[opendtv] IP video angst: For carriers, it's the Wild West

  • From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
  • To: <opendtv@xxxxxxxxxxxxx>
  • Date: Wed, 3 May 2006 13:23:31 -0400

I see much hype and not enough carefully considered tradeoffs, but at
least it gets the point across that there are a lot of confused
operators out there.

Bert

--------------------------------------------------
IP video angst: For carriers, it's the Wild West

Loring Wirbel
(05/01/2006 9:00 AM EDT)
URL: http://www.eetimes.com/showArticle.jhtml?articleID=3D187001816

Las Vegas -- The flexibility of Internet Protocol delivery soon will
make traditional RF broadcasting a thing of the past. But even as
operators of all stripes tackle the technical challenges that IP
television presents, they are entering a whole new world of content
choices.

It isn't clear that users will want traditional channels at all, once
they realize that all the standard programs plus a lot of untraditional
IP video content are available on demand at any time, Stephen Reeder,
director of product strategy at ANT Software Ltd., said last week.
"Interactivity itself is a revenue generator, but the type of
application can be very specific to region," he said.

For example, Reeder said, karaoke on de- mand is huge in Taiwan but has
been a lackluster seller everywhere else, while online gambling over
IPTV is big in Germany.

"This is more of a silver shrapnel than a silver bullet model," he said,
"and the deadly factor for service planning is that there's no way to
tell if a service will be successful until after you deploy it. In fact,
the highest revenue generators we have found are for on-demand subjects
that are very topical and short-lived. Background material for celebrity
reality shows, for example, may have a popularity spike measured in days
or hours."

Content was just one of the issues addressed at last week's National
Association of Broadcasters meeting here, as conference speakers
attempted to chart a course from legacy broadcast models to on-demand IP
or "clipcast" services. Indeed, observers said that even a full
broadcast programming model might best be provided as an IP flow.

"At the end of the day, IP isn't just a vehicle for on-demand service or
clipcasting. It can emulate real programming channels--assuming the
interactive user of the future is even going to care about channels,"
said Todd Waters, director of business development of IP subscriber
networks at Scientific-Atlanta. "The problem for the traditional
broadcast or cable carriers is, when do you cut across to IP services,
and how do you maintain an economical set-top box as you prepare for
that?"

Nor is the problem unique to the legacy analog broadcast channels.
Standard- and high-definition digital TV networks still follow the
push-broadcast model, and even new compression standards like MPEG-4
were not designed with TCP/IP transport in mind. Broadcasters and cable
TV operators could also share a common threat from telephony carriers
and Internet service providers as those companies move to IPTV.

Tom Giunta, vice president of product marketing at Motorola Inc.'s
wireline services business, warned that telco fiber-rich IPTV projects
like Verizon's FIOS and AT&T's Project Lightspeed could represent a
threat to all broadcast models, both in end-user bandwidth and in
management-plane control of hybrid delivery systems.

"The IMS [IP Multimedia Subsystem] architecture is absolutely essential
for common management of content based on IP, and so far the phone
companies are ahead in realizing this," Giunta said. "Meanwhile, they
are upgrading their access networks to a fiber-deep architecture, using
a mix of passive optical networks and ADSL2+."

Ryan Petty, vice president of the Myrio group within Siemens AG, said
that broadcasters and new video service providers alike must be prepared
to deliver the full range of broadcast possibilities: traditional
broadcast content, on-demand IP service, "local channel" equivalent
service, exclusive content and the extreme-niche specialized content
known as "long-tail" services (referring to the revenue obtained from a
large archive of obscure, less-popular content). A newcomer to IPTV
services should offer something available nowhere else, he said, such as
Belgacom's recent offer of exclusive access to Belgium's premier soccer
games.

In the extreme case, a carrier with no true broadcast legacy can offer a
streaming IP service that appears to be multichannel but is more
flexible than real-time broadcast. For example, said Petty, Shanghai
Telecom already offers 100 "channels" that consist of content stored and
archived in the network, giving users rewind and fast-forward capability
in what appears to them as a live-broadcast network.

The problem for broadcasters and cable operators in North America is
that current cable headends, as well as the "super headend" programming
centers used by digital broadcasters, are not prepared for bringing in
archived or near-real-time IP services. Jim Olson, chief executive of
Tandberg Television subsidiary SkyStream Inc., said that SkyStream
followed the move of intelligence to the network edge in developing
multifunction, multiband headend systems that could be adapted for HD
and IP mixes. It offers a full chassis-based multislot system for super
headends, while shrinking the same functions to a 1U system with
mezzanine modules for video-server edge applications. Parent Tandberg's
expertise in specific encoding functions may later show up as modular
"blade" technologies for the SkyStream iPlex, Olson said.

A difficult problem for cable operators with legacy RF traffic is the
difference in bill-of-materials costs between a simple IP set-top box,
common in many green-fields network locations in China, and a set-top
box that also must embed microtuners for traditional channels. An IP
system acts as a simple router, said Waters of Scientific-Atlanta, so
more pseudo-channels implemented as IP flows can be added with little
additional cost at the set-top.

But a cable set-top box must add microtuners to increase the number of
legacy channels, even if it moves some of its services to IP networks,
Waters said.

Paul Robinson, MPEG segment manager in Tektronix Inc.'s video test unit,
said that a similar gap exists between those familiar with monitoring
MPEG transport streams and those test engineers who grew up with IP
protocol analysis. Tektronix's partial solution was to develop two
generations of Video Quality Manager tools for its IP-centric Spectra 2
analyzer, while providing IP correlation information for the MP400
MPEG-4 transport stream analyzer. It will be a long and tortuous path to
bring MPEG and IP testing together, Robinson said, but the correlation
of IP packet problems with video artifacts represents an important first
step.

All material on this site Copyright 2006 CMP Media LLC. All rights
reserved.
 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts:

  • » [opendtv] IP video angst: For carriers, it's the Wild West