Assessing the Business Models Vertical Multiplay Stovepipes vs. the Horizontal Layered Model Seizing

1
Assessing the Business ModelsVertical Multiplay
Stovepipesvs.the Horizontal Layered
ModelSeizing the VoIP OpportunitiesRome,
October 5th, 2005

• Richard Stastny, ÖFEG

The opinions expressed here may or may not be
that of my company
2
Questions (I will try) to Answer

• What is the Impact of Voice and Everything over
  IP to the Telecom Industry?
• What is the future of Telcos?
• Providing Services in a multiplay vertical
  stovepipe e.g. the IMS NGN Model?
• Broadband access providers only in a horizontal
  layered model enabling third party service
  providers
• A combination of both?
• Are new approaches needed?
• What is the role of Viral and Peer 2 Peer
  Networks?
• Trusted User Identities by Telcos or by Third
  Parties?

3
Assumptions

• (End-)users want connectivity via any device,
  anywhere, at anytime
• Global mobility
• Personal Communication
• Which leads to
• the Portable Internet
• fixed-mobile convergence (FMC)
• multi-play (everything over IP)

4
Some Clarifications first

• What are stovepipes?
• from a regulatory perspective
• from a technical perspective
• What is vertical vs horizontal?
• What are Peer-to-Peer (P2P) and Viral Networks?
• What means multiplay?
• What means Everything over IP (EoIP)?

5
Regulatory Stovepipese.g. US Regulation
Currently services (and its underlying
technologies) are regulated in these vertical
Stove Pipes
Title I
Title II
Title III
Title III
Title VI
DATA
VOICE
VOICE
VIDEO
AUDIO / VIDEO
Broadcast Radio/TV
InfoServices
Wireline Telephony
Wireless Telephony
CableTelevision
6
A Change of Perspective
Cable
Traditional View
Broadcast
Telephony
Data
Wireless
Kevin Werbach
7
A Change of Perspective
Cable
Current View
Data
Broadcast
Telephony
Wireless
Kevin Werbach
8
A Change of Perspective
Content
Data
Reality
Web
Devices
Apps
Voice
Users
Video
Email
File Transfer
Cable
Wireless
Satellite
Fiber
Kevin Werbach
9
? Telcos are vertically integrated
Regulatory boundaries
T E LCO
T E LCO
T E LCO
T E LCO
Services
Transport
TISPAN3GPP ATISNGN
TISPAN3GPP ATISNGN
Access
10
Internet horizontal layering
Regulatory boundaries?
...
Services
SIP
MAIL
IM
WEB
GlobalInternet
Transport
P2P
...
xDSLCableFTTH
PSTN ISDN
GSMUMTS
Access
W-LAN
Viral or Meshed Networks
11
What about regulation?

• New approaches are needed
• e.g. US Digital Age Communications Act (DACA)
• Proposals exist e.g. from the DACA Regulatory
  Framework Working Group
• Discussed Models
• the Antitrust Model
• the Railroad Model
• the Layers Model
• the IP-Migration Model
• the EU-Model
• the FTC Act Model
• Proposal of the Regulatory Framework Working
  Group, Release 1.0
• New US Committee on Commerce and Energy Staff
  Draft on Broadband Legislation or Broadband
  Internet Transmission Services (BITS)

12
What is Peer 2 Peer?

• A peer-to-peer (or P2P) computer network relies
  on the computing power and bandwidth of the
  participants rather than concentrating it in a
  relatively few servers.
• P2P networks are typically used for connecting
  nodes via ad hoc connections.
• Such networks are useful for many purposes
• Sharing content files containing audio, video,
  data or anything in digital format is very
  common,
• and realtime data, such as voice, video, IM,
  Chat, etc. traffic, is also passed using P2P
  technology.
• A pure peer-to-peer network does not have the
  notion of clients or servers, but only equal peer
  nodes that simultaneously function as both
  "clients" and "servers" to the other nodes on the
  network.
• This model of network arrangement differs from
  the client-server model where communication is
  usually to and from a central server.
• Some networks and channels use a client-server
  structure for some tasks (e.g., searching) and a
  peer-to-peer structure for others.
• There exist many different flavors and algorithms

13
Viral communications, organic networks, meshed
networks

• Viral Communications Media Laboratory Research
  (Andrew Lippman, David P.Reed, A. Pentland)
• Viral communications derives directly from the
  end-to-end principle on which the Internet is
  based the intelligence is in the end nodes, the
  network itself maintaining as little state as
  possible.
• Communications are poised to become personal,
  embedded features of the world around us. New
  technologies allow us to make wired and wireless
  devices that are ad hoc, incrementally installed
  and populous almost without limit.
• They need no backbone or infrastructure in order
  to work instead, they use neighbors to
  bootstrap both bit delivery and geolocation.
• This re-distributes ownership of communications
  from a vertically integrated provider to the
  end-user or end-device and segregates bit
  delivery from services.
• Communications can become something you do rather
  than something you buy.

14
What is Multiplay?

• All types of communication
• TV video and audio broadcast
• fixed voice communication (telephony)
• mobile voice communication
• other real-time communications (video,
  conferencing, instant messaging, presence,
  location-based services, white-board, )
• other data transfer and access, telecontrol, etc.
  
• integrated in one access and one device
• cable companies add telephony (VoIP) and Internet
  access (need to integrate)
• mobile operators add video and web-access
• Telcos add DSL Internet Access, move to VoIP and
  add TV-Broadcast
• ? EoIP IPoE and OnePhone

15
What is Convergence?

• Convergence is a very abused term
• Voice data convergence
• Backbone people say this is done since more then
  20 years
• Fixed mobile convergence (FMC)
• IMS over Fixed Access (TISPAN, ATIS, )
• OnePhone
• GSM and WiFi access on one device
• Seamless handover, one number
• Application convergence
• e.g. using Outlook to establish calls, see
  presence information in MS Word, etc.
• The convergence is with the customer

16
The hourglass model
Everything over IP
IP to and from Everywhere
IP over Everything
17
The Internet Model
Tim Denton
18
The Internet view on the network is

• the Internet is the network. The next generation
  network (NGN) is IPv6,
• the Internet is transparent e2e or just "dumb"
  it is application unaware,
• user consent and control resides in the
  endpoints,
• service availability is what matters to users and
  not QoS. QoS is good as long as network
  congestion is avoided and if so, voice quality is
  an endpoint capability.
• the Internet is the result of a continuous
  evolution and the architecture changes constantly
  over time.

19
The ITU-T IMS NGN model
Tim Denton
20
The ITU-T view on the network is

• the NGN will be derived from the PSTN but using
  IP technology the IP Multimedia System (IMS),
• the NGN is application aware,
• control resides in the network,
• the NGN has ample QoS definitions and guarantees
  for the network service.
• all ITU-T NG networks, such as ISDN, BISDN/ATM,
  IMS/NGN/IP are based on grand designs and are not
  based on a continuous evolution. The changes from
  TDM to ATM to IP are significant discontinuities
  in the ITU-T architectures.

21
VoIP over the Internet

• The Internet is (or is intended to be) a network
  without central intelligence The stupid
  network (David S. Isenberg)
• The Internet is based on the end-to-end principle
• Every user may reach any other user via the IP
  address
• All services may be offered anywhere and may be
  accessed from everywhere
• This is of course also valid for voice and other
  communication services
• VoIP is not necessarily a service, it is an
  application - so no service must be provided
• Jon Peterson at the ITU-T/IETF NGN workshop in
  Geneva May 2005
• If there is no service provided, you do not need
  a service provider either.

22
How does VoIP work?

• What is the difference between the IETF approach
  for SIP and the NGN (IMS) approach?

23
The IETF SIP Trapezoid
DNS Server
Location Server
DNS
SIP
Inbound Proxy Server
Outbound Proxy Server
SIP
SIP
SIP
Media (RTP)
User Agent A
User Agent B
Henry Sinnreich and Alan Johnston
24
VoIP and IP Communications on the Internet
Internet
DNS SRV lookup fwd.pulver.com
SIP
SIP
server
server
sip19343_at_fwd.pulver.com
sip19343_at_fwd.pulver.com
session
sipaxelm_at_nic.at43.at
sipmah_at_nic.at43.at
sip18341_at_fwd.pulver.com
sip19343_at_fwd.pulver.com
25
Overall ETSI TISPAN IMS Architecture all
subsystems
Rf
/Ro
Ut
Rf
/Ro
Charging
Application Servers
Ut
Functions
Sh
Rf
/Ro
ISC
Dh
UPSF
Iw
SLF
IWF
Cx
Dx
Ib
P3
IMS / PSTN Simulation
PSTN Emulation (R2)
Mw
Mw/Mk/Mm
IBCF
Ic
I/S
-
CSCF
Mk
Mi
AGCF
P2
BGCF
Mk
Mw
Mr
Mj
Gq
'
Mg
e2
P
-
CSCF
SGF
MGCF
MRFC
Ie
Other IP Networks
P1
Gq
'
Gq
'
Network
Gm
SPDF
PSTN/ISDN
SPDF
Attachment
Resource Admission Control
Resource Admission Control
Mp
Mn
Subsystem
A-RACF
e4
MG
Re
Ia
MRFP
T
-
MGF
I
-
BGF
UE
BGF
RCEF
IP Transport (Access and Core)
CNG
-
26
ITU-T NGN System Architecture
27
What is your conclusion?

• Which approach will succeed, if the end-user has
  the choice what to use?
• The cheaper one?
• The one which is available (everywhere)?
• The one providing more features?
• The one easier to use?
• The one which will provide eventually more QoS
  for additional cost?

28
Cost Models
Distance and time dependent call charges
Old
Carrier selection Freephone
Regulation is based on this!
Backbone
Access
Access
Line rental
New
?????Do you sell calls or connectivity?
Access
Access
Backbone
John Horrocks
29
Transition from TDM to VoIP
BT 2015 DT 2019
30
Broadbands fast growth
Broadband access has quietly grown faster than
mobile phones in their early stages
Source ITU World Telecommunication Indicators
Database
31
Law and Policy Outmoded
BB Data Service
Infrastructure
New Model
Traditional Telecom Model
Brough Turner
32
Mike Powells 4 Internet Freedoms

• Freedom to Access Content Consumers should have
  access to their choice of legal content
• Freedom to Use Applications Consumers should be
  able to run applications of their choice
• Freedom to Attach Personal Devices Consumers
  should be permitted to attach any devices they
  choose to the connection in their homes and
• Freedom to Obtain Service Plan Information
  Consumers should receive meaningful information
  regarding their service plans.
• add Freedom to obtain your own location
  information

? regulation to concentrate on bottlenecks
33
One Problem remains Security

• Authentication, Authorization, Accounting,
  Non-repudiation,
• In the vertical model you trust the network (the
  carrier) and they trust each other (circle of
  trust)
• In the fixed network there is a trust-by-wire,
  
• in mobile networks you have a SIM-Card from the
  provider you trust (the only real asset of IMS)
• So you trust a Calling Line Identification
• But do you trust the VoIP provider of the other
  user?
• Especially if he is his own provider -)
• In the horizontal layered end-to-end model, also
  an end-to-end AAA is required.
• There is an additional requirement
• the end-user does not want to have too many
  identities,
• ideally only one for all types of
  communication- digital identity

Trusted End-User Identity
34
End-to-End Trust

• This requires the introduction of an end-to-end
  trust model
• An end-to-end trust model requires the use of
  certificates aka digital identities
• This is theoretically no problem, the necessary
  tools and protocols are available
• The problem is the trust model
• Hierarchical (e.g. PKI, X.509) or P2P (e.g. PGP)
• PGP does not scale
• In the PKI model you need a trusted 3rd party
• The problem both ends need the same trusted 3rd
  party
• - who is the root entity everybody trusts?
• Some developments are on the way, but very slowly
  and not very user-friendly (e.g. liberty
  alliance, eCards, etc. )

35
Two Global (IP-based) Networks
Global Phone Network
Global Internet

• heavily regulated
• optimized for speech
• end of lifetime
• accounting cascading,termination fees
• vertically integrated
• global connectivity
• mobility via roaming

• (still) unregulated
• multipurpose
• regarding IP Comm.
• begin of lifetime
• accounting peering
• horizontally layered
• global connectivity
• mobility and nomadic usage natively

NGN
P2P
36
Scott Bradner at NGN Conference

• Bradner says
• the ITU's model is designed to provide defined
  and guaranteed QoS,
• while the Internet is a best-effort model based
  on bandwidth capacity.
• He says both are applicable given the network
  circumstances - if there's plenty of bandwidth,
  there's no need for QoS controls if not, there
  is.
• the two models will not converge, but will
  interoperate.

Next Generation Networks Conference in
Washington, D.C. last week
37
Some Conclusions the next 5 years?

• Now lets assume the end-to-end Internet
  philosophy is taken seriously and the horizontal
  approach is followed up further
• The future developments will concentrate on
• EoIP IPoE IP Anywhere
• Broadband as Universal Service
• Multiplay, Computainment, rich and simple SW
• Personal, mobile/nomadic, general purpose devices
• Wireless communications
• P2P serverless communication (also with SIP)
• Meshed networks
• Viral communications ambient, organic networks
• Digital User Identities

38
The Impact?

• Based on these assumptions, what will be the
  impact of these developments on
• Regulation?
• Universal Service Obligations?
• Universal Service Funding?
• Emergency Services?
• Legal Intercept?
• On the Telcos?
• Side RemarkThe real regulatory battle of the
  futureDigital Rights Management

39
Résumé

• What is the Impact of Voice and Everything over
  IP to the Telecom Industry?
• What is the future of Telcos?
• Providing Services in a multiplay vertical
  stovepipe e.g. the IMS NGN Model?
• Broadband access providers only in a horizontal
  layered model enabling third party service
  providers
• A combination of both?
• Are new approaches needed?
• What is the role of Viral and Peer 2 Peer
  Networks?
• Trusted User Identities by Telcos or by Third
  Parties?

40
Telco Priority 1

• Provide local access, both for your own users and
  roamers
• Stick to the knitting
• Go where the money is
• Keep your customer
• There is only local competition
• Resale and wholesale
• The killer app here is speed
• Some nerds will have their own servers, but they
  still will need access

41
E.g. overcome Access Bottleneck
1000-to-1 disconnect !
Ethernet LAN Switches
Broadband Modem or Access Gateway
Internet Backbone
Brough Turner
42
Telco Priority 2

• Provide services
• OK, some nerds and large enterprises will have
  their own servers, they will not need a provider
  at all,
• but most end-users, SOHO and SME want their
  services hosted or at least configured
• Keep your customers
• Provide digital identities (they only real asset
  of IMS is the ISIM (SIM Card)
• Provide accounting services
• You may provide resale and wholesale also
• BUT, always keep in mind that the competition
  here will be global
• You have to compete against e.g. Skype, Google,
  eBay, Amazon, etc.
• You will need excellent products and excellent
  service

43
What happened in the last 4 weeks?

• eBay bought Skype for 4bn
• The Economist ?
• Mobile operators announce fixed line replacement
• Mobile operators will block VoIP
• Google announces Secure WiFi over VPN
• Neustar GSM Association team up to create an
  alternate DNS root to lock in customers
• Google proposes free WiFi access in SFO and later
  the whole US

44
The War is on

• Thank you

Richard Stastny ÖFEG 43 664 420
4100 richard.stastny_at_oefeg.at http//voipandenum.b
logspot.com