IEEE's Technology Winners & Losers of 2006

eldavojohn writes "As far as technologies go, there are clear winners and clear losers. This month's IEEE Spectrum issue contains an interesting list of winners and losers from 2006. Among the winners are a new radio technology, IP phone networks & memory technologies along with ethanol from sugarcane. Among the losers are tongue vision, LEDs in clothes, a flying car and ethanol from corn."

Re:Battery Life

[deimios666]

Software in this case probably means Firmware. As discussed previously firmware is a much cheaper alternative to specialised hardware. Besides it is more flexible than hardware.

Re:Brazil in the news...

[ulzeraj]

Ethanol from CORN is a bad idea. Here we make ethanol from sugarcane.

BT chosen? Look at KPN for really moving forward

[Raindeer]

I couldn't disagree more with the choice of BT as the leading company because of its 21CN network. As such it is in interesting choice of BT to go to Ethernet IP for its entire network. There are at least two other incumbents who are doing the same thing. KPN has a project called ALL-IP and and Telstra has a project called the Common Network.

However KPN is doing something more than just changing the backbone. KPN will roll-out VDSL2+ to the end-users as well. This will all be Ethernet/IP based for the backhaul and VDSL2+ for the last 450 meters, allowing 50/20mbit down/up. KPN will close 1350 swithch locations and roll out 28000 street cabinets to deliver the speeds to the end-user.
http://www.kpn.com/upload/1215076_9475_11328305981 77-1212162_9475_1132326712652-Op_weg_naar_All-IP_1 81105.pdf [kpn.com]
http://www.lightreading.com/document.asp?doc_id=69 419&print=true [lightreading.com]
(the lightreading article forgets the vdsl2+ bit, see presentation for that)

In contrast BT will only do ADSL in its network, they will not reach speeds above 24 mbit and in response to a question on access networks he says, that it is very hard to understand what a user will want to do with more than 24mbit. (hereby forgetting that most of the UK will not be living close enough to a dslam to actually get this 24mbit). He doesn't see a reason for fiber to the home or any other kind of access networks. This was said by its Chairman Ben Verwaayen at a recent Ofcom Event on convergence. http://www.ofcom.org.uk/event/presentations/sessio n6 [ofcom.org.uk] (minute 25 and onwards)

Re:Battery Life

[twiddlingbits]

Maybe, maybe not. Firmware requires an actual CPU such as an ARM or DSP to run the code. Those devices, while not ultra expensive, are not as cheap as FPGAs or ASICs. The cost in ASICs and FPGAs for the "dedicated hardware" phone is in the initial design, the cost to manaufacture is low. The cost for a "firmware phone" is also up front in design and development, and then the CPUs cost is added. For basic phones that do very little, I think the specialized hardware approach would be cheapest, for PDAs and higher end phones probably the Software driven approach is cheaper as you can get more functionality into the phone (which helps sell it at higher prices). Higher end phones and PDAs are really a combination of both approaches, there is dedicated special hardware to handle the signal processing and then OSes and Applications to provide the other features.