Transcript To Infinity and Beyond - Archive

To Infinity & Beyond!
Peter Cochrane
ConceptLabs CA Co-Founder
Personal Communication
Wireless communication
New Company Value
Faster & Faster Wealth Generation
US$Bn
100
90
80
70
60
50
40
30
20
10
0
Internet
Mobile
1
2
3
4
5
6
Year
7
PC
8
9
10
Manufacture Distribution
R
E
T
A
I
L
Recycle
Raw
Materials
Ecosystem
Trash
Trash
Consumer
Consumer
Raw
Materials
Recycle
B-B Manufacture
B-BDistribution
B-B
Ecosystem
Trash
Trash
C-B
C-C
Consumer
Consumer
R
E
T
A
I
L
achine
B-C
obile
2G - A field of dreams…
Build it and they will come…
…& 750M did, just
for Voice & SMS
Communicating on the move…
We value mobility above Quality
We value mobility above Price
Mobility is Personal
Text messaging is about Time Shifting
Data & text are more important than voice
Bandwidth is now the lubricant of our society
UK SMS Messages 99
300
250
200
Millions
150
100
50
0
Jan Feb MarAprMayJun Jul AugSep Oct NovDec
UK SMS Messages
3.5Bn
Year 2
0.3Bn
Year 1
UK SMS Messages
12Bn
Year 3
3.5Bn
Year 2
Year 1
3G - A field of dreams…
Build it and will they come…again?
Wire line quality
Seamless roaming
MP3 Sound Files
MPEG Movie Files
JPEG Digital Photos
On Line Games
IP Telephony
IP Video Conference
IP Radio & TV
GPS + TeleCare
+++New Markets
Why did WAP fail?
Over-hyped
Overdue
Under performed
-Slow
-Limited web access
-Nothing like a PC
Too complex to engineer
Too complex to use
Performance
Disruptive Technologies
3G
>>
2G
Time
13
Powerful mobile devices…
Machines
0.5 year
M-Media
1year
2 years
4 years
5 years
Java
Internet
Messaging
Voice
88 89 90 91 92 93 94 95 96
97 98 99 00 01 02 03 04 05
More
Mobile than fixed
Messages the voice
Net purchases than a PC
A
small screen
is not a
BIG SCREEN
Will 3G happen in the EU?
> 60Bn
> 60Bn
> 60Bn
>??Bn
= Licence fees
= Infrastructure
= Development
= Marketing & Sales
>200Bn = Total
Will 3G happen in the EU?
~ £400 per head of population
> 10 years to break even
> 18 months to first chip sets
> 24 months to first sales
> 48 months worst case
Mounting public pressure & changing
government position on masts & towers
Operator Options?
Merge companies, networks and
operations to offset costs
Load cost onto 2G service
- More expensive handsets
- More expensive calls
- Reduce Quality of Service
- Reduce customers services
Operator Options?
Merge companies, networks and
operations to offset costs
Load cost onto 2G service
- More expensive handsets
- More expensive calls
- Reduce Quality of Service
- Reduce customers services
Government Options?
Reset the clock
Give the money back
A once vibrant sector has been raped
and may never be the same again
Banks may begin to foreclose on
Mounting debts
The Only Realistic Option?
Use 3G to wipe out the last/first mile
Network Types
Copper, Fibre, Radio, Optical - Static - Mobile
Chip
to
Chip
Board
LANs
to Backplanes
WANs
Board
MANs
Spatial Density
1mm
10cm
1cm
Spectral Density
10m
1m
Long
Haul
100m
1km
1000km
10km 10000km
Everything is moving toward IP based operations
3G Threats & Limitations??
Technology options frozen too early
- Not IP based
- CDMA is band-limited
- 2.5G can deliver almost as much
- 4G coming faster than expected
- Parasitic networks are far superior
Quantity/Unit Time
100
Yearly Productivity
80
70
60
40
20
0
BW
Chip
SW
350% pa
The key constraint?
Carrier’s
the
bottleneck
150% pa
35% pa
Wireless WWW
Content / Apps
Infrastructure
Devices
Extrapolation from 1950s
Clock HD
1T
RAM
1G
Backbone
1M
Last/First Mile
1K
1
1947
1957
1967
1977
1987
1997
2007
Office
&
Home
Network
Always On
MStream
100Mbit/s
LAN
11Mbit/s
WLAN
2Mbit/s
Cable
ADSL
Dial Up
PSTN
Mobile
0.5Mbit/s 56kbit/s
256kbit/s 9.6kbit/s
Hard
Drives
100,000
MByte
Office
&
Clock
1000
MHz
Home
LAN
100
Mbit/s
WLAN
11Mbit/s
Network
MStream
2Mbit/s
Cable
ADSL
PSTN
Mobile
0.5Mbit/s 56kbit/s
256kbit/s 9.6kbit/s
Net Access
Bn
1.6
1.4
1.2
Mobile
phones
Mobile
Phones
With web
access
Wired phones
1.0
0.8
0.6
0.4
0.2
0.0
PCs
96
97
98
99
00
01
02
03
04
If Telco's sold cars…
Will the Telco/Cable Co’s Deliver?
NO
They just don’t get it!
They just don’t have it!
They don’t want to do it!
They don’t have to do it!
They can’t do it anyway!!
Wrong infrastructure
Wrong mindset
Mounting debts
Have we been here before?
YES
Have we been here before?
YES
Community Antenna TV
ABC don’t do terminals
ABC don’t do support
ABC don’t do delivery
ABC don’t do networks
Parasitic Nets - 802.11 et al
Who needs a Telco or Cableco?
PARASITIC NETWORKS
Zero infrastructure!
PARASITIC NETWORKS
Zero infrastructure!
PARASITIC NETWORKS
Zero infrastructure!
PARASITIC NETWORKS
With mobile base stations!
PARASITIC NETWORKS
New modes!
As today
Device
to
Device
Broadcast
Device
Repeater
Device
Wireless - the story so far…
3000GHz
0.1mm
300GHz
Microwave
Antenna
30GHz
Experimental Comm
Satellite
3GHz
Military
Satellite
Colour
TV
30 MHz
1mm
1cm
Low-Earth
Orbit Satellite
Marine Satellite
Cellular
Phone
300MHz
Black & White
TV
BroadBand
Satellite
10 cm
1m
WalkieTalkie
10 m
Modern
Shortwave
Short wave radio
1930
1940
1950
1960
1970
1980
1990
100m
2000
TV
Mobile
Opportunity
FM
Optical
Wireless Technologies
Radio
Optical
Unregulated
Regulated
Very secure
Secure
Short range
Medium-long range
Broadband
Narrowband
Immature
Mature
No radiation risk
Radiation risk?
Spectrum
> 50,000GHz
< 300GHz
Diffuse Optical
Network
Optical
transceiver
Broadband
Mobile
Secure
Cell
TelCo/CableCo Defence
Mobile
ISP
X
Switch
Latency
• Network latency should be dominated by
the speed-of-light delay
• Managed by tight integration of optical
and electronic technologies, architectures,
operating systems and programming
Minimising Latency
• provide abundant network bandwidth
• strip out layers and protocols
• minimise network crossings
- ideal: each ‘expedition’=1 crossing + return
• introduce ‘information geography’ (location
and distance of information) as the key
parameter in network architecture, operating
systems and programming
processor
memory
processor
memory
processor
memory
Current architecture
processor
memory
Future-matched architecture
Telephone Network
Random Events + Hierarchy
=> Time + Frequency Aggregation
=> Economy
=> Blocking/Freeze Out
Control - Total - Prescriptive
=> Fixed Bandwidth
=> Low Flexibility
=> Extremely Reliable
=> Reasonably Resilient
=> Fundamentally Unsuited to Data
Internet
Bursty Events + No Hierarchy
=> Time Aggregation
=> Economy
=> Variable Latency
Control - Little
=> Variable Bandwidth
=> High Flexibility
=> Flaky
=> Infinitely Resilient
=> Fundamentally Unsuited to Speech
Parasitic Nets
Bursty Events + No Hierarchy
=> Time + Space Aggregation
=> Economy
=> Huge Latency
Control - None
=> Variable Bandwidth
=> Infinite Flexibility
=> Highly Reliable
=> Highly Resilient
=> Fundamentally Unsuited to Real Time
Established Wisdom
Hardware
Software
Network
Applications
Convergence??
I don’t think so….
Planet wide universal access??
Thank you,
cochrane.org.uk
conceptlabs.net