VoIP Now: What Internet TV Needs: 7 Suggestions/ Concerns

February 23, 2007

What Internet TV Needs: 7 Suggestions/ Concerns

What's listed here doesn't preclude the possibility that some software or web service already does it. This is my list of ideal IPTV (Internet TV)-related functionality.

Mobile TV.
This is fine, but with wearable, comfortable goggles that project a virtual large screen. Little tiny phone screens won't cut it. The goggles are out there. They just need to be married with smartphones and PDAs. (i.e., maybe through Bluetooth, since cellular data plans are outrageously priced in some countries.)
Wireless streaming.
From my computer to my TV, if I want. (Though my computer screen is still larger than my TV, and I use an external TV capture box, which gives better performance than IPTV.) Apple's tentatively called iTV, for the digital living room, is one example.
Faster Internet connection speeds.
Let's face it, Joost might be nice (I'm still waiting for a Babelgum invite), but a faster connection would help, obviously. And what happens, for example, when everyone in my neighborhood on cable Internet starts watching at the same time? At that point, I turn back to regular cable TV, as will others. The success of IPTV hinges on much faster connection speeds.
More bandwidth.
My cable Internet provider caps me at 6 Gb/mth. I eat bandwidth for breakfast. I can use a Gigabyte in a single day sometimes. But can I buy more bandwidth? Noooooooo. Instead, if I go over in a given month, they'll warn me twice then cut me off until the next month - something I simply cannot afford to have happen, as a freelance writer. And with Joost's bandwidth consumption, this is important. Which is why I've stopped using it, beyond a few beta tests.
New compression coding.
Wavelets theory is an ultra-geeky discipline created by brilliant physicists in the 1970s but has roots in studies done in 1909. It's pure, advanced applied mathematics used to model a lot of phenomena, and a math professor told me that even most PhD's in math or physics don't understand it fully.

Data compression of images and video is one application, and depending on the algorithm used, the space savings are phenomenal. The benefit is that a crunched file would download very quickly. The problem is, that massive crunching requires a fairly significant amount of processing power to uncrunch for viewing. It certainly could not be done, with present home computers, in real-time. That is, you couldn't watch streaming video as it comes in over your Internet connection if the video data has been massively crunched with wavelet compression. The alternative is to not compress and have a faster connection, or more powerful graphics cards.
Quadcore video boards.
The whole net neutrality debate was sparked, from what I interpret, when Internet providers felt they had to apply a tiered price structure for connections based on expected usage. Fact is, if we suddenly had the billion or so current Internet users all using VoIP and/or IPTV simultaneously, the current infrastructure couldn't handle it. (Correct me if I'm wrong.)

We all probably want faster connection speeds, and they're coming, but will take time to roll out. What could come sooner is a new set of video compression codecs (last point) coupled with high-power graphics cards sporting their very own quad cores or more. If our graphics cards were powerful enough, and we used super-crunched video formats, we might possibly reduce bandwidth requirements down to a point where every Internet user could potentially watch Internet TV simultaneously. (Of course, it'd be nice to have something similar for VoIP communications: a quadcore sound card.)
More content, more choice.
Video sharing sites already have a great deal of content choice, though not all of it is necessarily watchable. Soft clients like Joost are young yet, but will need - in my honest opinion - a great variety of content, and a pay-per-view model without advertising. And that requires sign-on from production houses.