speed camera and receiving the notice. This can end up in a situation where someone with a miscalibrated speedo ends up losing their licence instead of finding out that there is a problem (and getting it fixed) the first time they are pulled over. That is both unfair, and also dangerous since the driver is (unintentionally) speeding the whole time they are waiting for the first notice to drop through the door.
we have a strict, accurate, reasonable measure of a scientifically-accurate value which is prominently displayed on all roads,
It really isn't that scientifically accurate - the speed limit placed on a specific road is based on a case by case judgement. To achieve a scientifically accurate speed limit you would need to drive a large number of vehicles along each road at various speeds and record how many accidents happened at each speed for each individual road and then set that road's speed limit according to the results.
What other activity do you do that requires you to control a piece of explosive machinery
Explosive? WTF sort of car are you driving?!
It's like saying that speed cameras are at fault because people brake heavily before them. They are not, they are exposing the problem that stupid drivers have always existed and yet nothing is done about them. You should ALREADY be at the speed limit (in fact, significantly less than, in almost all circumstances). If you have to brake heavily, the problem is YOU. YOU have created the hazard yourself. In the same way, you can't "blame" a plastic bag flying in front of your car for the accident that meant you hit someone in front, who was not a safe distance away. YOU were too close. YOU shouldn't be. YOU did not have a safe braking distance between you and the car in front. The plastic bag didn't press the throttle for you or cut your brake lines.
But the authority who is installing the speed cameras should be asking themselves why they are installing them. The possibilities that spring to mind are:
1. To make the road safer.
2. To blindly enforce the law for no reason other than it is the law.
3. To make money.
In the case of (1), if installing a speed camera increases the number of accidents then the authority has completely failed to accomplish their objective, no matter who's fault it is that the accidents are happening. There is no point in just blaming people after the accident has happened and saying it wasn't the camera's fault.
As for (2), why are we blindly enforcing laws? If people are safely exceeding the speed limit then that is surely an indication that the speed limit is set wrong, and punishing these drivers is accomplishes nothing.
And of course, no one admits to (3).
You will get more functionality than the freerunner and the N800 runs Linux (bigger screen) and is very hackable and stable right now.
And you get to fill your pockets with even more devices... I gave up carrying around a separate PDA and phone a long time ago because I just have far too much crap in my pockets already. At the moment I'm using a Symbian smartphone (which is, frankly, utter crap) - I've been following the OpenMoko project for a couple of years now and I'm sorely tempted to get a Freerunner, but the lack of 3G is putting me off.
I'm not sure why you think the N800 has "more functionality" anyway - the Freerunner runs Linux and Xorg so AFAIK you can run most of your desktop software on it (so long as the software plays nice with a small screen).
Well multi-touch definitely would have been nice for hackability.
Multitouch precludes the ability to use a stylus. (Multitouch uses capacitative coupling, similar to the trackpad on a laptop, rather than presure sensitivity).
It would have been nice if they designed it with a Android upgrade in mind and designed the Freerunner with a ARMv5 chip.
Since much of Android's development has been done behind closed doors, and the Neo 1973 (which the Freerunner is based on) was available before Android was ever announced, I don't see how they could have.
basically requires a stylus, and it isnt multitouch.
They should have included a stylus holder in the phone case (as far as I know there isn't one), but I consider the lack of multitouch to be a really good thing. Multitouch is fundamentally incompatible with a stylus, and the ability to use a stylus on my phone is pretty important to me (and is one of the many many reasons I wouldn't buy an iPhone).
it costs twice what this year's iPhone does
How do you figure that out?
The new 8GB iPhone costs £639 (£99 + £30 / month for 18 months: http://www.o2.co.uk/iphone/paymonthly), whilst the Freerunner will cost about £272 (up-front cost, no contract: https://www.truebox.co.uk/trueboxportal/index.php?wk=Openmoko).
I've thought for quite a long time that the Neo would make a great network debugging tool - plug in a USB ethernet adaptor and you can wander around a site running network tests from your phone.
The iPhone 3G has A-GPS, or assisted GPS. It is a real, honest-to-God GPS receiver, but queries a database of known wifi hotspots and their locations in the event GPS lock can't be obtained.
A-GPS has nothing to do with wifi - it simply means that the GPS chip needs some external support in order to work (i.e. it isn't a stand-alone chip). How much external support is needed is dependent on the exact hardware - in many cases it just requires the host CPU to do a bit of work, in other cases it requires that processing is offloaded to a remote server on the cellular network.
His wife is tubgirl?
I'm not arguing with this. I'm as surprised as anyone that his name isn't on the list - I'm just pointing out that the "maybe he wants to be neutral" arguments don't hold much water since he has been quite clear on his position previously (and it is anything but neutral).
Linus has made his opinions on the matter pretty clear in the past...
If Moore's law stated (roughly paraphrasing) that computer performance doubles each 2 years
Except it didn't state that at all. Moore observed that the number of transistors that can be inexpensively placed on an integrated circuit doubles every two years. Equating number of transistors == performance is vastly oversimplifying things. For one thing, it doesn't even take into account changes in clock speeds.
Also, increasingly the performance of a single piece of silicon is less important, since we are offloading some processing to specialist processors (e.g. GPUs, which are many many times faster at performing the calculations required for 3D graphics than your CPU). I do wonder if one of the next steps in general purpose computing will be to include a number of FPGAs on the motherboard (or in the CPU package itself) which can be reconfigured to do specialist operations on the fly - this would massively increase the performance of some operations that are normally run on the CPU.
Same with dates: MM/DD/YYYY, DD-MM-YYYY, YYYY.MM.DD, so on and so on.
There are real problems with date formats - how do you tell the difference between them unless someone explicitly tells you?
i.e. What date is 05/03/2008 (or worse - 05/03/08)?
I can't imagine such a conflict staying non-nuclear. And if it doesn't, oil will be the least of the world's worries.
How about civil war?
Attacking? For an attack to be meaningful, you'd have to invade the US and seize US oil fields. I can't see that succeeding for any nation or even continent-sized combination of countries. The US military would decimate any force before it got even close.
You're assuming that the US continues to remain as strong as it is now.
Imagine, for example, if the US found itself at war with most of the rest of the world - the US's forces may well be decimated. At that point, if another country decided they wanted the US's oil, the US would pretty much be at their mercy. (Think: Germany after the second world war).
The first of those was a pretty good example of the shit hitting the fan and nothing majorly bad happening as a result (i.e. a demonstration that fission can be pretty safe, even when things go to hell). The second was an example of why you don't use inherently unsafe reactors that wouldn't have been allowed in the West at the time anyway (not to mention: not allowing complete idiots to run them).
Let's use up the oil resources of the people who hate us while it's still relatively cheap, then tap our own resources at $300 a barrel and make them come crawling.
s/crawling/attacking/
There's plenty of U238 around anyway (thousands of years' worth) - it is only U235 which *may* be in short supply.
Actually, with digital TV mandated next year, cable TV is more and more of a rip off and excess expenditure than $9 a gallon gas.
I don't know how it works in the US, but in the UK the cost of upgrading to digital TV is a £15 DVB-T receiver or a £40 DVB-S receiver. That's about the cost of 13 litres of petrol around here.
And, when push comes to shove on the highway, we will survive while the Civic drive bites the dust.
I think that's a pretty bad attitude - "when I plough into another car, I don't care if I kill the family in the car I just hit so long as I'm not hurt."
Think of the children.
:)
Isn't that what people are getting arrested for?