EDGE will offer about 3x the data rates of GPRS whilst still using similar technology. So, it would be around the 350 kbps mark in (very) good radio conditions. 3G technology (in the current delayed timeframe) is looking less revolutionary and more evolutionary all the time.
...people will write them down.
Preferrably on post-it notes and stuck to the keyboard or the screen.
But that's not always a problem. In some situations, where outsiders don't wander round offices, this can be a good technique. If the office is "secure", writing down passwords is fine. This can certainly be put to good effect in the home.
Post-its stuck to monitors might not be the best place to write them down, I grant you.
DCS1800 is the only band offered by two out of the four UK networks (one2one and Orange). The other two (Vodafone and Cellnet) offer only GSM900. So, the 900/1900 could be used in the UK, but you'd have to choose the right operator. As another poster said, it would be far better if this device was triband - 900/1800/1900.
The RIM uses a different carrier and technology than GPRS, so it's not GPRS either but it works very similar to GPRS
In the UK, the Blackberry runs off GPRS on the BT Cellnet network. I imagine the US situation is different. Most (all?) the UK operators offer GPRS now. I think the Treo is likely to require some form of software upgrade before it'll do GPRS. However, if the US networks don't support it yet then launching without GPRS capability isn't as bad as I thought.
Hmm. It's a bit strange - most dual band phones support 900/1800, but this one appears to be 900/1900. That cuts out quite a large proportion of European users. It makes no mention of data rates either, so I assume it doesn't support HSCSD. That limits it to 14.4 kbps, which is pretty restrictive for a "Communicator" device.
It says here "GPRS upgradeable upon availability" - that rather implies it's not GPRS (General Packet Radio Service) at launch. That makes this rather less desirable - it's not "always on", and you're charged by time rather than by data. The fact it says "upon availability" makes this sound like a rushed out product. Compare this with the RIM Blackberry, which has been out for ages and has GPRS.
Now, instead of the 1000 minutes lasting all month your laptop eats them up in two days. That means you have to buy more minutes from the cell networks.
True, except this would almost certainly be charged as GPRS or some other packet service (EDGE, 3G, etc). So, it would be charged by the amount of data you used, not the amount of time you were connected to the network.
Um.. things are moving on. Take a look at this demo - it's a DirectX phone emulator with a game. The technology is not far from market, and is catching up with handheld gaming.
HTTP headers exist for explicitly requesting access to a resource in a certain language (though I dont know of any site that actually makes use of them)
One quick example: the Debian home page. I never really notice this (being an en-gb user), but it does work.
Just to make my comment above (about accuracy) look bloody stupid, this site gets me bang on. However, the other two I tried were both wrong - got the UK correctly, but the wrong end of the country.
That's a pretty big pin. Pinpoint an IP address maybe, but that doesn't tell you much about where someone really is. Ignoring the effect of proxies, some dynamic address allocation schemes can cover huge areas.
I think the more "Big Brother" aspects of this can probably be ignored for a while - until ISPs start getting more involved with content providers at least.
Sounds like you're confusing the active badge with the active bat. The active badge is entirely IR, but the active bat (which the article is mostly about) uses radio to the device and ultrasound back. It uses ultrasound because it's slow enough to get an accurate path difference from the time delay.
I'm currently taking a lecture course given by Prof. Andy Hopper (director of AT&T Cambridge Labs). The Active Badge was the original system, based on infrared. That just localised to a single room or area. The Active Bat is the ultrasound version, and if you put three on a single object you can tell its position and orientation to a pretty fine resolution. It's not just for people though - they've got them on computers, chairs, and pretty much everything in the office.
Don't know why this has suddenly appeared on/. though - it's been around quite a few years already.
The article is wrong - the mobile devices don't run at 1.6v. If you look at this page from AMD it's 1.4v. Assuming this is the same core rather than the upcoming Morgan core, then the power consumption should just be scaled down by the square of the suppy voltage. 21 W for the 600 MHz and 24 W for the 700 MHz, based on the power consumptions for the desktop versions.
If anyone else is after an ntl cable modem connection (how many people in Cambridge/East Anglia read this), as long as you've already got cable tv working it's not a big problem. A couple of friends of mine just bought a cable modem and connected it up - worked without a problem. For once, ntl's billing system actually worked, and it picked up they were using it.
Nice idea, and it's already used on a small scale. Rooms would be too big, as the beams would lose any sort of coherence or focus in that distance.
The use of optical frequencies is already being used for routing in lab systems, and possibly in commercial systems. This gives you an entirely optical network - very fast, with none of this crappy electrical stuff.
Most of the switching in current lab systems is done with MEMS mirrors, but there are other possibilities with arrayed waveguide gratings and lcd holograms that look promising.
The little bit of smoke isn't going to work I'm afraid - it's all IR....
A single span without amplification can currently reach about 30 km at 40 Gbps in the lab. When you add optical amplification, things start to get rather better, and transatlantic single span fibres have existed for a few years now. Couple this with the fact that a single optical amplifier can amplify over 100 40 Gbps channels at once when WDM (wavelength division multiplexing) is used, and you have quite ridiculous data rates.
I imagine the comment in the article about only.5% of the available bandwidth being used is for short span WDM systems, without optical amplifiers.
Gnutella obviously hasn't been tested under high stress - a decentralised system is always going to be hard to test realistically. Part of the disadvantage of being decentralised is there's no central server to upgrade when things slow down.
Slashdot Mirror
EDGE will offer about 3x the data rates of GPRS whilst still using similar technology. So, it would be around the 350 kbps mark in (very) good radio conditions. 3G technology (in the current delayed timeframe) is looking less revolutionary and more evolutionary all the time.
I'm a bit confused, though. Isn't this what GPRS is supposed to accomplish?
WiFi runs a lot faster than GPRS. Eight slot GPRS will give you about 115.2 kbps, whereas WiFi gives you ~10 Mbps. Both are best case figures.
But that's not always a problem. In some situations, where outsiders don't wander round offices, this can be a good technique. If the office is "secure", writing down passwords is fine. This can certainly be put to good effect in the home.
Post-its stuck to monitors might not be the best place to write them down, I grant you.
In the UK, the fastest commercialy available broadband is ADSL - 512 down, 256 up.
Not true any more - ntl now offer a 1mbit service for £50/month.
The main problem with ntl is wondering whether they'll still be around in a month, now they've gone for Chapter 11.
Hard disks are sold in Gigabytes, using the powers of 10 system. An 80GB hard disk is 80,000,000,000 bytes. This has been the case for a long time.
DCS1800 is the only band offered by two out of the four UK networks (one2one and Orange). The other two (Vodafone and Cellnet) offer only GSM900. So, the 900/1900 could be used in the UK, but you'd have to choose the right operator. As another poster said, it would be far better if this device was triband - 900/1800/1900.
The RIM uses a different carrier and technology than GPRS, so it's not GPRS either but it works very similar to GPRS
In the UK, the Blackberry runs off GPRS on the BT Cellnet network. I imagine the US situation is different. Most (all?) the UK operators offer GPRS now. I think the Treo is likely to require some form of software upgrade before it'll do GPRS. However, if the US networks don't support it yet then launching without GPRS capability isn't as bad as I thought.
Hmm. It's a bit strange - most dual band phones support 900/1800, but this one appears to be 900/1900. That cuts out quite a large proportion of European users. It makes no mention of data rates either, so I assume it doesn't support HSCSD. That limits it to 14.4 kbps, which is pretty restrictive for a "Communicator" device.
It says here "GPRS upgradeable upon availability" - that rather implies it's not GPRS (General Packet Radio Service) at launch. That makes this rather less desirable - it's not "always on", and you're charged by time rather than by data. The fact it says "upon availability" makes this sound like a rushed out product. Compare this with the RIM Blackberry, which has been out for ages and has GPRS.
Now, instead of the 1000 minutes lasting all month your laptop eats them up in two days. That means you have to buy more minutes from the cell networks.
True, except this would almost certainly be charged as GPRS or some other packet service (EDGE, 3G, etc). So, it would be charged by the amount of data you used, not the amount of time you were connected to the network.
Um.. things are moving on. Take a look at this demo - it's a DirectX phone emulator with a game. The technology is not far from market, and is catching up with handheld gaming.
HTTP headers exist for explicitly requesting access to a resource in a certain language (though I dont know of any site that actually makes use of them)
One quick example: the Debian home page. I never really notice this (being an en-gb user), but it does work.
Just to make my comment above (about accuracy) look bloody stupid, this site gets me bang on. However, the other two I tried were both wrong - got the UK correctly, but the wrong end of the country.
...some technologies can pinpoint one's location.
That's a pretty big pin. Pinpoint an IP address maybe, but that doesn't tell you much about where someone really is. Ignoring the effect of proxies, some dynamic address allocation schemes can cover huge areas.
I think the more "Big Brother" aspects of this can probably be ignored for a while - until ISPs start getting more involved with content providers at least.
Sounds like you're confusing the active badge with the active bat. The active badge is entirely IR, but the active bat (which the article is mostly about) uses radio to the device and ultrasound back. It uses ultrasound because it's slow enough to get an accurate path difference from the time delay.
I'm currently taking a lecture course given by Prof. Andy Hopper (director of AT&T Cambridge Labs). The Active Badge was the original system, based on infrared. That just localised to a single room or area. The Active Bat is the ultrasound version, and if you put three on a single object you can tell its position and orientation to a pretty fine resolution. It's not just for people though - they've got them on computers, chairs, and pretty much everything in the office.
/. though - it's been around quite a few years already.
Don't know why this has suddenly appeared on
The coffee pot went missing quite a while ago. Documented at http://ban.joh.cam.ac.uk/~ajp38/science/trojan/Cof fee.htm
The article is wrong - the mobile devices don't run at 1.6v. If you look at this page from AMD it's 1.4v. Assuming this is the same core rather than the upcoming Morgan core, then the power consumption should just be scaled down by the square of the suppy voltage. 21 W for the 600 MHz and 24 W for the 700 MHz, based on the power consumptions for the desktop versions.
> The only downside is its limited to 512kbps downstream.
Hardly a downside - the BT ADSL "service" is also at 512kbps. At least, that's about all I can find out from the BT ADSL website.
What's this all about? Local loop unbundling is happening in the UK, but very slowly. Is there an EU directive pushing it forward?
If anyone else is after an ntl cable modem connection (how many people in Cambridge/East Anglia read this), as long as you've already got cable tv working it's not a big problem. A couple of friends of mine just bought a cable modem and connected it up - worked without a problem. For once, ntl's billing system actually worked, and it picked up they were using it.
Nice idea, and it's already used on a small scale. Rooms would be too big, as the beams would lose any sort of coherence or focus in that distance.
The use of optical frequencies is already being used for routing in lab systems, and possibly in commercial systems. This gives you an entirely optical network - very fast, with none of this crappy electrical stuff.
Most of the switching in current lab systems is done with MEMS mirrors, but there are other possibilities with arrayed waveguide gratings and lcd holograms that look promising.
The little bit of smoke isn't going to work I'm afraid - it's all IR....
A single span without amplification can currently reach about 30 km at 40 Gbps in the lab. When you add optical amplification, things start to get rather better, and transatlantic single span fibres have existed for a few years now. Couple this with the fact that a single optical amplifier can amplify over 100 40 Gbps channels at once when WDM (wavelength division multiplexing) is used, and you have quite ridiculous data rates.
.5% of the available bandwidth being used is for short span WDM systems, without optical amplifiers.
I imagine the comment in the article about only
Gnutella obviously hasn't been tested under high stress - a decentralised system is always going to be hard to test realistically. Part of the disadvantage of being decentralised is there's no central server to upgrade when things slow down.
I'm sure it'll evolve. It's open after all.
Since plastic is EVERYWHERE right now
That's rather a short term approach isn't it? I suppose the devices themselves could be recycled more simply than current display technologies though.