Nov 22, 2012

WHAT WE DOWNLOAD

The first video I downloaded was a music video in 1999. I think it was an Offspring song and it took most of the day to download on dial-up.  A far cry from full HD, the frame was maybe half the size of standard definition and compressed to pieces, but it was incredibly exciting to bypass the TV. Finding what you wanted was fun, required some skill, and the wait while the video crept over telephone wire built anticipation. All together, it made the otherwise forgettable "Pretty Fly For a White Guy" very memorable to me.

My parents' iPad now has an HBO app on it that streams Game of Thrones in HD, and they expect it to work whenever a wifi signal is around. At their place it does, and ever since the cable company began offering broadband in 2002, the holdup on getting high quality content wasn't on the home end of the connection. Once we were plugged in, we began eagerly awaiting websites like Youtube to spring up and make use of all the bandwidth. Well, they did, but suddenly we discovered all that bandwidth really isn't as much as we thought.

Of all the kinds of information we download online, video is the most demanding consumer of bandwidth. Currently, high definition video is the quality consumers expect, and its demand on our internet connection is equally high. You'll be downloading 2.3 Gb worth of information in an hour just to watch an HD movie on Netflix. In an urban area you'll find this to be an easy task, but even on the lowest video quality settings on our rural connection we experience frequent "buffering" pauses.

If video quality were to pique here, there's a lot that can be done to improve the situation. Codecs, the software that compresses video down to smaller file sizes while attempting to maintain the highest video quality, are constantly being improved. Where we currently require 5Mbps to watch an HD video, we may only require a 2Mbps connection in a few years to view the same quality video. The postage stamp sized music video I spent hours downloading thirteen years ago would stream in HD today without buffering.

Video standards aren't stagnant, but we might be inclined to look at the evolution of television formats as very broad and gradual: black and white, color TV, and now high definition tv in the span of fifty or more years. The reality, though, is that it won't take another twenty years to see the next big improvement in viewing quality. 3D TVs are already on the market, and we've begun to hear talk of a "4K" TV which would be up to four times larger than current HD frame sizes. These new larger formats aren't the latest hot-out-of-the-lab inventions-- it's just an estimation by electronics manufacturers of what consumers will possibly invest in. If it were conceivable, for instance, that we'd be willing to re-buy all our cameras and televisions tomorrow, the 8K technology would be ready to go by tonight. Upgrade to higher quality video is dependent, then, only on the rate at which consumers are willing to digest it, and how profitable it will be to strategically stagger each move out.

So in building a communications network it makes sense to keep an eye on the progression of video standards. Netflix, one of the most used high quality online video services, recently announced it streamed 1 billion hours of video in a month for the first time. Check out this crazy bit of information from Ciena:

And while 1 billion hours is a huge number, there is one even bigger – the total amount of bandwidth that was consumed for these 1 billion hours of streaming video. 
In fact, it’s a number so big that it is almost impossible for the average brain (like mine) to comprehend.  How much bandwidth was actually used?  This much:
7,200,000,000,000,000,000 bits
I actually had to do a Google search to figure out what that number is.  Turns out it is7.2 quintillion. That’s how many bits of Netflix video traffic were sent over networks during those 1 billion streaming hours.

Although the merger of internet and conventional TV delivery systems is probably a good topic for a different post, it's important to note that we're already seeing the transition between traditional understandings of separate utilities like phone, TV, and internet to the one data connection that takes care of all your communication needs. It's a matter of time before all communication will be served through one type of digital connection and used on one multi-faced always connected computer. This isn't to mention the additional bundling of possibilities that could arrise from a high capacity network-- we shouldn't be thinking about simply meeting the demands we have right now, but about building an infrastructure that won't impede the development of those services we haven't yet dreamed of.

Amanda and I passed the time on the road yesterday by listening to old TV and music we hadn't heard since the 90's; it seems you can find anything on Youtube. 4G LTE is peppy, and it's easy to stream what we're looking for, but how long will it be before the content I'm looking for outstrips this network's capacity? Codec compression, for instance, may always provide a "mobile" version of the upcoming improvements in video quality, but at home there has to be an answer that addresses the need for "future proofing".