[opendtv] The top ten hang-ups in home networking
- From: "Manfredi, Albert E" <albert.e.manfredi@xxxxxxxxxx>
- To: <opendtv@xxxxxxxxxxxxx>
- Date: Wed, 3 Jan 2007 17:33:28 -0500
A friend of mine is putting off buying HDTV, until HDMI 1.3 becomes
available in all the components he needs. I chuckle at the "one issue"
thinking, considering how much more there is than just HDMI 1.3 to think
about.
There's no telling whether HDMI 1.3, or any version of HDMI, will last
long enough to become that commonplace, and you can be certain that
before the ink dries on HDMI 1.3 there will be any number of other
supposedly "must have" wonderful new interconnect innovations just
around the corner.
This article does a wonderful job of laying this out, including, more
broadly, home networking schemes. This is why it is futile to wait,
expecting sanity to establish itself anytime soon.
IMO fragmentation occurs because the bigger payoff for developers is to
create their own brand new standard, when at all possible, in spite of
the rhetoric. And hope to collect royalties on it. Not a lot of benefit,
to developers, to adopt an existing standard. Puts a whole new meaning
on the term "standard." When so many new approaches are aimed at
providing essentially the same thing, you have to know that no one is
really interested in a common, interoperable solution.
Wasn't the UWB fiasco a perfect example?
The losers are consumers. Just hang onto the analog interfaces, if you
want to be future-proof at all.
Bert
------------------------------
The top ten hang-ups in home networking
Rick Merritt
(01/02/2007 4:15 PM EST)
URL: http://www.eetimes.com/showArticle.jhtml?articleID=196800481
The home network is the next big frontier in electronics, but it's still
untamed territory today. A few data points provide a snapshot of the
opportunities. Market watcher iSuppli Corp. (El Segundo, Calif.)
predicts shipments of products with integrated wired home networking
will rise by more than a factor of 10 in the next four years, to hit
223.8 million units in 2010. Parks Associates estimates the number of
North American homes with networked digital-video recorders more than
tripled from 400,000 in 2005 to 1.7 million by the end of 2006.
But there are no easy pickings in this gold rush. Engineers face
historic levels of complexity building the digital home for several
reasons. An unprecedented number of players are competing for a piece of
the action. Coordination between these would-be architects is minimal.
What's more, the stakes are high. Consumers expect both top notch
quality and ease of use. Imagine routing high definition video streams
among various set-tops, players and TV screens without dropping a frame
or frustrating Joe Viewer who is trying to fire up HD versions of the
Super Bowl for his buddies in the living room and "Toy Story" for the
kids in the den.
"There's a huge set of things engineers have to put in place--digital
rights management, media formats--and you have to have all the pieces
implemented before the content flows," said Brendan Traw, chief
technology officer of Intel's digital home group. "If any piece of the
puzzle is not present, it doesn't work," he said in our preview of the
2006 Consumer Electronics Show.
The ultimate solution could require a realignment of the consumer
industry from vertically oriented companies to a more horizontal
structure in which different companies handle different pieces of the
problem. That would make the consumer sector look more like the computer
industry, said Mike Buckley, a director at Intel Capital, which manages
a $200 million consumer fund.
So for anyone who may have missed some of the many stories we've been
writing about the opportunities and pitfalls building the digital home,
here's our list of the top ten hang ups in home networking.
It's easy enough to show home networks are still too complicated-just
try them sometime. The analysts and even the industry players know
that's true.
"The glue that holds all this together is home networking, and it
stinks," said Van Baker, a consumer analyst with Gartner Dataquest, in
an early 2006 story. "If home networking stays like it is, it will stall
at 30 percent penetration," he said.
"We've all dealt with that rat's nest of wire behind our home
entertainment centers," said Bruce Watkins, president and COO of
Pulse-Link Inc. (San Diego) in the same article.
Pulse-Link is one of several companies that joined the High Definition
Audio-Video Network Alliance (HANA) to create guidelines making it
easier to link TVs, digital recorders, and storage devices via a single
IEEE 1394 cable. The group debuted in late 2005 promising systems will
emerge using its approach in 2007.
Another Gartner consumer analyst, Jon Erensen, summed up the situation
succinctly in a December 2006 briefing in San Jose. "There are a lot of
solutions out there, but not the real interoperability you need for
wired and portable products across multiple vendors," he said.
The problem is not that there is a lack of mechanisms to ensure what's
called quality-of-service (QoS) on the home network. To the contrary,
there are too many of them.
"Everybody has a different notion of what QoS should be, but if you've
got more than one QoS, you haven't got any," said Glen Stone, a director
of strategy, standards and architecture for Sony Electronics.
We explored this topic in a November story and found the problem is many
players see the capability of delivering multimedia over a home net as a
competitive advantage or core competency. "They fundamentally want to
have control over QoS in the home net because if something goes wrong
people will call them for support," said Stone.
Problems only get worse with the move to high definition TVs and DVDs.
"We are focused on the HD experience, and that content really
exacerbates the QoS issue," said Gary O'Neall, vice president of global
set-top development for Motorola.
Two groups are doing some fundamental work in this area. The Universal
Plug and Play Forum is tackling the issue from a high-level software
perspective. Meanwhile the 802.1 Audio/Video Bridging Task Group is
trying to make changes in silicon for QoS.
Two other groups may leverage their work. The Home Gateway Initiative
represents telcos and their suppliers, and Cable Labs handles R&D in
areas including home networking for the cable TV industry.
Despite the rise of 802.11n broadband wireless links in 2007, Wi-Fi will
be no panacea for the digital home, we concluded in our preview for the
2007 CES.
Other wired and wireless solutions will come on strong this year. They
include coax-based approaches from the Multimedia over Coax Alliance
(MoCA) and Home Phoneline Networking Alliance (HPNA), the
powerline-based HomePlug 2.0 and various flavors of ultrawideband.
HPNA leapfrogged its competition in throughput in November, announcing a
version 3.1 spec with an aggregate 320 Mbits/second over two
simultaneous channels. But MoCA officials say cable TV suppliers will
come out supporting their approach in 2007. We profiled the competition
among these and other players in a January 2006 story.
"11n will make streaming audio better and video possible--as long as you
are not using your microwave--but I still think the digital home will be
a heterogeneous environment and no one physical layer will win," said
Scott Smyers, vice president of network systems architecture at Sony
Electronics Inc. and chairman of the Digital Living Network Alliance.
The service providers are calling the shots in home networking for
set-top box makers such as the Scientific Atlanta group of Cisco that
sees itself as agnostic, said Dave Davies, vice president of strategy
and product marketing for SA's digital set-tops.
"The cable operators are more interested in MoCA, the telcos focus on
MoCA and HPNA-over-coax and others are thinking about HomePlug, so we'll
see multiple flavors of home networks in 2007," said Davies.
Service providers see wireless as expensive, insecure and unreliable.
They fear costly service calls due to routine interference as well as
thefts of service from apartment owners picking up a neighbor's wireless
TV signals. Thus SA has no plans to integrate any wireless networking in
its set-tops in 2007, though it does already sell Wi-Fi peripherals that
attach to them via USB.
"I personally see wired nets taking off in 2007. They will definitely
start shipping in volume, but no one will be in the tens of millions for
the next couple years," said John Hussey vice president of the
high-speed signal processing group at Analog Devices Inc.
Life isn't any easier for those who just want to add a little
intelligence to managing their lights, security and HVAC systems. There
are at least three new and three traditional approaches to handling home
automation with plenty of companies backing each camp.
Our Embedded.com online editor Bernie Cole laid out the fundamentals in
a forward-looking article way back in 1999. Cole said any home net needs
to be simple, low cost and not require new wiring in his profile that
included a look at CEBus, X-10, LonWorks and more.
More recently, much of the home and industrial-control buzz has centered
around Zigbee. The open spec got an update in 2006 supporting star and
mesh topologies.
Before the ink dried on the first Zigbee spec, chip and software
developer Zensys gathered more than 60 companies to form an alliance
that will push for the adoption of Zensys' Z-Wave wireless protocol in
the home automation market.
Before either of those alternatives got off the ground, Smarthome Inc.,
a large maker and retailer of home automation products, announced
another option. Insteon is a hybrid power line/wireless networking
technology that it claimed will fix the reliability problems of current
X-10 home control networks while retaining backward compatibility with
them. Analysts were cool on Insteon because it lacks the bandwidth and
standards backing of Zigbee.
Short-range personal-area networks (PANs) are every bit as fragmented as
their counterparts that try to stretch across the entire home. The main
underlying transport networks--Bluetooth and ultrawideband (UWB)-come in
flavors with various protocols running on top of them.
We profiled the sector in a May story from the Windows Hardware
Engineering Conference (WinHEC) where Microsoft announced its plans to
support WiNet, a form of Internet Protocol over UWB.
Cellphone makers are stirring the pot with their own PANs. Nokia rolled
out in October Wibree, a low-power derivative of Bluetooth for links to
toys and gadgets running on button-cell batteries. Handset companies are
also pushing for near-field communications as a transport for electronic
payments. How these mobile PANs will interact in the home network
environment is still unclear.
"Right now, it's a mess," said Liam Quinn, chief technology officer for
communications and peripherals at Dell Inc, speaking at WinHEC. "We are
collectively doing a poor job articulating what is Bluetooth,
ultrawideband and so on. If we can't do this as savvy technology people,
how can users do it?"
The problem is that everyone has their favorite protocol, and developers
differ on what they consider to be the best ideas for making it easy to
associate nodes on an ad hoc wireless net. But systems have limited
space for all the wide-, local- and personal-area radios and antennas
that the new efforts are generating.
"At this point, nothing's clear. We will be six to eight months working
through this," said Alec Gefrides, a wireless strategist for Intel Corp.
"People are still putting up flags for new camps."
The area of dedicated point-to-point connections between consumer
systems is also seeing a rising tide of new options, potentially
confusing and confounding both systems makers and their customers. Once
again, the intensity is greatest at the high end among new options to
handle the heady bandwidth requirements of high def video.
On the wired side, DisplayPort and the Unified Display Interface are
vying to become the standard for a secure digital link in consumer
systems and computers. The pair will compete against two digital
interfaces already in use-- Digital Visual Interface and High Definition
Multimedia Interface (HDMI).
We've ended up with a nightmare scenario of multiple standards. It's a
frightening mess," said Bob O'Donnell, vice president of clients and
displays at International Data Corp. (Framingham, Mass.). "The notion of
a converged display interface may just go away," he said in a May report
examining the conflict.
On the wireless side, startup SiBeam announced it has gathered an ad hoc
consortium called WirelessHD around its approach using 60GHz radios to
deliver up to 5 Gbits/second. It will compete with companies using
ultrawideband to deliver 480 Mbits/s or more over an approach dubbed
WirelessHDMI.
Then there are startups like Amimon doing proprietary twists with
802.11n to make it a high-def ready interconnect.
"It is completely unpredictable at this point whether one technology
will win or not, but I don't think that will happen," said Craig
Mathias, a wireless analyst at Farpoint Group.
Home networks are built to carry both personal and paid-for content,
like songs and movies. But there's no standard way to protect the
so-called premium content from being copied and freely distributed.
A whole new category of mainly software security products is growing up
around different digital rights management approaches (DRMs). According
to our 2007 CES preview, most observers believe the industry will
struggle with an increasing number of proprietary DRMs for a long time.
Many see Microsoft's Windows Media DRM gaining momentum as a de facto
standard due to its broad use in PCs, a prospect some industry observers
and consumer OEMs said they find "too frightening." Others are angry
that Apple has not opened up the FairPlay DRM used in its wildly popular
iPod.
Cable and satellite TV companies share some of the blame. They tend to
want to have a single DRM as part of a closed end-to-end system.
That leaves the digital home with one DRM on their PC, another on their
iPod and a third on their TV content-and none of them talk with each
other.
Several groups are working on DRM standards including the
cellphone-focused Open Mobile Alliance and the Europe based Digital
Video Broadcasting Project. One of the most closely watched is the ad
hoc Coral Consortium which has already released an initial specification
for passing content between different DRMs. However none of the
approaches have made their way into real products yet, and none have the
cooperation of Microsoft and Apple for linking to mainstream PCs and
iPods.
In addition, Intel is promoting Digital Transmission Content Protection
over IP as a core standard for this area. The Digital Living Network
Alliance has adopted DTCP-IP as a requirement and Windows Media DRM as
an option as part of its first pass at a standard for content
protection. Separately, Sun Microsystems has been promoting its concept
for an open-source, royalty free DRM, or what it calls its Dream project
in Sun Labs.
"We don't see much light at the end of this tunnel. There are efforts
out there, but they aren't making much progress," said Van Baker, a
consumer analyst for market watcher Gartner Dataquest. "Vendors are
determined to monetize their own assets through proprietary
technologies," he said in our preview of the 2006 Consumer Electronics
Show.
Engineers had hoped they could concoct a variant of Linux for consumer
electronics systems, providing an open platform that could lead to
greater interoperability between systems on a home network. The
operating system would need to support digital media, broadband
networking, near real time response and do that on an extremely low-cost
and low-power budget.
As of our May story that covered this issue, developers have pretty much
let go of that ambitious goal, settling instead for promoting as much
collaboration around Linux as possible. Scott Smyers, a Sony executive
who also chairs the Consumer Electronics Linux Forum, said the
three-year old group has evolved from pursuing a common framework to
providing a forum to share software components. "Over time, a common
framework may emerge, but at every step there may be opportunities for
differentiation," Smyers said. "For any one problem like power
management there may be more than one open-source solution."
Intel Corp. is one of the few companies that has not given up hope. It
tried to rally support for the desktop-oriented Linux Standard Base
effort at a poorly attended session of the Intel Developer Forum earlier
this year.
Linux Standard Base is a functional spec working at the library and
command level--not at the kernel level--to act as an application
portability standard for GNU/Linux. Intel called for developers to get
involved in the definition of a version 4.0 now in the works.
An executive from Motorola's cell phone group expressed familiar
frustrations with consumer Linux. The company made a commitment in 2005
to consolidating work on various software platforms to Linux as a free
and open code base that does not depend on any third party for enabling
new hardware. But the lack of a standard is its major drawback, said
Christy Wyatt, vice president of third-party relations for Motorola.
Some fault chip companies such as Marvell and Broadcom that provide
Linux stacks that are just a little different from those of their
competitors so that OEMs are encouraged to stick with their chip sets.
Microsoft shares some of the blame for making OEMs liable for any
compromises of the Windows Media DRM, a legal technicality that scares
some OEMs away from open-source Linux and into the arms of more tightly
controlled RTOSes.
Home networks are in such an early stage of development that even the
interoperability efforts are still fragmented, we reported in a May
article.
At least three major efforts are trying to address interoperability in
the digital home-the Digital Living Network Alliance (DLNA), the
Universal Plug and Play Forum (UPnP) and Intel's Networked Media Product
Requirements (NMPR). The programs are roughly coordinated, but have
their gaps and areas of overlap.
Typically, DLNA adopts all the work of UPnP, which creates application
programming interface standards so systems can work together over home
nets. UPnP does endeavor to set new standards where there are none.
Ultimately, DLNA aims to take the widest possible scope, filling in gaps
left by UPnP and other groups.
Trying to nudge the pace of progress, Intel created its own suite of
interoperability standards with NMPR. The effort was "a time-to-market
play. A consensus process like DLNA takes longer than you might like,"
said Brendan Traw, chief technologist in Intel's digital home group.
China has kicked off a standards group called Intelligent Group and
Resource Sharing Alliance that roughly parallel's the objectives of
UPnP. The two groups have made efforts at to work together.
In addition to these broad efforts, there are a number of groups working
on home net standards for specific constituencies. The Home Gateway
Initiative represents telcos and their suppliers, Cable Labs handles R&D
for the cable TV industry, the Open Mobile Alliance is focused on the
cellular industry and the Digital Video Broadcasting Project has efforts
for Europe's TV broadcasters, to name a few.
And finally,
10. There are too many cooks in the kitchen
Roll it all together and one big theme emerges: There are just too many
loosely connected players in this rapidly growing field.
As we reported in our CES 2007 preview, it's a heady mix of players.
They include Hollywood studios; cable and satellite TV providers;
cellular carriers; traditional telcos; consumer, computer and
communications OEMs; chip makers and more. Every player in every
industry is looking for its own silver bullet, an optimal
software/hardware platform that advantages its own content or devices
and guarantees a piece of the action every time content shifts places or
devices.
"The proliferation of formats and desire to interconnect everything ends
up creating a mad scramble," said Bill Bucklen, a segment director for
advanced TV at Analog Devices Inc.
The net result: The digital home will face technology fragmentation for
some time to come, and that is its biggest hang up.
All material on this site Copyright 2007 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] The top ten hang-ups in home networking