[opendtv] The Boldest Dumb Idea I've Ever Seen

Here's a post of mine that some on this list read about a month ago.

I see several killer services that should and could be delivered to handsets 
and that could make money.  Mobile video isn't one of them, unless content 
creators own the services, and even then ...

John Willkie 

The other day, I attended my first Society of Broadcast Engineers (SBE) section 
36 meeting.  I was intrigued by the location and the pitch.  The meeting was at 
the brand spanking new headquarters of Qualcomm's MediaFlo USA headquarters on 
Eastgate Mall in San Diego.  The pitch for broadcasters was that MediaFLO could 
be a new source of revenue for broadcasters.

Fat chance.

The challenges of out-of-home video are probably insurmountable.  Most of the 
time, people are at work, or at home, or are sleeping, or are driving their 
cars.  The only one of those activities that has a remote potential for revenue 
from MediaFLO is at home, but I suspect that few will enjoy the down-res. I 
recently read of a survey where 80 per cent of the people with video-enabled 
cell phones said they'd never use the service: and that was without regard to 
the cost.

As I pointed out before the meeting started (I have no interest in selling 
anything to Qualcomm), I am of the understanding that people are entertained by 
content and not technology.  The executive in charge -- with a cable 
engineering background, but who claimed to be a broadcaster -- kept on 
mentioning "content providers" but I didn't see any participating in this 
system, and last I heard television programmers call people who provide 
programs "content providers" and they all have their own extensive distribution 
networks.

The basics: MediaFLO uses proprietary and very ingenious Qualcomm technology to 
transmit, via a COFDM (but not DVB-compatible) single frequency network, up to 
20 video (320x240, up to 30 fps) and 20 stereo audio channels to 
Qualcomm-designed cell phones.  Right now, they have almost 20 markets 
pre-operational, and will have up to 100 by the end of the year.    The 
technology is designed to minimize power usage and to minimize the digital 
cliff, by using a base and enhanced layer.  Up to 15 fps comes from the base 
layer, and up to 30 fps when one is within a strong signal layer.   To minimize 
bandwidth usage they employ two-pass encoding. Total system delay (input to 
handset: 15 seconds)

The handset is quite impressive, and so are the installations -- last summer, I 
saw the early install of their facility on Mount Wilson.

Qualcomm has obtained (under 47 CFR part 27)  the rights to use channel 55 
around the U.S., at least once broadcasters vacate that channel.  On an interim 
basis, they are using channel 53 in San Diego, due to channel 55 being used by 
Los Angeles and San Diego DTV stations.

They have a national multiplex, distributed via a Ku transponder Intelsat IA8.  
In short order, they will have move their uplink from the main Qualcomm campus 
a few blocks away to this site.  Therein lies the first technical issue for 
them to discover, since this site is under the takeoff pattern of the Miramar 
Marine Corps Air Station.  As anybody with a satellite dish in this portion of 
San Diego knows, when jets takeoff from the airbase, their radar systems 
interfere with satellite reception and transmission.  It's been known for 
DECADES; several firms that weren't even near the takeoff pattern have had to 
move their uplinks and downlinks behind hills to shield them from routine 
transmissions.  When "there is a war (or battle) on, all bets are off."

The amount of money that they've spent on this facility is impressive.  The 
rack room has 110 racks.  Two separate 6mw power feeds; two mw UPS systems, 
computer flooring, main and back up a/c systems, 40,000 sf main building, with 
two more at the ready, with all expected to be occupied by the end of this 
year.  On site generator, and two 9 meter uplinks, several simulsats and more 
to be installed by the end of the year.  The nine meter dishes will be very 
good, I predict, at collecting RFI.

The live control room has 5 positions, in a 4:1 configuration.  Omnibus 
automation.  The main control room features a video wall with 18 5--inch 
screens, showing network status video signals, weather, input, uplink, downlink 
confidence displays.  Kind of funny that the display showing the work being 
done on the network showed the time as being PST, since we're currently in PDT.

The wiring in the rack room was the finest I've ever seen.  Each rack row has 
an "eye in the sky", the rooms are access-controlled, and have multiple 
fire-suppression systems.  

Our visit ended in the programming "pit", which was slightly largest than the 
the programming pit at KGB-AM/FM circa 1978.  "If we can't buy 
programming,we'll produce it ourselves."  That's the rub.  They have no 
content, and I didn't detect the slightest knowledge on their part that the 
hardest thing to bridge is the soft skills needed to work with talent, 
creatives and crafts to produce programs on time and within sight of a budget 
line.  I tried to discern if they understood the difference between the way 
cable firms and networks deal with producers, and the way that broadcasters and 
their networks deal with producers. The answer had something to do with HBO 
buying programs and producing unique content, which gave me the answer: they 
figure if they build the network, the content providers will come to them.

Yes, and they'll start off intending to sell old content at high cost, but when 
they see the facilities, they'll triple their prices, since the lack of 
contracts and content -- and the intention of competing with local and national 
broadcast networks -- since nobody is required to work with them, only breeds 
desperation.

They talk of having local aggregators to provide local content, perhaps with 
equipment installed at local tv stations.  Gee, we can then watch clips from 
local newscasts.  That 15 second throughput delay might be an issue.

One of my brothers is an EVP at Qualcomm.  When I left the meeting, what came 
to mind is his description of Java circa 1994:  a technological "solution" in 
search of a problem to solve. A few years later, he said that Java would NEVER 
be in cell phones. Two years after that, he was talking about how one could run 
a version of Java in 4K of memory. (Virtually all phones are Java-enabled now; 
Qualcomm found a way to embrace Java and defend their proprietary system.)

To seed the market for it's CDMA phones more than a decade ago, Qualcomm and 
Sony entered into a partnership,called Qualcomm Personal Electronics, that made 
cellphones.  It didn't really succeed, at least in the long run, but it was 
do-or-die for them, and they now thrive.  However,the breakup of the division 
was a painful one for Qualcomm.

I suspect that that experience informed Qualcomm in pursuing the MediaFLO 
strategy.  They are likely to spend much money before getting out of the 
content distribution business.  I do suspect they'll end up with several sports 
contracts, and little in the way of sports revenues, before costs are taken 
into account.

On engineer told me that he thought that MediaFLO equipped cellphones would end 
up being docked into a media center at home. Yeah,that's the trick: expensive 
subscription video with lower resolution than NTSC.

I do see something that could be a real revenue source for the system: local, 
subscription-only radio stations.  They have plenty of bandwidth,and commercial 
radio has gotten real crappy.  Of course, that can be done in a closet.

One attendee said his niece just got a video-equipped cell phone, but would 
never use it, as the cost for video was $.10 per minute.  

John Willkie




 
 
----------------------------------------------------------------------
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: